JP5790820B2 - Inconsistency detection apparatus, program and method, correction support apparatus, program and method - Google Patents

Description

  The present technology relates to a technology for grasping or improving document quality.

  In system development, various design documents are created in each development process. Since the quality of the design document greatly influences the progress of the system development project and eventually the quality of the completed system, a technique for appropriately managing the quality of the design document is required.

  By the way, in a UI (User Interface) process, for example, a sentence describing the specifications of processing logic (hereinafter referred to as processing details) and a fixed form (hereinafter referred to as item definition) that defines item names included in the processing details. A design document including An example of such a design document is shown in FIG.

  In quality control of such a design document, it is necessary to check whether there is any inconsistency between item definitions and processing details. An inconsistency here is an inconsistency that the process name is not described in spite of the item name defined in the item definition, or that the item name is included in the process details. It is an inconsistency that is not defined.

  The former inconsistency can be easily found. For example, in the example of FIG. 1, it may be determined whether or not there is a description of “establishment code” in the process details by searching for process details using the item name “establishment code”.

  On the other hand, the latter inconsistency cannot be easily found. Since the processing details include many words other than the item name, a method such as simply extracting a noun phrase from the processing details and checking whether the extracted noun phrase is defined in the item definition Even a noun phrase that should not be defined in the item definition is determined to be an undefined item name. For example, in the example of FIG. 1, noun phrases such as “screen” and “notification message” are determined to be undefined item names, which is not appropriate.

  Conventionally, as a technique for extracting a phrase from a document, a search key (keyword) for extracting an item name is prepared in advance, character strings including keyword characters are identified from a specification document, and unnecessary characters are further extracted. Is known, and a check target item name to be subjected to a consistency check is extracted.

  However, in the above prior art, unless an appropriate keyword can be selected by any method, it is not possible to extract a check target item name suitable for a consistency check with high accuracy. In addition, a method of preparing a keyword used for a search in advance according to the contents of the specification document has a problem that the cost is high and the versatility is low.

  In addition, the former inconsistency has the following problems. Specifically, in the prior art, it is possible to detect inconsistencies and extract data related to item names not included in the processing details from the processing details of other design documents, but supplement the extracted data. There is a problem that a position to be specified cannot be specified.

JP 2008-186356 A

  Accordingly, an object of the present technology is, according to one aspect, to provide a technology for complementing missing data in a document.

  The inconsistency detection apparatus according to one aspect of the present technology is extracted from an item definition including (A) an independent word group extracted from a sentence included in a document, an item name included in the sentence, and a definition of the item name. Stored in the document database, (B) a data storage unit storing the independent word group and item name group of the first document to be diagnosed, and (C) the document database. The degree of similarity between the independent word group of each document and the independent word group stored in the data storage unit is calculated, and a document whose similarity is equal to or greater than a predetermined threshold is identified as a similar document and identified. Similar document specifying means for extracting the independent word group and item name group of the similar document from the document database, and (D) the first independent word that is an independent word included in the independent word group stored in the data storage unit Item name to be used for similar document identification means Item candidate extracting means for extracting from the extracted item name group, and (E) item names not included in the item name group stored in the data storage unit among the item names extracted by the item candidate extracting means And an inconsistent item specifying means for specifying as an inconsistent item name.

  A correction support apparatus according to another aspect of the present technology includes (A) a database that stores item definition data that defines item names related to processing, and detailed processing data that defines the details of the processing for each processing; ) Read out the first item definition data and the first process detail data corresponding to the first item definition data from the database, and are defined in the first item definition data and the first process detail data An item extraction unit that extracts item names that are not included as inconsistent item names, and (C) second processing detailed data including inconsistent item names among the processing detailed data stored in the database. A replenishment data specifying unit for specifying a position where the item name appears in the second processing detailed data and storing position information representing the position in the storage device; and (D) the first item from the database. The item names defined together in the second item definition data corresponding to the definition data and the second process detail data are extracted as common item names, and the common item names are the first process detail data and the second process. Using the corresponding position specifying unit that specifies each position appearing in the detailed data, stores the position information representing the position in the storage device in association with the common item name, and (E) the data stored in the storage device A range specifying unit for specifying the common item name that appears immediately before the inconsistent item name and the common item name that appears immediately after the common item name as the immediately preceding item name and the immediately following item name in the second processing detailed data. Have.

  The missing data in the document can be supplemented.

FIG. 1 is a diagram for explaining the premise of the present embodiment. FIG. 2 is a functional block diagram of the inconsistency detection apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an example of data stored in the design document DB. FIGS. 4A to 4C are diagrams for explaining a method of generating an independent word list with processing details. FIGS. 5A and 5B are diagrams for explaining a method of generating an item name list of item definitions. FIG. 6 is a diagram showing a main processing flow in the first embodiment. FIG. 7 is a diagram illustrating a process flow of the similar design document specifying process. FIG. 8 is a diagram showing a processing flow of item candidate extraction processing. FIG. 9 is a diagram illustrating an example of data stored in the matching item data storage unit. FIG. 10 is a diagram illustrating a processing flow of item candidate extraction processing. FIG. 11 is a diagram illustrating an example of data stored in the item candidate storage unit. FIG. 12 is a diagram illustrating a process flow of the appearance portion comparison process. FIG. 13 is a diagram illustrating an example of data stored in the common item data storage unit. FIG. 14 is a diagram illustrating a process flow of the appearance part comparison process. FIG. 15 is a diagram illustrating an example of data stored in the similar partial data storage unit. FIG. 16 is a diagram illustrating a process flow of the appearance part comparison process. FIG. 17 is a diagram illustrating an example of data stored in the distance data storage unit. FIG. 18 is a diagram illustrating a process flow of the item definition comparison process. FIG. 19 is a functional block diagram of the correction support apparatus according to the second embodiment. FIG. 20 is a diagram illustrating an example of item definition data stored in the design document DB. FIG. 21 is a diagram illustrating an example of detailed processing data stored in the design document DB. FIG. 22 is a diagram showing a main processing flow in the second embodiment. FIG. 23 is a diagram illustrating an example of item definition data stored in the input data storage unit. FIG. 24 is a diagram illustrating an example of detailed processing data stored in the input data storage unit. FIG. 25 is a diagram illustrating an example of data stored in the supplementary sentence storage unit. FIG. 26 is a diagram illustrating a processing flow of range narrowing processing. FIG. 27 is a diagram illustrating an example of common item name data. FIG. 28 is a diagram illustrating an example of data stored in the line number list storage unit. FIG. 29 is a diagram illustrating an example of data stored in the corresponding position data storage unit. FIG. 30 is a diagram illustrating a process flow of the corresponding position specifying process. FIG. 31 is a diagram illustrating an example of data stored in the narrowing processing result storage unit. FIG. 32 is a diagram for explaining the range narrowing processing. FIG. 33 is a diagram for explaining processing performed by the corresponding position specifying unit. FIG. 34 is a diagram for explaining processing performed by the corresponding position specifying unit. FIG. 35 is a diagram illustrating a processing flow of the first replenishment position determination processing. FIG. 36 is a diagram illustrating an example of window data. FIG. 37 is a diagram for explaining processing performed by the search unit. FIG. 38 is a diagram illustrating an example of data stored in the output data storage unit. FIG. 39 is a diagram for describing the first replenishment position determination process. FIG. 40 is a diagram illustrating a process flow of the second replenishment position determination process. FIG. 41 is a diagram showing an example of window data. FIG. 42 is a functional block diagram of a computer.

  Hereinafter, the present embodiment will be described in detail. In the present embodiment, it is assumed that the design document includes item definitions and processing details.

[Embodiment 1]
First, a process for detecting inconsistency (first inconsistency) that is not defined in the item definition although it is an item name included in the processing details will be described.

  FIG. 2 shows a functional block diagram of the inconsistency detection apparatus according to the first embodiment. The inconsistency detection apparatus according to the first embodiment includes (A) an input data processing unit 1 that accepts input of process details and item definition of a design document to be diagnosed, and generates an independent word list and an item name list; (B) An input data storage unit 3 that stores the independent word list and item name list generated by the input data processing unit 1, and (C) a design document DB 7 that stores an independent word list and item name list for each design document. (D) A similar design document specifying unit that performs a similar design document specifying process to be described later based on the independent word list stored in the input data storage unit 3 and the independent word list stored in the design document DB 7 9, (E) a similar design document storage unit 11 for storing an independent word list and item name list of the similar design document specified by the similar design document specifying unit 9, and (F) an input data storage unit 3. Data and data Based on the data stored in the design document storage unit 11, the item candidate extraction unit 5 that performs an item candidate extraction process, which will be described later, and (G) the item candidate data extracted by the item candidate extraction unit 5 are stored. A first inconsistent item that identifies an inconsistent item from the data stored in the item candidate storage unit 13 based on the item candidate storage unit 13 to be executed and (H) the item name list stored in the input data storage unit 3 The identification unit 15, (I) a first inconsistency item storage unit 17 that stores data of inconsistency items identified by the first inconsistency item identification unit 15, and (J) the first inconsistency item storage unit 17 And an output unit 19 that performs processing for outputting stored data.

  The item candidate extraction unit 5 includes a matching item extraction unit 501, a matching item data storage unit 503, a narrowing unit 505, a common item data storage unit 507, a similar portion data storage unit 509, and a distance data storage unit 511. And have. The narrowing-down unit 505 includes an item definition comparison unit 5051 and an appearance part comparison unit 5053.

  The matching item extraction unit 501 generates matching item data based on the data stored in the input data storage unit 3 and the data stored in the similar design document storage unit 11 and stores the matching item data in the matching item data storage unit 503. The narrowing-down unit 505 specifies item candidates from the data stored in the matching item data storage unit 503 based on the data stored in the input data storage unit 3 and the similar design document storage unit 11, and the item candidate storage unit 13 To store. The item definition comparison unit 5051 compares the item name stored in the similar design document storage unit 11 and the item name stored in the input data storage unit 3 with respect to the data stored in the matching item data storage unit 503. Based on the list, an item definition comparison process, which will be described later, is performed, and the processing result is stored in the item candidate storage unit 13. The appearance part comparison unit 5053 matches an item name included in the item name list stored in the input data storage unit 3 among the item names included in the item name list stored in the similar design document storage unit 11. The name is stored in the common item data storage unit 507. In addition, the appearance part comparison unit 5053 generates similar part data based on the independent word list stored in the similar design document storage unit 11 and the data stored in the common item data storage unit 507, and the similar part data storage unit 509 is stored. Furthermore, the appearance part comparison unit 5053 generates distance data based on the data stored in the similar part data storage unit 509 and the data stored in the matching item data storage unit 503 and stores the distance data in the distance data storage unit 511. .

  FIG. 3 shows an example of data stored in the design document DB 7. In the example of FIG. 3, a screen ID, a process name, a process detail file name, an item definition file name, a process detail free word list, and an item definition item name list are stored. . Thus, the design document DB 7 stores data for each design document specified by the screen ID. Note that the independent word list and the item name list are generated by the method described below.

  A method for generating an independent word list and an item name list will be briefly described with reference to FIGS. First, a method for generating an independent word list will be described with reference to FIG. In the present embodiment, by performing morphological analysis on the processing details, the sentence is decomposed into morphemes, and the part of speech of each morpheme is determined. For example, when the morphological analysis is performed on the processing details as shown in FIG. 4A, an analysis result as shown in FIG. 4B is obtained. Furthermore, a self-supporting word is extracted by performing a process of connecting the morpheme to the compound word with respect to the analysis result of the morpheme analysis. For example, when the concatenation process is performed on the analysis result as shown in FIG. 4B, a list of independent words as shown in FIG. 4C is obtained. In addition, since the method of extracting an independent word by such a process is a well-known technique, it is not described in detail here.

  A method for generating an item name list will be described with reference to FIG. The item name list is generated by extracting item names from the item definition. For example, when processing is performed on the item definition as shown in FIG. 5A, an item name list as shown in FIG. 5B is generated. The number attached to the left of the item name is the item ID and is automatically assigned.

  Next, processing contents of the inconsistency detection apparatus shown in FIG. 2 will be described with reference to FIGS. First, the input data processing unit 1 of the inconsistency detection apparatus receives processing details and item definition input of a design document to be diagnosed and stores it in a storage device such as a main memory (FIG. 6: Step S1). Here, the input may be received directly from the user via an input device such as a keyboard or a mouse, the list of design documents is presented to the user, and the processing details and item definitions of the user-specified design document are not shown. The processing may be acquired from a storage device or the like (or another computer connected via a network).

  Then, the input data processing unit 1 extracts an independent word from the input process details to generate an independent word list, extracts an item name from the input item definition, generates an item name list, and inputs data Store in the storage unit 3 (step S3). Since the method for generating the independent word list and the item name list has already been described with reference to FIGS. 4 and 5, the description thereof is omitted here. The input data storage unit 3 stores an independent word list in the same data format as in FIG. 4C, and stores an item name list in the same data format as in FIG. 5B.

  And the similar design document specific | specification part 9 implements a similar design document specific process (step S5). The similar design document specifying process will be described in detail with reference to FIG. The similar design document specifying unit 9 acquires one unprocessed independent word list from the design document DB 7 (FIG. 7: Step S21). Further, the similar design document specifying unit 9 calculates the similarity between the acquired independent word list and the independent word list stored in the input data storage unit 3 (step S23). In step S23, for example, a value obtained by dividing the number of independent words included in both independent word lists by the number of independent words included in the independent word list stored in the input data storage unit 3 is calculated as the similarity. To do. Further, the similar design document specifying unit 9 determines whether the similarity is equal to or greater than a predetermined threshold (step S25).

  When the similarity is less than the predetermined threshold (step S25: No route), the process proceeds to step S29. On the other hand, when the similarity is equal to or greater than a predetermined threshold (step S25: Yes route), the similar design document specifying unit 9 stores the acquired independent word list and the item name list corresponding to the independent word list in the similar design document. Stored in the unit 11 (step S27). The item name list is acquired from the design document DB 7. The similar design document storage unit 11 stores data in the same data format as the design document DB 7.

  And the similar design document storage part 11 judges whether it processed about all the independent word lists in design document DB7 (step S29). When all the independent word lists have not been processed (step S29: No route), the process returns to step S21 in order to execute the process for the next independent word list. On the other hand, when all the independent word lists have been processed (step S29: Yes route), the processing returns to the original processing.

  By carrying out the processing as described above, it is possible to appropriately specify a design document whose contents are similar to the design document to be diagnosed.

  Returning to the description of FIG. 6, the item candidate extraction unit 5 performs an item candidate extraction process (step S <b> 7). The item candidate extraction process will be described in detail with reference to FIGS. First, the matching item extraction unit 501 of the item candidate extraction unit 5 identifies one unprocessed independent word from the independent word list stored in the input data storage unit 3 (FIG. 8: step S31). Further, the matching item extraction unit 501 specifies one unprocessed item name list from the similar design document storage unit 11 (step S33). Further, the matching item extraction unit 501 specifies one unprocessed item name from the item name list specified in step S33 (step S35). Then, the matching item extraction unit 501 calculates the similarity of the notation between the independent word specified in step S31 and the item name specified in step S35 (step S37). In step S37, the similarity is calculated using, for example, a technique disclosed in Japanese Patent Laid-Open No. 6-83871.

  Then, the matching item extraction unit 501 determines whether the similarity calculated in step S37 is greater than or equal to a predetermined threshold (step S39). When the similarity calculated in step S37 is less than a predetermined threshold (step S39: No route), the process proceeds to step S43.

  On the other hand, when the similarity calculated in step S37 is equal to or greater than a predetermined threshold (step S39: Yes route), the matching item extraction unit 501 includes a match including the independent word and item name whose similarity is calculated in step S37. Item data is generated and stored in the matching item data storage unit 503 (step S41).

  FIG. 9 shows an example of data stored in the matching item data storage unit 503. The example of FIG. 9 includes an independent word ID column, an independent word column, a similar design document ID column, an item ID column, an item name column, and a coincidence column. In the column of the independent word ID, for example, a number assigned according to the number of the independent word specified in step S31 is stored. In the similar design document ID and item ID column, the screen ID and item ID corresponding to the item name specified in step S35 are acquired from the similar design document storage unit 11 and stored. The degree of coincidence is “complete match” when the degree of similarity calculated in step S37 is 1, that is, the independent word and the item name completely match, and the degree of similarity calculated in step S37 is a predetermined threshold value. If it is less than 1, it is determined as “partial match”. For example, in FIG. 9, since the independent word “office code” and the item name “office code” are the same word, “complete match” is stored in the match degree column. Further, since the independent word “establishment name” and the item name “establishment name” have high similarity but are not the same word, “partial match” is stored in the coincidence column. On the other hand, when the item name is “payment / payment number” with respect to the independent word “office division name”, the similarity calculated in step S37 does not satisfy the predetermined threshold value, and therefore the matching item data storage unit 503 stores the similarity. Does not store any data.

  Returning to the description of FIG. 8, the matching item extraction unit 501 determines whether all item names have been processed (step S43). If all the item names have not been processed (step S43: No route), the process returns to step S35 in order to execute the process for the next item name.

  On the other hand, when all item names have been processed (step S43: Yes route), the matching item extraction unit 501 determines whether all item name lists have been processed (step S45). If all the item name lists have not been processed (step S45: No route), the process returns to step S33 to execute the process for the next item name list.

  On the other hand, when all item name lists have been processed (step S45: Yes route), the matching item extraction unit 501 determines whether all independent words have been processed (step S47). If all the independent words are not processed (step S47: No route), the process returns to the process of step S31 in order to execute the process for the next independent word.

  On the other hand, when all independent words have been processed (step S47: Yes route), the process proceeds to the process of step S49 (FIG. 10) via the terminal A.

  As described above, a matching item having a similarity equal to or higher than a predetermined threshold is first identified as an item name that may become an item candidate. Then, candidate items are specified by further narrowing down as described below.

  Shifting to the description of FIG. 10, the narrowing-down unit 505 identifies one unprocessed independent word from the matching item data storage unit 503, and the item name, similar design document ID, and matching that are associated with the independent word The degree is specified and stored in a storage device such as a main memory (FIG. 10: Step S49). Then, the matching item data storage unit 503 determines whether or not the identified matching degree includes a matching degree that is “perfect matching” (step S51). When the degree of coincidence that is “complete match” is included (step S51: Yes route), the narrowing-down unit 505 stores the item name that is “complete match” in the item candidate storage unit 13 (step S63). For example, in FIG. 9, the item name associated with the independent word “payment / payment number” includes “payment / payment number”. Since the matching degree of this item name is “complete match”, the item name “payment / payment number” Is stored in the item candidate storage unit 13. This is because if the item name defined in the similar design document completely matches the independent word in the processing details, it is highly likely that the item name should be defined in the design document to be diagnosed.

  FIG. 11 shows an example of data stored in the item candidate storage unit 13. In the example of FIG. 11, item names are stored.

  Returning to the description of FIG. 10, when the degree of coincidence “completely coincident” is not included (step S51: No route), the narrowing-down unit 505 determines whether there is one item name identified in step S49 ( Step S53). When the number of item names specified in step S49 is one (step S53: Yes route), the narrowing down unit 505 stores the item name specified in step S49 in the item candidate storage unit 13 (step S63). For example, in FIG. 9, the item name associated with the self-supported word “establishment division name” is only the item name “establishment division” having the degree of coincidence “partial match”. It is stored in the storage unit 13. This is because it is not necessary to narrow down item names if only one item name is specified even if the degree of coincidence is “partial match”.

  On the other hand, when it is determined that there is not one item name specified in step S49 (step S53: No route), the narrowing unit 505 determines that the item name that matches the item name specified in step S49 is input data. It is determined whether it is included in the item name list stored in the storage unit 3 (step S55). If included (step S55: Yes route), the narrowing-down unit 505 stores the item names included in the item name list stored in the input data storage unit 3 in the item candidate storage unit 13 (step S63). For example, in FIG. 9, the independent name “office name” is associated with the item names “office name” and “own office name”, but in the item name list stored in the input data storage unit 3 When the item name includes "Establishment Code", "Establishment Name", "Establishment Category", "Acceptance Number", "Payment Number", "Owner Site Code", and "Owner Site Name" , The item name “business establishment name” is stored in the item candidate storage unit 13. This is because an item name already defined in the design document to be diagnosed should be identified as an item candidate. Even if the item is stored in the item candidate storage unit 13 in step S63, it is determined that the item name is included in the item name list stored in the input data storage unit 3 in the process of step S9 described later. Therefore, it is not specified as an inconsistent item.

  On the other hand, when not included (step S55: No route), the narrowing down unit 505 determines whether or not the similar design document ID specified in step S49 includes a plurality of identical similar design document IDs (step S57). When a plurality of the same similar design document IDs are included (step S57: Yes route), the appearance part comparison unit 5053 of the narrowing-down unit 505 performs an appearance part comparison process (step S59). For example, in FIG. 9, the similar design document ID associated with the independent word “payment state” is “G002” on the 6th line, “G003” on the 7th line, “G004” on the 8th line, and 9th line. Since it is “G004”, the appearance partial comparison process is performed for the independent word “payment / payment state”. For convenience, among the item names identified in step S49, the item name corresponding to the similar design document ID determined to include a plurality of IDs in step S57 is referred to as an ID duplicate item name. In the example of FIG. 9, “payment / payment source state” and “total payment / payment state” correspond to ID duplication item names. The appearance part comparison process is a process for selecting an item name that is most suitable as an item candidate from among ID duplicated item names.

  Here, the appearance portion comparison processing will be described in detail with reference to FIGS. First, the appearance portion comparison unit 5053 selects one unprocessed similar design document ID (hereinafter referred to as a similar design document ID to be processed) from among similar design document IDs determined to include a plurality of IDs in step S57. It identifies (FIG. 12: step S81). Then, the appearance portion comparison unit 5053 identifies an item name list corresponding to the similar design document ID to be processed from the similar design document storage unit 11, and identifies one unprocessed item name from the item name list (step S1). S83). In step S83, the item ID corresponding to the unprocessed item name is also specified from the similar design document storage unit 11.

  Then, the appearance portion comparison unit 5053 determines whether or not the item name that matches the item name specified in step S83 is included in the item name list stored in the input data storage unit 3 (step S85). If not included (step S85: No route), the process proceeds to step S89.

  On the other hand, when it is included (step S85: Yes route), the appearance portion comparison unit 5053 stores the item name and item ID specified in step S83 in the common item data storage unit 507 (step S87).

  FIG. 13 shows an example of data stored in the common item data storage unit 507. In the example of FIG. 13, an item ID column and an item name column are included.

  Returning to the description of FIG. 12, the appearance portion comparison unit 5053 determines whether all item names have been processed (step S89). If all the item names have not been processed (step S89: No route), the process returns to step S83 in order to execute the process for the next item name. On the other hand, if all item names have been processed (step S89: Yes route), the process proceeds to the process of step S91 (FIG. 14) via the terminal B.

  Shifting to the description of FIG. 14, the appearance portion comparison unit 5053 identifies an independent word list corresponding to the similar design document ID to be processed from the similar design document storage unit 11, and an unprocessed independent word from the independent word list. Is identified (step S91). In step S91, an independent word ID is assigned to the specified independent word. The independent word ID is a number assigned according to the number of the independent word specified in step S91, for example.

  Then, the appearance portion comparison unit 5053 calculates the similarity of the notation between the independent word specified in step S91 and each item name stored in the common item data storage unit 507 (step S93). The method of calculating the similarity in step S93 is the same as that in step S37. In addition, the appearance portion comparison unit 5053 determines whether or not the maximum similarity among the calculated similarities is equal to or greater than a predetermined threshold (step S95). For example, let us consider a case where the independent word specified in step S91 is “partner business name” and the common item data storage unit 507 stores the data shown in FIG. In step S93, it is assumed that the similarity calculated for the item name “partner office code” is 0.71, and the similarity calculated for the item name “partner office name” is 0.92. In this case, in step S95, it is determined whether the similarity 0.92 calculated for the item name “partner office name” is equal to or greater than a predetermined threshold.

  If the maximum similarity among the calculated similarities is greater than or equal to a predetermined threshold (step S95: Yes route), the appearance portion comparison unit 5053 displays the independent word and item name for which the maximum similarity is calculated. The included data is stored in the similar partial data storage unit 509 (step S97). On the other hand, when the maximum similarity is less than the predetermined threshold among the calculated similarities (step S95: No route), the appearance portion comparison unit 5053 includes data including the independent word for which the maximum similarity is calculated. It stores in the similar partial data storage unit 509 (step S99).

  FIG. 15 shows an example of data stored in the similar partial data storage unit 509. The example of FIG. 15 includes an independent word ID column, an independent word column, an item ID column, an item name column, and a similar part ID column. In the example of FIG. 15, the data of the fourth, sixth, 18th and 21st lines are stored in step S97, and the data of the other lines are stored in step S99. It should be noted that no data is stored in the similar part ID column when the process of step S103 described later is not performed.

  Returning to the description of FIG. 14, the appearance portion comparison unit 5053 determines whether or not all independent words have been processed (step S101). When all the independent words are not processed (step S101: No route), the process returns to the process of step S91 in order to execute the process for the next independent word. On the other hand, when all the independent words are processed (step S101: Yes route), the appearance portion comparison unit 5053 identifies a similar portion in the similar portion data storage unit 509, and associates the similar portion ID with the independent portion of the similar portion. And stored in the similar partial data storage unit 509 (step S103). Then, the process proceeds to step S105 (FIG. 16) via the terminal C.

  In step S103, for example, a determination is made based on whether n or more item names are stored in association with m consecutive independent words. For example, consider the case where m = 5 and n = 2 are set in FIG. First, there are three independent words from the independent word with the independent word ID “4” to the independent word with the independent word ID “6”, and the item names associated with these independent words are “the partner office name”. ”And“ partner office code ”. Therefore, the parts from the independent word IDs “4” to “6” are specified as similar parts. Moreover, since there are 13 independent words from the independent word with the independent word ID “6” to the independent word with the independent word ID “18”, the “partner establishment code” and the “partner establishment name” are included in these independent words. Are associated with each other, they are not identified as similar parts. In addition, there are four independent words from the independent word with the independent word ID “18” to the independent word with the independent word ID “21”, and the item names associated with these independent words are “the partner office name”. ”And“ partner office code ”. Accordingly, the parts from the independent words ID “18” to “21” are specified as similar parts. In step S103, if it is desired to specify many similar parts, m is set to be large or n is set to be small. If it is not desired to specify similar parts very much, m is set to be small or n is set to be large. May be set in advance.

  Shifting to the description of FIG. 16, the appearance portion comparison unit 5053 identifies one unprocessed independent word stored in the similar portion data storage unit 509 (FIG. 16: step S105). Then, the appearance part comparison unit 5053 calculates the similarity of the notation between the identified independent word and each ID duplicate item name (step S107). The method for calculating the similarity in step S107 is the same as in step S37.

  Then, the appearance portion comparison unit 5053 determines whether the maximum similarity among the calculated similarities is equal to or greater than a predetermined threshold (step S109). When it is less than the predetermined threshold (step S109: No route), the process proceeds to step S115. On the other hand, when it is equal to or greater than the predetermined threshold (step S109: Yes route), the appearance portion comparison unit 5053 calculates the distance between the independent word specified in step S105 and the preceding and following similar portions in the similar portion data storage unit 509. (Step S111). In addition, the appearance portion comparison unit 5053 generates distance data including the smallest one of the calculated distances and the ID duplication item name used for calculating the maximum similarity among the similarities calculated in step S107. And stored in the distance data storage unit 511 (step S113). In addition, when the independent word specified in step S105 is included in the similar part, the distance is set to “0”.

  The processing in steps S105 to S113 will be briefly described using a specific example. For example, the similar partial data is the data shown in FIG. 15, the self-supporting word specified in step S105 is the “payment / payment source state” of the self-supporting word ID “8”, and the ID duplicate item name is the eighth line in FIG. Consider a case in which the item name is on the ninth line (that is, “payment source state” and “total payment state”). In step S107, the similarity calculated for the duplicate ID item name “payment / payment source state” is 1.0, and the similarity calculated for the duplicate ID item name “total payment / payment state” is 0.8. . In this case, in step S109, it is determined whether or not the similarity 1.0 is equal to or greater than a predetermined threshold. However, the similarity 1.0 indicates a perfect match and is equal to or greater than the predetermined threshold. Therefore, in step S111, the “partner office code” (independent word) that is the closest independent word among the independent words included in the similar part SP1 from the independent word “payment / payment source state” having the independent word ID “8”. ID “6”) and the independent word “8” that is the closest independent word from the independent word “payment source state” having the independent word ID “8” included in the similar part SP2. A distance “10” to “place name” (independent word ID “18”) is calculated. In step S 113, distance data including the distance “2” and the ID duplication item name “payment / payment source state” is generated and stored in the distance data storage unit 511.

  FIG. 17 shows an example of data stored in the distance data storage unit 511. In the example of FIG. 17, an independent word ID column, an item ID column, an item name column, and a distance column between similar parts are included.

  Returning to the description of FIG. 16, the appearance portion comparison unit 5053 determines whether or not all independent words have been processed (step S115). If all the independent words have not been processed (step S115: No route), the process returns to step S105 in order to execute the process for the next independent word.

  On the other hand, when all independent words are processed (step S115: Yes route), the appearance portion comparison unit 5053 causes the ID data items other than the ID duplication item name associated with the minimum distance in the distance data storage unit 511. The name is specified, and the data for the specified ID duplicate item name is deleted from the matching item data storage unit 503 (step S117). In the example of FIG. 17, since the item name associated with the minimum distance “2” is “payment / payment source state”, the item name other than “payment / payment source state” is “total payment / payment state”. Is identified. If the data of FIG. 9 is stored in the matching item data storage unit 503 and the ID duplicate item name is “payment source state” on the eighth line and “total payment state” on the ninth line, the ninth line The data on “total payment state” is deleted.

  Returning to the description of FIG. 16, the appearance portion comparison unit 5053 determines whether or not all similar design document IDs have been processed (step S119). When all the similar design document IDs have not been processed (step S119: No route), the process returns to step S81 via the terminal D in order to perform the process for the next similar design document ID. On the other hand, when all the similar design document IDs are processed (step S119: Yes route), the process returns to the original process.

  By performing the above processing, even if multiple item names are extracted from the item name list of one similar design document for one independent word, the item name most suitable as an item candidate Can be identified.

  Returning to the explanation of FIG. 10, when a plurality of identical similar design document IDs are not included (step S57: No route), the item definition comparing unit 5051 of the narrowing down unit 505 performs an item definition comparing process (step S61). For example, in FIG. 9, the similar design document IDs associated with the independent word “payment / payment state classification” are “G002” and “G003”, and a plurality of identical similar design document IDs are not included. Item definition comparison processing is performed for “payment / payment status classification”.

  The item definition comparison process will be described in detail with reference to FIG. First, the item definition comparison unit 5051 identifies one unprocessed similar design document ID from the similar design document IDs identified in step S49 (FIG. 18: step S121). Then, the item name list corresponding to the specified similar design document ID is specified from the similar design document storage unit 11, and the similarity between the specified item name list and the item name list stored in the input data storage unit 3 Is calculated and stored in a storage device such as a main memory (step S123). In step S123, for example, the number of item names commonly included in both item name lists is set as the similarity. For example, the item name “office code”, “acceptance number”, “acceptance number”, “acceptance status”, “acceptance status category”, “business category” and “inventory category” in the item name list specified in step S123. The item name list stored in the input data storage unit 3 includes “establishment code”, “establishment name”, “estate classification”, “acceptance number”, “acceptance number”, “partner” When the “establishment code” and the “partner establishment name” are included, the item names “establishment code”, “acceptance number”, and “payment number” are common, and the similarity is “3”.

  Then, the item definition comparison unit 5051 determines whether all similar design document IDs have been processed (step S125). When all the similar design document IDs are not processed (step S125: No route), the process returns to step S121 in order to perform the process for the next similar design document ID.

  On the other hand, when processing is performed for all similar design document IDs (step S125: Yes route), the item definition comparison unit 5051 has the maximum similarity calculated in step S123 among the item names specified in step S49. The item name corresponding to the similar design document ID is stored in the item candidate storage unit 13, and the process returns to the original process (step S127). For example, in FIG. 9, the item name list of the similar design document whose similarity degree calculated for the item name list of the similar design document whose similar design document ID is “G002” is 5 and whose similar design document ID is “G003”. When the similarity calculated for the item is 3, the item name “payee status category” among the item names “payee status category” and “acceptance status category” associated with the independent word “payment status category” Are stored in the item candidate storage unit 13.

  By performing the above processing, even when item names are extracted from a list of item names of a plurality of similar design documents for one independent word, it is most suitable for extracting item names. The item name list can be determined.

  Returning to the description of FIG. 10, after step S63 or step S61, the narrowing-down unit 505 determines whether or not all independent words have been processed (step S65). When all the independent words have not been processed (step S65: No route), the process returns to step S49 in order to execute the process for the next independent word. On the other hand, if all independent words have been processed (step S65: Yes route), the process returns to the original process.

  By performing the processing as described above, it is possible to appropriately extract item names that should be item candidates.

  Returning to the description of FIG. 6, the first inconsistent item specifying unit 15 includes items that are not included in the item name list stored in the input data storage unit 3 among the item names stored in the item candidate storage unit 13. An inconsistent item name that is a name is specified and stored in the first inconsistent item storage unit 17 (step S9). For example, when the data shown in FIG. 11 is stored in the item candidate storage unit 13 and the data shown in FIG. 5B is stored in the input data storage unit 3, the item name “payee status” and The “payee status classification” is stored in the first inconsistent item storage unit 17.

  And the output part 19 implements the process which outputs the item name stored in the 1st inconsistent item storage part 17 (step S11). When a display device or a printing device is connected to the mismatch detection device, it may be displayed on the display device or output to the printing device. Furthermore, the data may be output to another computer connected via a network.

  As mentioned above, since the item name list of similar design documents is used, it is possible to specify the item name to be defined with high accuracy while preventing the independent word that is not the item name from being defined as the item name. It becomes like this.

[Embodiment 2]
Next, you should detect the item name that has been inconsistent (second inconsistency) that is defined in the item definition but not included in the processing details, and replenish the data related to the item name. Processing for specifying the position will be described.

  FIG. 19 shows a functional block diagram of the correction support apparatus according to the second embodiment. The correction support apparatus according to the second embodiment includes an input processing unit 101, an input data storage unit 103, a design document DB 105, a similar design document specifying unit 107, a similar design document storage unit 113, and a second memory. The matching item specifying unit 109, the second inconsistent item storage unit 111, the supplementary sentence extraction unit 115, the supplementary sentence storage unit 117, the narrowing processing unit 119, the narrowing processing result storage unit 121, and the supplementary position determination unit 123. And an output data storage unit 125 and an output unit 127.

  The input processing unit 101 receives processing details and item definition input of a design document to be diagnosed, and stores it in the input data storage unit 103. The similar design document specifying unit 107 extracts the similar design document from the design document DB 105 based on the data stored in the input data storage unit 103 and the design document DB 105 and stores the similar design document in the similar design document storage unit 113. The second inconsistent item specifying unit 109 extracts the inconsistent item name using the data stored in the input data storage unit 103 and stores it in the second inconsistent item storage unit 111. The supplementary sentence extraction unit 115 supplements the input processing details using the inconsistent item name stored in the second inconsistent item storage unit 111 and the processing details stored in the similar design document storage unit 113. A process for extracting power data is performed and stored in the supplementary sentence storage unit 117. The narrowing processing unit 119 performs range narrowing processing, which will be described later, using data stored in the input data storage unit 103, the similar design document storage unit 113, and the supplementary sentence storage unit 117, and narrows down the processing result. The result is stored in the result storage unit 121. The replenishment position determination unit 123 uses the data stored in the input data storage unit 103, the similar design document storage unit 113, and the narrowing-down process result storage unit 121 to perform a replenishment position determination process to be described later. Is stored in the output data storage unit 125. The output unit 127 outputs the data stored in the output data storage unit 125 to the display device.

  The narrowing processing unit 119 includes a line number list storage unit 1191, a corresponding position data storage unit 1192, a line number list generation unit 1193, a corresponding position specifying unit 1194, and a range specifying unit 1195. The line number list generation unit 1193 uses the data stored in the input data storage unit 103 and the similar design document storage unit 113 to perform a process of generating a line number list, which will be described later, and outputs the processing result. Stored in the number list storage unit 1191. The corresponding position specifying unit 1194 uses the data stored in the line number list storage unit 1191 and the input data storage unit 103 to perform a corresponding position specifying process, which will be described later, and the processing result is stored in the corresponding position data storage unit. 1192. The range specifying unit 1195 performs processing based on the data stored in the supplementary sentence storage unit 117 and the corresponding position data storage unit 1192 and stores the processing result in the narrowing processing result storage unit 121.

  The replenishment position determination unit 123 includes a similarity storage unit 1231, a window generation unit 1232, and a search unit 1233. The window generation unit 1232 performs processing for generating window data, which will be described later, and outputs the generated window data to the search unit 1233. The search unit 1233 performs processing using the window data received from the window generation unit 1232 and the data stored in the input data storage unit 103, and stores the processing result in the output data storage unit 125.

  20 and 21 show an example of data stored in the design document DB 105. FIG. FIG. 20 shows an example of item definition data stored in the design document DB 105. In the example of FIG. 20, an item definition document ID, a file name, and a process name are stored, and a number column, an item name column, an item type column, Columns and I / O (input / output) columns are included. FIG. 21 shows an example of processing detail data stored in the design document DB 105. In the example of FIG. 21, the detailed processing definition document ID and the file name are stored, and a row number column and a statement column representing the contents of the processing are included. In the design document DB 105, item definitions and process details are stored as a set for each process. For example, FIG. 20 and FIG. 21 show item definitions and processing details for the process of “stock receipt registration”.

  Next, processing contents of the correction support apparatus shown in FIG. 19 will be described with reference to FIGS. First, the input processing unit 101 of the correction support apparatus receives processing details and item definition input of a design document to be diagnosed and stores it in the input data storage unit 103 (FIG. 22: step S201). Here, the input may be received directly from the user via an input device such as a keyboard or a mouse, the list of design documents is presented to the user, and the processing details and item definitions of the user-specified design document are not shown. The processing may be acquired from a storage device or the like (or another computer connected via a network). Further, item definitions and processing details may be extracted from the design document DB 105.

  23 and 24 show an example of data stored in the input data storage unit 103. FIG. FIG. 23 shows an example of item definition data stored in the input data storage unit 103, and the data format is the same as the data shown in FIG. FIG. 24 shows an example of detailed processing data stored in the input data storage unit 103, and the data format is the same as the data shown in FIG.

  Returning to the description of FIG. 20, the second inconsistent item specifying unit 109 selects the input data among the item names defined in the item definition (hereinafter referred to as input item definition) stored in the input data storage unit 103. An inconsistent item name not included in the processing details stored in the storage unit 103 (hereinafter referred to as input processing details) is specified and stored in the second inconsistent item storage unit 111 (step S203). The inconsistent item name is an item name in which data regarding the processing content is missing in the processing details although it is defined in the item definition.

  Then, the similar design document specifying unit 107 reads out the process details having the highest similarity to the input process details and the item definition corresponding to the process details from the design document DB 105, and stores them in the similar design document storage unit 113 (step S205). ).

  In step S205, for example, a process similar to the process performed in steps S3 and S5 described in the first embodiment may be performed. In this case, the similar design document specifying unit 107 extracts an independent word from the input processing details to generate an independent word list, and also generates an independent word list for each of the processing details stored in the design document DB 105. In addition, the similar design document specifying unit 107 calculates, for each processing detail stored in the design document DB 105, a similarity between the independent word list of the processing details and the independent word list of the input processing details. Then, the similar design document specifying unit 107 reads out the process details having the highest calculated similarity and the item definition corresponding to the process details from the design document DB 105 and stores them in the similar design document storage unit 113. Hereinafter, the process details stored in the similar design document storage unit 113 are referred to as similar process details, the item definitions stored in the similar design document storage unit 113 are referred to as similar item definitions, and a set of similar process details and similar item definitions is set. Is called a similar design document.

  Next, the supplementary sentence extraction unit 115 searches for similar processing details using the inconsistent item name stored in the second inconsistent item storage unit 111, and includes the sentence including the identified inconsistent item name and the line of the sentence. The number is stored in the supplementary sentence storage unit 117 (step S207).

  FIG. 25 shows an example of data stored in the supplementary sentence storage unit 117. In the example of FIG. 25, the line number and the sentence to be supplemented are stored.

  Returning to the description of FIG. 22, the narrowing-down processing unit 119 performs the range narrowing-down process (step S209). The range narrowing process will be described with reference to FIGS.

  First, the line number list generation unit 1193 of the narrowing processing unit 119 extracts item names included in both the input item definition and the similar item definition as common item names and stores them in a storage device such as a main memory (FIG. 26: Step S221). For example, when the similar item definition is data as shown in the example of FIG. 20 and the input item definition is data as shown in the example of FIG. 23, the data as shown in FIG. 27 is common. Extracted as an item name.

  Then, the line number list generation unit 1193 generates the line number list by extracting the common item name extracted in step S221 and the line number of the sentence including the common item name for each of the input process details and the similar process details. Then, it is stored in the line number list storage unit 1191 (step S223).

  FIG. 28 shows an example of data stored in the line number list storage unit 1191. In the example of FIG. 28, the ID, line number, and item name of the detailed processing definition document are stored. The data shown on the left side of FIG. 28 shows an example of the line number list generated for the input process details, and the data shown on the right side shows an example of the line number list generated for the similar process details.

  Returning to the description of FIG. 26, the corresponding position specifying unit 1194 extracts the item name next to the last extracted item name from the line number list for the input process details and the line number list for the similar process details (that is, the last extracted item name). Each item name having the next smallest row number) is extracted and stored in a storage device such as a main memory (step S225). When the process of step S225 is performed for the first time, the item name with the smallest row number is extracted. Hereinafter, the item name extracted from the line number list for the input process details is referred to as a first item name, and the item name extracted from the line number list for the similar process details is referred to as a second item name.

  Then, the corresponding position specifying unit 1194 determines whether the first item name and the second item name are extracted (step S227). For example, when the item name having the largest line number has already been processed, the item name is not extracted because there is no item name to be processed next. When the first item name and the second item name are extracted (step S227: Yes route), the corresponding position specifying unit 1194 determines whether the first item name and the second item name are the same. (Step S229). When the first item name and the second item name are the same (step S229: Yes route), the corresponding position specifying unit 1194 sets the line number corresponding to the item name to the line for the input processing details. The number is identified from the number list and the line number list for details of similar processing, and stored in the corresponding position data storage unit 1192 in association with the item name (step S231).

  FIG. 29 shows an example of data stored in the corresponding position data storage unit 1192. 29 includes an order column, an item name column, a row number column in the input processing details, and a row number column in the similar processing details.

  Returning to FIG. 26, when the first item name and the second item name are not the same (step S229: No route), the corresponding position specifying unit 1194 performs the corresponding position specifying process (step S233). The corresponding position specifying process will be described with reference to FIG.

  First, the corresponding position specifying unit 1194 specifies the last stored item name from the corresponding position data storage unit 1192 (FIG. 30: Step S241). In addition, the corresponding position specifying unit 1194 selects an item name defined after the item name specified in step S241 from the input item definition (that is, an item name having a “number” larger than the specified item name). Extraction is performed, and a candidate list including the extracted item names is generated (step S243). In addition, the corresponding position specifying unit 1194 specifies the item name defined first (that is, the item name having the smallest “number”) among the unprocessed item names included in the candidate list (step S245). .

  Then, the corresponding position specifying unit 1194 determines whether the item name specified in step S245 is the same as either the first item name or the second item name (step S247). When it is determined that neither the first item name nor the second item name is the same (step S247: No route), the process returns to the process of step S245 to execute the process for the next item name.

  On the other hand, when it is determined that it is the same as either the first item name or the second item name (step S247: Yes route), the corresponding position specifying unit 1194 determines that the item name specified in step S245 is the first item name. It is determined whether it is the same as the item name of 1 (step S249).

  When it is determined that the name is the same as the first item name (step S249: Yes route), the corresponding position specifying unit 1194 has an item whose line number is larger than the second item name in the line number list for similar processing details. An item name identical to the first item name is searched from the names, and the item name and the line number corresponding to the item name are extracted (step S251).

  On the other hand, when it is determined that it is not the same as the first item name (that is, the same as the second item name) (step S249: No route), the corresponding position specifying unit 1194 performs a line for details of the input process. In the number list, an item name having the same line number as the second item name is searched from item names having a line number larger than that of the first item name, and the item name and the line number corresponding to the item name are extracted (step) S253).

  Then, the corresponding position specifying unit 1194 associates the item name specified in step S245 with the line number corresponding to the item name and stores it in the corresponding position data storage unit 1192 (step S255). If it is determined in step S249 that the item name specified in step S245 is the same as the first item name, the line number corresponding to the first item name in the line number list for the input processing details is The line number extracted in step S251 is stored. If it is determined in step S249 that the item name specified in step S245 is not the same as the first item name (that is, the same as the second item name), the line number for the similar process details In the list, the line number corresponding to the second item name and the line number extracted in step S253 are stored. Then, the process returns to the original process.

  Returning to the description of FIG. 26, if the item name is not extracted from any of the line number lists (step S227: No route), the corresponding position specifying unit 1194 reads the line number and corresponding position data of the sentence to be supplemented. Based on the data stored in the storage unit 1192, a process for specifying a range in which a sentence to be supplemented is specified is performed, and the processing result is stored in the narrowing-down processing result storage unit 121 (step S235). Then, the process returns to the original process.

  FIG. 31 shows an example of data stored in the narrowing processing result storage unit 121. The example of FIG. 31 includes an order column, an item name column, a row number column in the input processing details, and a row number column in the similar processing details.

  Here, the process performed in step S235 will be described. For example, as shown in FIG. 25, it is assumed that the line number of the sentence to be supplemented is “331”. On the other hand, it is assumed that the corresponding position data storage unit 1192 stores data as shown in FIG. Then, it can be seen that the sentence to be supplemented is located between the line including the common item name “operation date” and the line including the common item name “office code” in the similar process details. Therefore, in the present embodiment, as shown in FIG. 32, the input processing details are sandwiched between a line including the common item name “operation date” and a line including the common item name “office code”. It is estimated that there is a position to fill in the missing data in the range. This uses the property that the processing details similar to the input processing details (similar processing details) are similar in the order in which the item names appear.

  In addition, a description of the processing performed by the corresponding position specifying unit 1194 will be added. FIG. 33 is a diagram schematically showing line number association performed by the corresponding position specifying unit 1194. First, the corresponding position specifying unit 1194 determines whether the item names included in the respective line number lists are the same in order from the top (that is, in the order from the smallest row number). In the example of FIG. 33, first, it is determined whether the “operation date” in the line number list for the input process details is the same as the “operation date” in the line number list for the similar process details. Here, since they are the same, the corresponding position data storage unit 1192 stores data as in the first row. The same processing is performed on the “item code” that is the next item name and the “operation date” that appears thereafter.

  Then, as the next item name, “establishment code” is specified from the line number list for the input process details, and “partner classification” is specified from the line number list for the similar process details. Since these item names are not the same, the corresponding position specifying unit 1194 performs the corresponding position specifying process described above. For example, if the input item definition data is as shown in FIG. 23, the “establishment code” is defined before the “business partner classification”, so the “establishment code” has priority. Process. Here, in the line number list for the details of similar processing, the “establishment code” is identified from the item names located after the “partner classification”, and the line number “337” corresponding to this item name is input. Corresponding to the line number “294” in the line number list for processing details.

  As described above, in the present embodiment, when the item names extracted from the respective line number lists are not the same, the item name defined earlier in the input item definition is preferentially processed. Yes.

  On the other hand, in the example of FIG. 33, in the case where processing is performed with priority given to the item name “partner classification” in the line number list for similar processing details without considering the order in which the item names are defined in the input item definition explain. In this case, in the line number list for the details of the input process, the “partner category” is identified from the item names located after the “office code”, and the line number “421” corresponding to this item name is Corresponding to the line number “321” in the line number list for similar processing details. The corresponding position data obtained as a result is as shown in FIG. 34, for example. According to this data, the sentence to be supplemented should be supplemented between the line number “421” and the line number “439” in the input processing details. However, this has resulted in incorrect refinement.

  Therefore, in order to reduce the possibility of such an erroneous narrowing, in this embodiment, line numbers are associated in a form that considers the order in which item names are defined in the input item definition. Yes.

  By performing the range narrowing process as described above, it is possible to narrow down the range where it is estimated that data is missing in the input process details with high accuracy.

  Returning to the description of FIG. 22, the replenishment position determination unit 123 performs a replenishment position determination process (step S211). First, the first replenishment position determination process will be described with reference to FIGS.

  First, the window generation unit 1232 of the replenishment position determination unit 123 sets the window width n to 1 (FIG. 35: step S261). Then, the window generation unit 1232 reads the data of the line number of the sentence to be supplemented from the supplementary sentence storage unit 117, extracts n lines before and after the sentence to be supplemented from the details of similar processing, and generates window data. The data is stored in a storage device such as a memory (step S263).

  FIG. 36 shows an example of window data. In the example of FIG. 36, the window ID, the line number, and the data of n sentences before and after the sentence to be supplemented (here, n = 3) are stored.

  Returning to the description of FIG. 35, the search unit 1233 has the highest degree of similarity with the generated window data in the range where the data is estimated to be missing in the input process details (the range specified in step S235). A high position is searched (step S265).

  The process performed in step S265 will be described with reference to FIG. In FIG. 37, the range from the line number “278” to the line number “294” is shown as the range in which the data is estimated to be missing in the input processing details. For example, when n = 3, first, the similarity between the data included in the range 371 and the window data is calculated. As the similarity, for example, the number of words that match the free words included in the range 371 and the free words included in the window data may be used. Similarly, when the similarities are calculated for the ranges 372 and 373 and the similarities are calculated up to the range 374, the calculation of the similarity is terminated. And the predetermined position (for example, center) in the range corresponding to the highest similarity among the calculated similarities is specified.

  Returning to the description of FIG. 35, the search unit 1233 stores the degree of similarity specified by the search processing in step S <b> 265 and the corresponding degree of similarity in association with the window width n and stores the degree of similarity. The data is stored in the unit 1231 (step S267).

  Next, the search unit 1233 increments the window width n by 1 (step S269). Then, the search unit 1233 determines whether the window width n exceeds the upper limit value (Step S271). It is desirable that the range for extracting window data does not exceed the range specified in step S235. Therefore, the upper limit value of the window width is “6” when, for example, the data shown in FIG. 31 is stored in the narrowing processing result storage unit 121 and the line number of the sentence to be supplemented is “331”. .

  If the window width n does not exceed the upper limit (step S271: No route), the process returns to step S263. On the other hand, when the window width n exceeds the upper limit value (step S271: Yes route), the search unit 1233 determines the position corresponding to the highest similarity among the similarities stored in the similarity storage unit 1231. Data is extracted and stored in the output data storage unit 125 (step S273). Then, the process returns to the original process.

  FIG. 38 shows an example of data stored in the output data storage unit 125. In the example of FIG. 38, the line number of the part to be supplemented and the data of the sentence included in the part to be supplemented are stored. The example of FIG. 38 indicates that data should be supplemented between the sentence with line number “289” and the sentence with line number “290” in the input processing details.

  FIG. 39 is a diagram illustrating a concept of processing performed in step S273. In the first replenishment position determination process, the window width n is changed as n = 1, 2, 3,..., And the maximum similarity is calculated for each window width. Then, the position corresponding to the highest similarity (similarity in the case of n = 5 in the example of FIG. 39) among the calculated similarities is specified as a position where data is missing.

  By performing the processing as described above, it is possible to identify the most probable position as the position where data is missing from the narrowed down range.

  Instead of the first refill position determination process, a second refill position determination process as described below may be performed.

  The second replenishment position determination process will be described with reference to FIGS. 40 and 41. FIG. First, the window generation unit 1232 of the replenishment position determination unit 123 sets 1 to the upper and lower widths of the window (FIG. 40: step S281). Then, the window generation unit 1232 generates window data based on the set upper width and lower width, and stores the window data in a storage device such as a main memory (step S283).

  Next, the window generation unit 1232 determines whether the uppermost sentence (that is, the sentence with the smallest line number) included in the window data includes data representing the chapter or section heading (step S285). If the top sentence does not include data representing a chapter or section heading (step S285: No route), the window generation unit 1232 increases the upper width of the window by 1 (step S287), and the process returns to step S283. .

  On the other hand, when the top sentence includes data representing the chapter or section heading (step S285: Yes route), the window generator 1232 displays the bottom sentence (that is, the sentence with the largest line number) included in the window data. ) Includes data representing a chapter or section heading (step S289). When the sentence representing the heading of the chapter or section is not included in the sentence at the bottom (step S289: No route), the window generating unit 1232 increases the lower width of the window by 1 (step S291), and the process returns to step S289. .

  On the other hand, when data representing a chapter or section heading is included in the sentence at the bottom (step S289: Yes route), the window generation unit 1232 generates window data with the set upper and lower widths, and the main memory Or the like (step S293). Further, the search unit 1233 searches for the position having the highest similarity with the generated window data in the details of the input process (step S295), and stores the data about the position with the highest similarity in the output data storage unit 125. Store. The process performed in step S295 is the same as the process performed in step S265. Then, the process returns to the original process.

  An example of window data generated by the second replenishment position determination process is shown in FIG. In the example of FIG. 41, the window ID, the line number, and the data of the sentences before and after the sentence to be supplemented (here, the upper width is 2 and the lower width is 4) are stored. In the example of FIG. 41, since the bottom sentence included in the window data is deleted, only the top sentence includes data representing the chapter or section heading.

  It should be noted that the processing details created in the UI process are characterized in that the contents are described in a relatively short sentence and the content changes drastically unlike a general document. Therefore, if the window width is too wide, data about various processing contents is included in the window data, so that the characteristics are diminished and it is difficult to narrow down candidate positions. On the other hand, if the window width is too narrow, there is a lack of data as a clue, so that an incorrect position is often specified. Therefore, instead of roughly determining the window width, using a single piece of data as window data, such as a chapter or a section, a position that is likely to be missing is high. It will be specified with accuracy.

  Returning to the description of FIG. 22, the output unit 127 outputs the data stored in the supplementary sentence storage unit 117 and the output data storage unit 125 to the display device (step S215). Then, the user confirms the sentence to be replenished and the position to be replenished, and takes measures such as correcting the data of the input processing details as necessary.

  By performing the processing as described above, it is possible to specify with high accuracy the position where the missing data in the processing details should be replenished, so that it is possible to reduce the cost required for correcting the processing details. become.

  Although one embodiment of the present technology has been described above, the present technology is not limited to this. For example, the functional block diagrams of the inconsistency detection apparatus and the correction support apparatus described above do not necessarily correspond to an actual program module configuration.

  Further, the configuration of each table described above is an example, and the configuration as described above is not necessarily required. Further, in the processing flow, the processing order can be changed if the processing result does not change. Further, it may be executed in parallel.

  For example, in the example described above, the item definition comparison process in step S61 is performed after the appearance partial comparison process in step S59. However, even if this order is reversed, the same processing result can be obtained. it can.

  In the present embodiment, the process details and item definition input of the design document to be diagnosed are received in step S1, and an independent word list and item name list are generated from the process details and item definitions input in step S3. Yes. However, the self-supporting word list and the item name list of the design document to be diagnosed may be read from the design document DB 7 and the processes after step S5 may be performed.

  In the corresponding position specifying process (step S233) described above, a candidate list is generated each time the process is performed. However, items that are not common item names and items that do not include inconsistent item names from item names included in the input item definition may be prepared in advance as a priority list and used instead of the candidate list.

  In the example described above, the similar design document is used. However, the process may be performed using, for example, a design document that receives an input from the user without using the similar design document.

  In the example described above, an example in which the present technology is applied to a design document created in the UI process of system development is shown. However, even if it is not such a design document, it is included in the document and the document. The present embodiment can be applied to any document that includes an item definition that defines an item name.

  The inconsistency detection device and the correction support device described above are computer devices, and as shown in FIG. 42, a display connected to a memory 2501, a CPU 2503, a hard disk drive (HDD) 2505, and a display device 2509. A control unit 2507, a drive device 2513 for a removable disk 2511, an input device 2515, and a communication control unit 2517 for connecting to a network are connected by a bus 2519. An operating system (OS) and an application program for executing the processing in this embodiment are stored in the HDD 2505, and are read from the HDD 2505 to the memory 2501 when executed by the CPU 2503. If necessary, the CPU 2503 controls the display control unit 2507, the communication control unit 2517, and the drive device 2513 to perform necessary operations. Further, data in the middle of processing is stored in the memory 2501 and stored in the HDD 2505 if necessary. In an embodiment of the present technology, an application program for performing the above-described processing is stored in a computer-readable removable disk 2511 and distributed, and installed from the drive device 2513 to the HDD 2505. In some cases, the HDD 2505 may be installed via a network such as the Internet and the communication control unit 2517. Such a computer apparatus realizes various functions as described above by organically cooperating hardware such as the CPU 2503 and the memory 2501 described above, the OS, and necessary application programs.

  The embodiment described above is summarized as follows.

  The inconsistency detection apparatus according to the present embodiment is extracted from an item definition including (A) an independent word group extracted from a sentence included in a document, an item name included in the sentence, and a definition of the item name. A document database that stores an item name group for each document; (B) a data storage unit that stores an independent word group and an item name group of the first document to be diagnosed; and (C) a document database that stores the item name group. The degree of similarity between the independent word group of each document and the independent word group stored in the data storage unit is calculated, the document having the similarity equal to or higher than a predetermined threshold is identified as the similar document, and the identified The similar document specifying means for extracting the independent word group and item name group of the similar document from the document database, and (D) the first independent word that is an independent word included in the independent word group stored in the data storage unit Use the similar document identification method to specify the item name. Item candidate extraction means for extracting from the extracted item name group, and (E) among the item names extracted by the item candidate extraction means, item names not included in the item name group stored in the data storage unit And inconsistent item specifying means for specifying as an inconsistent item name.

  In this way, since the item names included in the item name group of similar documents are used, it is possible to accurately define undefined item names while preventing independent words that are not originally item names from being defined as item names. It becomes possible to specify with. In addition, since it is not necessary to prepare a keyword pattern or a learning example for item name extraction for each document to be diagnosed, the cost is low. Furthermore, since it can be applied to any document including sentences and item definitions, it is highly versatile.

  Further, the item candidate extraction means described above calculates, for each first independent word, the similarity between the first independent word and each item name included in the item name group extracted by the similar document specifying means, If the calculated similarity is equal to or greater than the first threshold, the item name related to the calculation of the similarity and the matching degree set based on the similarity are associated with the first independent word and stored as matching item data For each first independent word stored in the matching item data storage unit and the matching item extraction means stored in the part, from the item name stored in association with the first independent word, at least based on the degree of matching, You may make it have a narrowing-down means to specify one item name as an item candidate. Although multiple item names may be extracted for one first independent word, the above-described processing can identify the most appropriate item name and eliminate unnecessary item names. .

  In addition, the matching item extraction unit described above sets a matching degree indicating that the calculated similarity is a partial match when the calculated similarity is greater than or equal to the first threshold and less than the second threshold, and the calculated similarity When the degree is greater than or equal to the second threshold, a degree of coincidence indicating complete match is set, and the above-described narrowing means, for each first independent word, item name whose degree of coincidence is perfect match, The first item name that is the only item name that is associated with the first independent word and that has a partial match, or the item name that is included in the item name group stored in the data storage unit An item name that matches any of the above may be specified as an item candidate. Of course, when the degree of coincidence is a perfect match, even if it is a partial match, the item name associated with the first independent word is one and there is no room for selection. . Further, even if the item is already defined in the item definition of the first document, it is naturally identified as an item candidate.

  Further, the narrowing-down means described above is associated with the first independent word, and the second item name that is an item name included in the item name group of the second document that is one similar document among the similar documents. When there are a plurality of words, the second independent word whose similarity to the second item name is equal to or greater than a predetermined threshold among the second independent words that are independent words included in the independent word group of the second document, 1 item name is identified from the second item name based on the distance from the similar part in the independent word group of the document 2 and the data for the item name other than the identified item name is deleted from the matching item data storage unit And when there are a plurality of item name groups including the item name associated with the first independent word, the item name groups stored in each of the item name groups and the data storage unit The item name group for which the calculated similarity is the maximum The item names included may further have a field definition comparing means for identifying as an item candidate. Thereby, even if it is a case where an item candidate cannot be specified only by narrowing down using a matching degree and the item name group of the 1st document, an item candidate can be specified appropriately.

  Further, the appearance part comparison means described above identifies an item name that matches any of the first item names among the item names included in the item name group of the second document, and stores it in the common item data storage unit. The similarity between the second independent word and each item name stored in the common item data storage unit is calculated for each second independent word, and the independent word of the second document is calculated based on the calculated similarity. A similar part in the group is specified, data for specifying the similar part is generated and stored in the similar part data storage unit, and for each second independent word, the second independent word and each second item name When the similarity is calculated and the maximum similarity among the calculated similarities is equal to or greater than a predetermined threshold, the second independent word and the similar part are determined using data for specifying the similar part. The distance is calculated, and the first distance that is the minimum distance among the calculated distances is specified. The second item name related to the calculation of similarity is associated with the identified first distance and stored in the distance data storage unit, and the first distance associated with the distance data storage unit is the smallest. The second item name may be specified, and data regarding the second item name other than the specified second item name may be deleted from the matching item data storage unit. Narrowing is performed based on the idea that the closer the distance to the similar part is, the higher the relevance to the first document is, and the suitable item candidate is.

  The correction support apparatus according to the present embodiment includes (A) a database that stores item definition data that defines item names related to processing, and detailed processing data that defines the details of the processing, and (B) a database. The first item definition data and the first process detail data corresponding to the first item definition data are read out, and are defined in the first item definition data and included in the first process detail data An item extraction unit that extracts an inconsistent item name as an inconsistent item name, and (C) inconsistent item name for second processing detailed data including the inconsistent item name among the processing detailed data stored in the database Identifies a position that appears in the second process detailed data, and stores a position data representing the position in the storage device, and (D) a first item definition from the database. The item name defined in the second item definition data corresponding to the data and the second process detailed data is extracted as a common item name, and the common item name is the first process detailed data and the second process. Using the corresponding position specifying unit that specifies each position appearing in the detailed data, stores the position information representing the position in the storage device in association with the common item name, and (E) the data stored in the storage device A range specifying unit for specifying the common item name that appears immediately before the inconsistent item name and the common item name that appears immediately after the common item name as the immediately preceding item name and the immediately following item name in the second processing detailed data. Have.

  With such a configuration, it is possible to narrow down the position where the data relating to the inconsistent item name is missing with high accuracy, so that the cost required for the correction work can be reduced.

  In addition, a window generation unit that performs processing for extracting data included in a specific range before and after the inconsistent item name as window data from the second processing detailed data while changing the specific range; In the processing detailed data of, in the range between the immediately preceding item name and the immediately following item name, the window data extracted by the window generation unit is searched for the position having the highest similarity with the window data, and the window -You may make it further have a search part which specifies a position with the highest similarity among the positions detected about each data as a position where the data about an inconsistent item name is missing. In this way, it is possible to identify the most probable position as the position where the data relating to the inconsistent item name is missing from the identified range.

  In addition, before and after the inconsistent item name in the second processing detail data, data indicating a processing content delimiter is specified, and data included in a range sandwiched by the data indicating the processing content delimiter is displayed as window data. In the range sandwiched between the immediately preceding item name and the immediately following item name in the first processing detail data and the window generating unit to be extracted as, the position having the highest similarity between the window data extracted by the window generating unit, You may make it further have a search part which specifies as a position where the data regarding an inconsistent item name are missing. In this way, the window data will contain the data of the processing contents related to the inconsistent item name, so the position that is likely to be the position where the data related to the inconsistent item name is missing is highly accurate. To be specified.

  In addition, the corresponding position specifying unit described above, according to the order in which the common item name appears in the first item definition data, the position information indicating the position where the common item name appears in the first process detail data, and the common item You may make it match | combine with the positional information showing the position where a name appears in 2nd process detailed data. In this way, the possibility of specifying an incorrect range can be reduced.

  In addition, the second process detail data described above may include an inconsistent item name and have a high degree of similarity with the first process detail data. This is because a more reliable result can be obtained by using processing detail data having a high degree of similarity.

  The inconsistency detection method according to the present embodiment includes an independent word group extracted from a sentence included in a document, an item name group extracted from an item definition including an item name included in the sentence and the definition of the item name, This is executed by a computer that can access a document database for storing each document and a data storage unit for storing the independent word group and item name group of the first document to be diagnosed. The inconsistency detection method calculates (A) the degree of similarity between the independent word group of each document stored in the document database and the independent word group stored in the data storage unit. A similar document specifying step of specifying a document that is equal to or greater than a predetermined threshold as a similar document, and extracting the independent word group and item name group of the specified similar document from the document database; and (B) stored in the data storage unit An item candidate extraction step for extracting an item name that matches the first independent word that is an independent word included in the independent word group from the item name group extracted in the similar document specifying step; and (C) the item candidate extraction step. Identifying an item name that is not included in the item name group stored in the data storage unit among the extracted item names as an inconsistent item name.

  The correction support method according to the present embodiment includes (A) a first item from a database that stores item definition data that defines item names related to processing and detailed processing data that defines the details of the processing for each processing. First definition data and first process detail data corresponding to the first item definition data are read, and item names that are defined in the first item definition data and are not included in the first process detail data. The step of extracting as the inconsistent item name, and (B) the second process detailed data including the inconsistent item name among the process detailed data stored in the database, the inconsistent item name becomes the second process detailed data. Identifying the position where it appears, and storing the position information representing the position in the storage device; (C) from the database to the first item definition data and the second processing detail data The item names defined together in the corresponding second item definition data are extracted as common item names, and the positions where the common item names appear in the first process detail data and the second process detail data are specified. , Storing the position information representing the position in the storage device in association with the common item name, and (D) using the data stored in the storage device, the second processing detailed data in the common item name Identifying a common item name that appears immediately before the inconsistent item name and a common item name that appears immediately after.

  A program for causing a computer to perform the processing according to the above method can be created. The program can be a computer-readable storage medium such as a flexible disk, a CD-ROM, a magneto-optical disk, a semiconductor memory, a hard disk, or the like. It is stored in a storage device. The intermediate processing result is temporarily stored in a storage device such as a main memory.

  The following supplementary notes are further disclosed with respect to the embodiments including the above examples.

(Appendix 1)
A document database that stores for each document an independent word group extracted from a sentence included in the document, and an item name group extracted from an item definition including the item name and the definition of the item name included in the sentence;
A data storage unit for storing the independent word group and the item name group of the first document to be diagnosed;
The degree of similarity between the independent word group of each document stored in the document database and the independent word group stored in the data storage unit is calculated, and a document whose similarity is equal to or greater than a predetermined threshold is calculated as a similar document. Similar document specifying means for extracting from the document database the independent word group and item name group of the specified similar document
Item candidate extraction means for extracting an item name that matches the first independent word that is an independent word included in the independent word group stored in the data storage unit from the item name group extracted by the similar document specifying means; ,
Among the item names extracted by the item candidate extraction unit, an inconsistent item specifying unit that specifies an item name that is not included in the item name group stored in the data storage unit as an inconsistent item name;
A mismatch detection device having

(Appendix 2)
The item candidate extracting means is
For each first independent word, a similarity between the first independent word and each item name included in the item name group extracted by the similar document specifying unit is calculated, and the calculated similarity is the first If it is greater than or equal to a threshold value, the matching item extraction stores the item name relating to the calculation of the similarity and the matching degree set based on the similarity in the matching item data storage unit in association with the first independent word Means,
For each of the first independent words stored in the matching item data storage unit, one item name is selected as an item candidate based on at least the matching degree from item names stored in association with the first independent word. A narrowing-down means specified as:
The inconsistency detection device according to supplementary note 1, comprising:

(Appendix 3)
The matching item extracting means includes
When the calculated similarity is greater than or equal to the first threshold and less than the second threshold, a degree of coincidence indicating partial match is set, and the calculated similarity is greater than or equal to the second threshold. If there is, set the degree of match to indicate an exact match,
The narrowing means is
For each first independent word, the item name whose degree of coincidence is a perfect match, the item name that is the only item name associated with the first independent word and the degree of coincidence is a partial match, or 3. The inconsistency detection apparatus according to appendix 2, wherein an item name that matches any of the first item names that are item names included in the item name group stored in the data storage unit is specified as the item candidate.

(Appendix 4)
The narrowing means is
When there are a plurality of second item names that are associated with the first independent word and are included in item name groups of a second document that is one similar document among the similar documents, A second independent word whose similarity to the second item name is greater than or equal to a predetermined threshold among the second independent words that are independent words included in the independent word group of the second document, and the independent word of the second document Appearing portion that identifies one item name from the second item name based on the distance from the similar portion in the word group, and deletes data for item names other than the identified item name from the matching item data storage unit A comparison means;
When there are a plurality of item name groups including item names associated with the first independent words, the similarity between each of the item name groups and the item name group stored in the data storage unit is determined. An item definition comparing means for calculating and identifying an item name included in the item name group with the calculated maximum similarity as an item candidate;
The mismatch detection apparatus according to appendix 3, further comprising:

(Appendix 5)
The appearance part comparing means is
Among the item names included in the item name group of the second document, an item name that matches any of the first item names is specified, and stored in the common item data storage unit,
For each second independent word, a similarity between the second independent word and each item name stored in the common item data storage unit is calculated, and based on the calculated similarity, the second document Identify similar parts in independent words, generate data for identifying the similar parts and store it in the similar part data storage unit,
For each second independent word, the similarity between the second independent word and each of the second item names is calculated, and when the maximum similarity among the calculated similarities is equal to or greater than a predetermined threshold , Using the data for specifying the similar part, calculating a distance between the second independent word and the similar part, and specifying a first distance that is a minimum distance among the calculated distances; The second item name related to the calculation of the similarity is associated with the identified first distance and stored in the distance data storage unit,
From the distance data storage unit, the second item name having the smallest associated first distance is specified, and data about the second item name other than the specified second item name is the matching item. The inconsistency detection apparatus according to appendix 4, wherein the inconsistency detection apparatus is deleted from the data storage unit.

(Appendix 6)
An independent word group extracted from a sentence included in a document and an item name extracted from an item definition including an item name included in the sentence and the definition of the item name are stored in a document database for each document. Calculating the degree of similarity between the independent word group of each document and the independent word group stored in the data storage unit storing the independent word group and the item name group of the first document to be diagnosed, A similar document specifying step of specifying a document having a similarity equal to or greater than a predetermined threshold as a similar document, and extracting the specified independent document group and item name group of the similar document from the document database
An item candidate extraction step for extracting item names that match the first independent word, which is an independent word included in the independent word group stored in the data storage unit, from the item name group extracted in the similar document specifying step; ,
Of the item names extracted in the item candidate extraction step, specifying an item name that is not included in the item name group stored in the data storage unit as an inconsistent item name;
Inconsistency detection program for causing a computer to execute.

(Appendix 7)
A document database for storing for each document an independent word group extracted from a sentence included in a document, an item name included in the sentence, and an item name group extracted from an item definition including the definition of the item name, and a diagnosis target An inconsistency detection method executed by a computer accessible to the data storage unit storing the independent word group and the item name group of the first document,
The degree of similarity between the independent word group of each document stored in the document database and the independent word group stored in the data storage unit is calculated, and a document whose similarity is equal to or greater than a predetermined threshold is calculated as a similar document. A similar document specifying step for extracting the independent word group and item name group of the specified similar document from the document database;
An item candidate extraction step for extracting item names that match the first independent word, which is an independent word included in the independent word group stored in the data storage unit, from the item name group extracted in the similar document specifying step; ,
Of the item names extracted in the item candidate extraction step, specifying an item name that is not included in the item name group stored in the data storage unit as an inconsistent item name;
Inconsistency detection method including:

(Appendix 8)
A database that stores, for each process, item definition data that defines an item name related to the process and process detail data that defines the content of the process;
First item definition data and first process detail data corresponding to the first item definition data are read from the database, and are defined in the first item definition data and the first process detail data. An item extraction unit that extracts item names that are not included as inconsistent item names,
For the second process detail data including the inconsistent item name among the process detail data stored in the database, the position where the inconsistent item name appears in the second process detail data is specified, and the position A replenishment data specifying unit that stores position information representing
An item name defined together in the second item definition data corresponding to the first item definition data and the second process detail data is extracted as a common item name from the database, and the common item name is A corresponding position identifying unit that identifies each position appearing in the first process detail data and the second process detail data, and stores position information representing the position in the storage device in association with the common item name;
Using the data stored in the storage device, among the common item names, the common item name that appears immediately before the inconsistent item name and the common item name that appears immediately after the inconsistent item name in the second processing detailed data A range identifying part that identifies the immediately preceding item name and the immediately following item name;
A correction support apparatus.

(Appendix 9)
A window generation unit for performing processing for extracting data included in a specific range before and after the inconsistent item name as window data from the second processing detailed data while changing the specific range;
In the range between the immediately preceding item name and the immediately following item name in the first processing detailed data, each of the window data extracted by the window generation unit has the highest similarity to the window data. A search unit that searches for a high position and identifies a position having the highest similarity among the positions detected for each of the window data as a position where data relating to the inconsistent item name is missing,
The correction support device according to appendix 8, further comprising:

(Appendix 10)
Before and after the inconsistent item name in the second processing detail data, data indicating a processing content delimiter is specified, and data included in a range sandwiched by the data indicating the processing content delimiter is displayed as window data A window generator to extract as
In the first processing detailed data, in the range sandwiched between the immediately preceding item name and the immediately following item name, the position having the highest similarity with the window data extracted by the window generating unit is determined as the inconsistency. A search unit that identifies the position where data relating to the item name is missing;
The correction support device according to appendix 8, further comprising:

(Appendix 11)
The corresponding position specifying unit is
In accordance with the order in which the common item names appear in the first item definition data, position information indicating the position where the common item names appear in the first process detail data, and the common item name is the second process. The correction support device according to any one of appendices 8 to 10, wherein the correction information is associated with position information representing a position that appears in the detailed data.

(Appendix 12)
The correction support device according to any one of appendices 8 to 11, wherein the second processing detailed data includes the inconsistent item name and has a high degree of similarity with the first processing detailed data.

(Appendix 13)
The first item definition data and the first item definition data corresponding to the first item definition data are stored in the database storing the item definition data defining the item name related to the process and the process detail data defining the contents of the process for each process. Reading out the processing details data of 1 and extracting the item names defined in the first item definition data and not included in the first processing details data as inconsistent item names;
For the second process detail data including the inconsistent item name among the process detail data stored in the database, the position where the inconsistent item name appears in the second process detail data is specified, and the position Storing position information representing
An item name defined together in the second item definition data corresponding to the first item definition data and the second process detail data is extracted as a common item name from the database, and the common item name is A corresponding position identifying step of identifying each position appearing in the first process detail data and the second process detail data, and storing the position information representing the position in the storage device in association with the common item name;
Using the data stored in the storage device, among the common item names, the common item name that appears immediately before the inconsistent item name and the common item name that appears immediately after the inconsistent item name in the second processing detailed data Identifying the immediately preceding item name and the immediately following item name;
And a correction support method executed by a computer.

(Appendix 14)
A window generating step of performing processing for extracting data included in a specific range before and after the inconsistent item name as window data from the second processing detailed data while changing the specific range;
In the range between the immediately preceding item name and the immediately following item name in the first processing detailed data, each of the window data extracted in the window generating step has the highest similarity to the window data. Searching for a high position and identifying a position having the highest similarity among the positions detected for each of the window data as a position where data relating to the inconsistent item name is missing;
The correction support method according to supplementary note 13, further comprising:

(Appendix 15)
Before and after the inconsistent item name in the second processing detail data, data indicating a processing content delimiter is specified, and data included in a range sandwiched by the data indicating the processing content delimiter is displayed as window data A window generation step to extract as
In the first processing detailed data, in the range sandwiched between the immediately preceding item name and the immediately following item name, the position having the highest similarity with the window data extracted in the window generating step is determined as the inconsistency. Identifying the location where the data relating to the item name is missing,
The correction support method according to supplementary note 13, further comprising:

(Appendix 16)
The corresponding position specifying step includes
In accordance with the order in which the common item names appear in the first item definition data, position information indicating the position where the common item names appear in the first process detail data, and the common item name is the second process. The correction support method according to any one of supplementary notes 13 to 15, further comprising a step of associating positional information representing a position appearing in the detailed data.

(Appendix 17)
The correction support method according to any one of appendices 13 to 16, wherein the second process detailed data includes the inconsistent item name and has a high degree of similarity with the first process detailed data.

(Appendix 18)
The first item definition data and the first item definition data corresponding to the first item definition data are stored in the database storing the item definition data defining the item name related to the process and the process detail data defining the contents of the process for each process. Reading out the processing details data of 1 and extracting the item names defined in the first item definition data and not included in the first processing details data as inconsistent item names;
For the second process detail data including the inconsistent item name among the process detail data stored in the database, the position where the inconsistent item name appears in the second process detail data is specified, and the position Storing position information representing
An item name defined together in the second item definition data corresponding to the first item definition data and the second process detail data is extracted as a common item name from the database, and the common item name is A corresponding position identifying step of identifying each position appearing in the first process detail data and the second process detail data, and storing the position information representing the position in the storage device in association with the common item name;
Using the data stored in the storage device, among the common item names, the common item name that appears immediately before the inconsistent item name and the common item name that appears immediately after the inconsistent item name in the second processing detailed data Identifying the immediately preceding item name and the immediately following item name;
Is a correction support program for causing a computer to execute.

(Appendix 19)
A window generating step of performing processing for extracting data included in a specific range before and after the inconsistent item name as window data from the second processing detailed data while changing the specific range;
In the range between the immediately preceding item name and the immediately following item name in the first processing detailed data, each of the window data extracted in the window generating step has the highest similarity to the window data. Searching for a high position and identifying a position having the highest similarity among the positions detected for each of the window data as a position where data relating to the inconsistent item name is missing;
The correction support program according to appendix 18, for causing a computer to execute the above.

(Appendix 20)
Before and after the inconsistent item name in the second processing detail data, data indicating a processing content delimiter is specified, and data included in a range sandwiched by the data indicating the processing content delimiter is displayed as window data A window generation step to extract as
In the first processing detailed data, in the range sandwiched between the immediately preceding item name and the immediately following item name, the position having the highest similarity with the window data extracted in the window generating step is determined as the inconsistency. Identifying the location where the data relating to the item name is missing,
The correction support program according to appendix 18, for causing a computer to execute the above.

(Appendix 21)
The corresponding position specifying step includes
In accordance with the order in which the common item names appear in the first item definition data, position information indicating the position where the common item names appear in the first process detail data, and the common item name is the second process. The correction support program according to any one of appendices 18 to 20, comprising a step of associating position information representing a position appearing in the detailed data.

(Appendix 22)
The correction support program according to any one of appendices 18 to 21, wherein the second processing detailed data includes the inconsistent item name and has a high degree of similarity with the first processing detailed data.

1 Input Data Processing Unit 3 Input Data Storage Unit 5 Item Candidate Extraction Unit 7 Design Document DB
DESCRIPTION OF SYMBOLS 9 Similar design document specific | specification part 11 Similar design document storage part 13 Item candidate storage part 15 1st inconsistency item specification part 17 1st inconsistency item storage part 19 Output part 501 Matching item extraction part 503 Matching item data storage part 505 Narrowing part 507 Common item data storage unit 509 Similar part data storage part 511 Distance data storage part 5051 Item definition comparison part 5053 Appearance part comparison part 101 Input processing part 103 Input data storage part 105 Design data DB 107 Similar design data specification part 109 Matching item specifying unit 111 Second inconsistent item storage unit 113 Similar design document storage unit 115 Supplementary sentence extraction unit 117 Supplementary sentence storage unit 119 Narrowing processing unit 1191 Line number list storage unit 1192 Corresponding position data storage unit 1193 Line number list generation unit 1194 Corresponding position specifying unit 1195 Range specifying unit 121 Result storage unit 123 supplemented position determination unit 1231 similarity storage unit 1232 the window generator 1233 searching unit 125 outputs the data storage unit 127 output unit

Claims (6)

A database that stores, for each process, item definition data that defines an item name related to the process and process detail data that defines the content of the process;
First item definition data and first process detail data corresponding to the first item definition data are read from the database, and are defined in the first item definition data and the first process detail data. An item extraction unit that extracts item names that are not included as inconsistent item names,
For the second process detail data including the inconsistent item name among the process detail data stored in the database, the position where the inconsistent item name appears in the second process detail data is specified, and the position A replenishment data specifying unit that stores position information representing
An item name defined together in the second item definition data corresponding to the first item definition data and the second process detail data is extracted as a common item name from the database, and the common item name is For each of the first common item names that are the common item names that appear in the first process detail data, respectively, identify the positions that appear in the first process detail data and the second process detail data. A second common item name that is a common item name that matches the first common item name is specified from the second processing detailed data, and the first position where the first common item name appears and the second A corresponding position specifying unit that stores the second position where the common item name appears in the storage device in association with the second position ,
Using the data stored in the storage device, of the second common item name, the second common item names appearing immediately before the mismatched item name in the second processing detailed data and after The appearing second common item name is specified as the immediately preceding item name and the immediately following item name, and the first position associated with the second position where the immediately preceding item name appears and the immediately following item name appear. A range specifying unit for specifying a first position associated with the position of
A correction support apparatus. A window generation unit for performing processing for extracting data included in a specific range before and after the inconsistent item name as window data from the second processing detailed data while changing the specific range;
In the first processing detailed data, a first position where the first position and the immediately item names corresponding to the second position in which the immediately preceding item name appears is associated with the second position which appears In the range between the window data, the window data extracted by the window generator is searched for a position having the highest similarity to the window data, and the position detected for each of the window data is searched. A search unit that identifies a position having the highest similarity as a position where data relating to the inconsistent item name is missing,
The correction support device according to claim 1, further comprising: Before and after the inconsistent item name in the second processing detail data, data indicating a processing content delimiter is specified, and data included in a range sandwiched by the data indicating the processing content delimiter is displayed as window data A window generator to extract as
In the first processing detailed data, a first position where the first position and the immediately item names corresponding to the second position in which the immediately preceding item name appears is associated with the second position which appears A search unit that identifies a position having the highest similarity to the window data extracted by the window generation unit as a position where data relating to the inconsistent item name is missing,
The correction support device according to claim 1, further comprising: The second processing detailed data, any one of claims 1 to 3, wherein the similarity between the mismatched item name hints and said first processing detailed data is the most detailed process data 1 Correction support device described in one. Computer
The first item definition data and the first item definition data corresponding to the first item definition data are stored in the database storing the item definition data defining the item name related to the process and the process detail data defining the contents of the process for each process. Reading out the processing details data of 1 and extracting the item names defined in the first item definition data and not included in the first processing details data as inconsistent item names;
For the second process detail data including the inconsistent item name among the process detail data stored in the database, the position where the inconsistent item name appears in the second process detail data is specified, and the position Storing position information representing
An item name defined together in the second item definition data corresponding to the first item definition data and the second process detail data is extracted as a common item name from the database, and the common item name is For each of the first common item names that are the common item names that appear in the first process detail data, respectively, identify the positions that appear in the first process detail data and the second process detail data. A second common item name that is a common item name that matches the first common item name is specified from the second processing detailed data, and the first position where the first common item name appears and the second Storing in the storage device the second position where the common item name appears in association with the second position ;
Using the data stored in the storage device, of the second common item name, the second common item names appearing immediately before the mismatched item name in the second processing detailed data and after The appearing second common item name is specified as the immediately preceding item name and the immediately following item name, and the first position associated with the second position where the immediately preceding item name appears and the immediately following item name appear. Identifying a first position associated with the position of
Modify support how to run. The first item definition data and the first item definition data corresponding to the first item definition data are stored in the database storing the item definition data defining the item name related to the process and the process detail data defining the contents of the process for each process. Reading out the processing details data of 1 and extracting the item names defined in the first item definition data and not included in the first processing details data as inconsistent item names;
For the second process detail data including the inconsistent item name among the process detail data stored in the database, the position where the inconsistent item name appears in the second process detail data is specified, and the position Storing position information representing
An item name defined together in the second item definition data corresponding to the first item definition data and the second process detail data is extracted as a common item name from the database, and the common item name is For each of the first common item names that are the common item names that appear in the first process detail data, respectively, identify the positions that appear in the first process detail data and the second process detail data. A second common item name that is a common item name that matches the first common item name is specified from the second processing detailed data, and the first position where the first common item name appears and the second Storing in the storage device the second position where the common item name appears in association with the second position ;
Using the data stored in the storage device, of the second common item name, the second common item names appearing immediately before the mismatched item name in the second processing detailed data and after The appearing second common item name is specified as the immediately preceding item name and the immediately following item name, and the first position associated with the second position where the immediately preceding item name appears and the immediately following item name appear. Identifying a first position associated with the position of
Is a correction support program for causing a computer to execute.

Images