JPH08153108A - Method and device for retrieving information - Google Patents

Abstract

PURPOSE: To provide a method and a device for information retrieval which enables retrieval based upon a correct retrieval key word, specially, without depending upon the change of a retrieval key word unless the inputted retrieval key word is present, concretely, a method and device for information retrieval which can retrieve a department name after organization alteration even for a department name before organization alteration even when the department name to be retrieved is not present in a department data base owing to the alteration of the department name. CONSTITUTION: When the inputted retrieval key word is not a retrieval key word of the data base, the retrieval key word of the data base which is closest to the inputted retrieval key word is estimated as the correct retrieval key word. In concrete, this device has the department data base 101 which stores a list of department names of the organization, a department retrieval part 102 which retrieves whether or not the department name to be retrieved is present in the department data base, and a department estimation part 103 which estimates the correct department name from the said department name to be retrieved unless the department name to be retrieved is present in the data base.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information search method and apparatus for searching information based on an input search keyword.

[0002]

2. Description of the Related Art In a conventional information search apparatus, information search is performed based on a character string that matches an input search keyword.

[0003]

However, if there is no input of the search keyword or there is no search keyword input due to the change of the search keyword, the desired search result cannot be obtained and the search needs to be retried. was there. Such an example is not limited to the search keyword. For example, after the department name is changed by a company or the like due to the organization change, the department after the organization change corresponding to the mail etc. sent with the department name before the organization change It was not possible to find a name and send an email, or to search the database by the department name before the organizational change.

The present invention eliminates the above-mentioned drawbacks of the prior art, and provides an information retrieval method and an apparatus therefor, which makes it possible to perform retrieval by a correct retrieval keyword without changing the retrieval keyword when an inputted retrieval keyword does not exist. To do. In addition, specifically, even if the department name to be searched does not exist in the department database due to a change in the department name, the department name after the organization change can be searched for the department name before the organization change. A search method and a device therefor are provided.

[0005]

In order to solve the above-mentioned problems, the information search method of the present invention is an information search method for searching a database based on a search keyword.
When the input search keyword is not a search keyword of the database, the search keyword of the database closest to the input search keyword is estimated as a correct search keyword. Here, the database is a database which is at least partially changed to a search keyword associated with the search keyword before the change when the search keyword is changed, and does not store the old search keyword. It is a database.

Further, the information search apparatus of the present invention is a database in which when a search keyword is changed, the new search keyword is at least partially associated with the search keyword before the change, and the old search keyword is stored. A database that does not exist, a search means that searches the database with the input search keyword, and a correct search for the closest search keyword to the input search keyword when the input search keyword does not exist in the database And a keyword estimating means for estimating as a keyword. Here, the keyword estimation unit calculates the number of characters that match the changed search keyword with respect to the characters that form the input search keyword, and estimates the search keyword with the largest number of matching characters as the correct search keyword. And a means for performing morphological analysis on the input search keyword to decompose it into words, calculate the number of words that match the changed search keyword, and set the maximum value as the correct search keyword. Estimating means, and giving points corresponding to the number of consecutive matches with the changed search keyword for two or more consecutive characters of the input search keyword,
Means for estimating the search keyword in the database with the maximum number of points as the correct search keyword;
Means for providing a synonym dictionary such as a thesaurus, calculating the number of words matching the search keyword of the database with respect to the synonyms for the input keyword, and estimating the word having the largest number of matching words as the correct search keyword And at least one estimation means of Further, each of the estimating means includes an excluding means for excluding a generally common character from the input keyword from a comparison target.

Further, the information retrieval apparatus of the present invention comprises a department database storing a list of department names of an organization, department searching means for searching whether or not a department signature to be searched exists in the department database, When the department name to be searched does not exist in the department database, the department estimating unit estimates the correct department name from the department name to be searched. Here, the department estimation unit calculates the number of characters that match each department name of the department database with respect to the characters that form the department name of the search target, and the department signature in the department database that maximizes the number of matching characters. Is estimated as the correct department name. Further, the department estimation means performs morphological analysis on the department name of the search target to divide into words, and for each divided word, calculates the number of words that match each department signature of the department database, The department name in the department name database with the maximum number of matching words is estimated as the correct department name. In addition, the department estimation means,
For two or more consecutive characters that form the department name to be searched, consecutive characters that match each department name in the department database are obtained, and the number of points corresponding to the given word is set for the number of consecutive characters. It is calculated and the department name in the department database with the maximum number of points is estimated as the correct department name. Further, the department estimation means includes a synonym dictionary such as a thesaurus, calculates the number of words that match each department name of the department database with respect to the synonyms for the department name to be searched, and the number of matching words becomes maximum. The department name in the department database is estimated as the correct department name. Further, the department estimation unit excludes, from among the department names to be searched, characters that are almost common between the department names from the department names to be compared. In addition, when the electronic mail is delivered, a delivery means for delivering to the estimated department is further provided.

[0008]

With the above configuration, it is possible to perform a search by estimating a close search keyword with respect to the input search keyword that does not exist in the database.

[0009]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. <Embodiment 1> FIG. 1 is a diagram showing a configuration example of an information search apparatus according to the present embodiment. In the figure, 1 is a central processing unit (CPU) for processing the device, 2 is a keyboard for operation input, 21 is a pointing vice for pointing a screen such as a mouse, and 3 is a display output to an operator. For display, 4 is a bus for connecting the CPU 1 and other components,
Reference numeral 5 denotes a RAM or ROM memory, which includes a system program 104 for performing system processing, a department search unit 102 including a department search program and a department estimation unit 103 including a department estimation program, which are components of this embodiment. Including.
Note that these programs may be loaded in the memory 5 in another storage unit before execution of processing. A floppy disk 6 is used as a medium for inputting and outputting necessary data and programs. Reference numeral 7 denotes a large-capacity disk, which is a large-capacity storage unit such as a hard disk or a magneto-optical disk, and a department database 1 which is a constituent element of this embodiment.
Including 01. The department database 101 may be in the floppy disk 6 or the memory 5. Reference numeral 8 is a communication control device for accessing the information retrieval device via a communication line.

FIG. 2 shows the main components of the information retrieval apparatus of this embodiment. In FIG. 2, 101 is the department database that stores a list of department names,
Reference numeral 102 is the department search unit that searches whether or not the department name to be searched exists in the database, and 103 is the department estimation unit that estimates a correct department name for an incorrect department name that does not exist in the department database.

FIG. 3 is a flow chart showing the operation process of the information retrieval apparatus of this embodiment. In step S101,
The operator inputs the department name to be searched. Step S10
In 2, the department search unit 102 determines whether the department name input in step S101 exists in the department database 101.
It is searched by. If the department name exists in the database 101, the process proceeds to step S103. If it does not exist, the process proceeds to step S104. In step S103, the input department name is determined as the department name to be output.

If the department name does not exist in the database 101 in step S102, the department estimating unit 103 estimates a department name close to the incorrect department signature that is input but does not exist in the department database 101 in step S104. In step S105, the department name estimated in step S104 is set as the department name to be output. In step S106, step S103 or step S10.
The department name obtained in step 5 is output.

FIG. 4 is a flowchart for explaining in more detail the process of department estimation shown in step S104 of FIG. In step S201, for each department name in the department database 101, the number of characters that matches the input department name is calculated, and in step S202, the department name with the largest number of characters calculated in step S201 is estimated to be the correct department name.

An actual example will be described in more detail below. FIG. 5 shows an example of the department database 101 before and after the organization change. In this example, in the department database 101, the original artificial intelligence research department, computer research department,
The Computer Image Research Department and the Patent Department have been renamed as Intelligent Systems Research Department, Parallel Computer Research Department, Image Information Research Department, and Intellectual Property Protection Department, respectively. Suppose there is an email addressed to the "Artificial Intelligence Research Department" after the organization change.

The flow of processing in this case is as follows. That is, in step S101 of FIG. 3, the department name is input and set as “artificial intelligence research department”. In the next step S102, the "Artificial Intelligence Research Department" has a department database 1
Since it is found that the wrong department name does not exist in 01, the process proceeds to step S104. Step S104
First, the number of matching characters in step S201 shown in FIG. 4 is calculated. The calculated result is shown in FIG. Step S2
In 02, the "Intelligent Systems Research Department" having the largest number of matching characters in FIG. 6 is estimated to be the correct department name. In this way, the process of step S104 in FIG. 3 ends.

In the next step S105, the department name to be output is defined as "Intelligent Systems Research Department", and step 106
Then, "Intelligent Systems Research Department" is output as the department name.
Therefore, a mail addressed to the "artificial intelligence research department" can be sent to the "intelligent system research department". It should be noted that in step S101 of FIG. 3, an operator can make an input from a communication line, and an electronic mail can be automatically transmitted instead of outputting the department name.

In the above embodiment, the number of matching characters is calculated for all the characters forming the department name. Not limited to this, the number of matching characters may not be calculated for general common characters such as “copy” and “section”.
Further, in the above-described embodiment, the number of matching characters is calculated and the value is used to estimate the correct department name. Not limited to this, morphological analysis may be performed to divide into words, the number of matching words may be calculated, and the one with the maximum value may be estimated as the correct department name.

It is also possible to give a point corresponding to a word to the number of consecutive character strings that consecutively match two or more characters, and assume that the point with the largest point is the correct department name. In addition to the matching words, the number of words that are synonymous with information such as thesaurus is also added. For example, "patent"
On the other hand, a synonym “intellectual property right” may be added to the number.

Further, the above methods may be combined appropriately. <Embodiment 2> In the above embodiment, the processing of the information retrieval apparatus when the department name of the company organization or the like is changed, or the mail or the like is sent with the department name before the change is described. , It became possible to search for new department name in the sent mail.

Therefore, in the information retrieval apparatus for retrieving the database, there is a possibility that the name may be partially changed like the department database of the present embodiment, and not only the department name but also the database in which the old name is not left. It goes without saying that the information retrieval device described in this embodiment can be used as it is. That is, there is a database in which the keyword part of the search may be partially or entirely changed to the keyword associated with the previous keyword, and the information search device does not store the old keyword part. In the above, the database, a keyword search unit for bundling the database with keywords, and if the search target keyword does not exist in the database, by providing a keyword estimation unit that estimates the closest keyword to the search target keyword, Search is possible.

The keyword estimation unit calculates the number of matching characters for all the characters forming the keyword and estimates the keyword having the largest number of matching characters, as in the above embodiment. Morphological analysis is performed on the keywords, and the number of matching words is calculated by dividing them into words and the one with the largest value is estimated as the keyword, or for consecutive characters of two or more characters of the keyword,
A point corresponding to the number of consecutive matches with the keyword is given in advance, and the keyword in the database with the maximum number of points is estimated as the matched keyword. On the other hand, the number of matching words of each keyword in the database is calculated, and the word having the largest number of matching words is estimated as the keyword. In addition, characters that are generally common between the keywords are excluded from the keywords to be compared.

As described above, the information retrieval method of the present invention is applied to a database in which there is a possibility that some or all of the keywords associated with the previous keyword may be changed and the old keyword does not remain. Can be universally used. The present invention may be applied to a system including a plurality of devices or an apparatus including a single device. Further, it goes without saying that the present invention can be applied to the case where it is achieved by supplying a program to a system or an apparatus.

[0023]

As described above, according to the present invention, when the input search keyword does not exist, the correct search keyword is estimated from the input search keyword, so that it is correct regardless of the change of the search keyword. The search by the search keyword was enabled.

Further, according to the present invention, when the department name to be searched does not exist in the department database, the department estimating unit for estimating a correct department name for an incorrect department name is provided.
There is an effect that the department name after the organization change can be searched for the department name before the organization change.

[Brief description of drawings]

FIG. 1 is a configuration example of an information search device for explaining a first embodiment.

FIG. 2 is a diagram showing a logical structure of the present invention.

FIG. 3 is a flowchart showing an operation process of the embodiment.

FIG. 4 is a flowchart for explaining in more detail the step 104 of department estimation in the first embodiment.

FIG. 5 is a diagram illustrating an example of a department database 101 according to the first embodiment.

FIG. 6 is a diagram illustrating a department estimation step 104 according to the first embodiment.
It is a figure which shows the example of calculation of the number of matching characters.

[Explanation of symbols]

 1 CPU 2 keyboard 3 display 4 bus 5 memory 6 floppy disk 7 large capacity disk 8 communication control device 21 pointing device 101 department database 102 department search unit 103 department estimation unit 104 system program

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Shogo Shibata 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc. (72) Inventor Takaya Ueda 3-30-2 Shimomaruko, Ota-ku, Tokyo Kya Non-Incorporated (72) Inventor Yuji Ikeda 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc. (72) Inventor Minoru Fujita 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon Inc.

Claims (12)

[Claims] 1. An information search method for searching a database based on a search keyword, wherein when the input search keyword is not the search keyword of the database, the search keyword of the database closest to the input search keyword is selected. , An information search method characterized by estimating as a correct search keyword. 2. The database is a database that is changed to a search keyword associated with the search keyword before the change when the search keyword is changed, and stores the old search keyword. The information retrieval method according to claim 1, wherein the information retrieval method is a non-existent database. 3. A database in which when a search keyword is changed, a new search keyword is at least partially associated with the search keyword before the change, and a database in which an old search keyword is not stored is input. Search means for searching the database with the search keyword, and keyword estimating means for estimating a search keyword closest to the input search keyword as a correct search keyword when the input search keyword does not exist in the database An information search device comprising: 4. The keyword estimating means calculates the number of characters matching the changed search keyword with respect to the characters forming the input search keyword, and the search keyword having the largest number of matching characters is the correct search keyword. And a means for estimating the input search keyword, performing morphological analysis on the input search keyword to decompose it into words, calculating the number of words that match the changed search keyword, and finding the one with the maximum value as the correct search. Means for estimating as a keyword, and a search keyword in a database in which the number of consecutive points corresponding to the changed search keyword is given to two or more consecutive characters of the input search keyword, and the number of points is maximum And a dictionary of synonyms such as a thesaurus, At least one estimating means for calculating the number of words that match the search keyword in the database with respect to the synonyms for the entered keywords and estimating the word having the largest number of matching words as the correct search keyword The information search device according to claim 3, wherein 5. The information retrieving apparatus according to claim 4, wherein each of the estimating means includes an excluding means for excluding a generally common character from the input keyword from a comparison target. 6. A department database storing a list of department names of an organization, department search means for searching whether or not a department signature to be searched exists in the department database, and the department signature to be searched is a department database. And a department estimation unit that estimates a correct department name from the department name to be searched when the information retrieval device does not exist. 7. The department estimation unit calculates the number of characters that match each department name of the department database with respect to the characters that form the department name of the search target, and selects the number of characters in the department database in which the number of matching characters is the maximum. The information retrieval apparatus according to claim 6, wherein the department name is estimated as a correct department name. 8. The department estimation unit performs morphological analysis on the department name to be searched and divides it into words, and for each divided word, determines the number of words that match each department signature of the department database. 7. The information retrieval apparatus according to claim 6, wherein the calculated section name in the section name database having the maximum number of matching words is estimated as a correct section name. 9. The department estimation unit obtains a continuous character that matches each department signature of a department database for two or more consecutive characters that form the department signature to be searched,
7. The information retrieval according to claim 6, wherein the number of points corresponding to a given word is calculated with respect to the number of consecutive characters, and the department name in the department database having the maximum number of points is estimated as a correct department name. apparatus. 10. The department estimation unit includes a dictionary of synonyms such as a thesaurus, calculates the number of words that match each synonym of the department name of the search target with each department name of the department database, and determines the number of matching words. 7. The information retrieval device according to claim 6, wherein the department name in the largest department database is estimated as a correct department name. 11. The department estimation unit excludes characters that are substantially common to each department signature from the department signatures to be searched, from the department signatures to be compared.
The information search device described in any one of 0. 12. The information retrieving apparatus according to claim 6, further comprising a distribution unit that distributes to the estimated department when distributing the electronic mail.