JP2019101149A - Question automatic generation program and question automatic generator - Google Patents

Abstract

To provide a question automatic generator that can automatically generate questions from various sentences.SOLUTION: The present invention relates to a question automatic generator for automatically generating questions about the contents of read sentences. The generator includes: a text data acquisition unit for acquiring text data of sentences on which questions are to be generated; a logic structure extraction unit for extracting technical terms in the acquired text data and also extracting the logic structure between technical terms; and a question generation unit for generating questions about technical terms, using the logic structure.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technology for automatically generating a question regarding the content of a read sentence.

  In the past, it has often been up to each faculty member to carry out the task of creating a question in a test based on a textbook, reference book, technical book, etc. in an educational setting, and the task of creating a test question becomes a burden on the staff. There was a case. In addition, there is a problem that sharing questions among teachers and staff, accumulation of data across fiscal years, and handover can not be performed well if questions are prepared for each individual.

  From the above situation, a system that can automatically generate questions was desired. As what produces a problem automatically, patent document 1 is mentioned, for example. This patent document 1 discloses an automatic problem generation apparatus for automatically generating a problem sentence from content described in a predetermined format existing on the Internet.

JP 2012-018300 A

  The patent document 1 has a function of automatically generating a problem sentence from content described in a predetermined format. However, since the problem sentences can not be generated from the sentences out of the fixed format, there is a problem that the documents which can generate the problem sentences are limited.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a question automatic generation program and a question automatic generation device capable of automatically generating questions from various sentences.

  The question automatic generation program according to the present invention is a question automatic generation program for causing a computer to realize a process of automatically generating a question regarding the content of a read sentence, and the computer is provided with the text data of the sentence to be generated. A text data acquisition function for acquiring a logical term, a logical structure extraction function for extracting a technical term included in the acquired text data and extracting a logical structure among a plurality of technical terms, and a question regarding the technical term using the logical structure And generating a question generation function for generating

  In addition, the logical structure extraction function includes a function of classifying logical structure relationships among a plurality of technical terms by logical structure classification including definition, substitution, inclusion, incidental property, etc., and the question generation function It is characterized by including a function of generating a question based on an algorithm according to the logical structure classification.

  Further, the logical structure extraction function is characterized by including a function of automatically generating a reverse logical structure of the logical structure based on the extracted logical structure.

  The text data acquisition function also includes an image area classification function of classifying an image into areas according to predetermined classifications such as a table of contents, a text, a diagram, and an index for image data of sentences for which questions are to be generated; And an image text conversion function of acquiring text data converted into text data by optical character recognition and having tag information of a predetermined classification.

  Further, the extraction of the technical terms in the logical structure extraction function is characterized by being performed using a technical term database.

  Further, the technical term database is characterized in that text data on an index portion of a book including text data of a sentence for question generation is used at least in part.

  Further, the logical structure extraction function is characterized in that the logical structure between the technical terms included in the text data is extracted based on a predetermined keyword corresponding to the logical structure.

  Further, the logical structure extraction function is characterized in that the acquired text data is decomposed for each sentence, and the logical structure is extracted for each sentence.

  Further, the logical structure extraction function is characterized by including a function of extracting a logical structure from a text data portion related to the acquired chart.

  Further, the computer is characterized by realizing a logical structure table output function of outputting the relationship of the logical structure between the technical terms represented by the logical structure extracted by the logical structure extraction function in the form of a table.

  The automatic question generation apparatus according to the present invention is an automatic question generation apparatus for automatically generating a question related to the content of a read sentence, and a text data acquisition unit for acquiring text data of a sentence for question generation, and the acquired text A logical structure extraction unit that extracts a technical term included in data and extracts a logical structure among a plurality of technical terms, and a question generation unit that generates a question regarding the technical term using the logical structure I assume.

  According to the apparatus for automatically generating questions according to the present invention, the text data of the sentence for question generation is acquired, the technical terms included in the acquired text data are extracted, and the logical structure between the plural technical terms is extracted. Since the logical structure is used to generate questions about technical terms, questions can be generated automatically from text data such as textbooks, technical books, and reference books, which makes it possible to reduce the burden on faculty members. Also, if the generated questions are stored in a database, it can be used for sharing among faculty and staff, accumulation of data across fiscal years, and handover. Furthermore, because it is possible to output a logical structure table that expresses the relationship of the logical structure used to generate the question, the logical structure table is used as an auxiliary teaching material for understanding textbooks, etc., or as a material for relearning after testing. It is possible to

It is a block diagram showing composition of a question automatic generation device 10 concerning the present invention. FIG. 6 is a flowchart showing a flow of question automatic generation processing in the question automatic generation device 10. It is explanatory drawing showing an example of the textbook containing the content which wants to perform question generation. It is explanatory drawing showing the content which text-ized the page of the textbook shown in FIG. It is explanatory drawing showing the extraction result at the time of extracting a logical structure about each of the extraction sentence shown in FIG. It is an explanatory view showing an example of generation of a logic structure of a reverse direction. It is explanatory drawing showing an example of the question automatically produced | generated based on a logical structure. It is explanatory drawing showing an example of the logical structure table | surface which represented the relationship of logical structure. It is explanatory drawing showing an example of the chart contained in a textbook etc.

  Hereinafter, an example of the question automatic generation device according to the embodiment will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an automatic question generation apparatus 10 according to the present invention. Although the question automatic generation device 10 may be a device designed as a dedicated machine, it is assumed that it can be realized by a general computer. In this case, the automatic question generation apparatus 10 may include a central processing unit (CPU), a graphics processing unit (GPU), a memory, and a hard disk drive that a general computer would normally have. Storage (not shown). Further, it is needless to say that various processes are executed by a program in order to cause these general computers to function as the question automatic generation device 10 of this example. In addition, the question automatic generation apparatus 10 itself does not have to have all the components of the question automatic generation apparatus 10 described below, and a part of the configuration can be provided in another apparatus, for example, connection via a communication network is possible It is also possible for the automatic question generation apparatus 10 to use the configuration provided in the server apparatus while performing communication, as the specific server apparatus is provided with the partial configuration.

  As shown in FIG. 1, the automatic question generation apparatus 10 comprises at least a text data acquisition unit 11, a logical structure extraction unit 12, a question generation unit 13, a logical structure table output unit 14, and a storage unit 15. There is. Further, in the present example, the text data acquisition unit 11 includes an image area classification unit 111 and an image text conversion unit 112.

  The text data acquisition unit 11 has a function of acquiring text data of a sentence for which a question is to be generated. It is a configuration for acquiring the text data of the sentence including the contents for which you want to generate questions such as textbooks, reference books, technical books, etc., in addition to the function of importing the texts into the question automatic generation device 10 It also has the function of converting texts, specialized books, etc. into image data using an image scanner and processing the text content in the image data into text in the image area classification unit 111 and image text conversion unit 112 below. . The acquired text data may remain in a state where a plurality of sentences are continuously described, but in the present example, in order to execute processing one sentence at a time in the logical structure extraction unit 12 described later, In the text data acquisition unit 11, the acquired text data is divided into sentences and stored. At this time, cells may be divided and stored one sentence at a time in tabular data.

  The image region classification unit 111 has a function of classifying images into regions according to a predetermined classification such as a table of contents, a text, a diagram, and an index, for image data of sentences for which question generation is to be performed. In books such as textbooks, there are written parts with different contents such as table of contents, text, figures, indexes, etc., so these should be classified. As a method of classification, an algorithm (may be discrimination by machine learning) for automatically judging from the written content may be adopted, or the user may designate the classification for each area.

  The image text converting unit 112 converts the text data of the classified image data into a text data by optical character recognition (OCR) for each region after classification in the image region classification unit 111, and generates a tag of a predetermined classification. It has a function of acquiring text data attached with information. The area classified into the text is acquired as text data of the text, and the area classified into the diagram is acquired as text data with tag information of each classification, such as acquiring as the text data of the diagram. In addition, text data of the area classified into the index may be used as a technical term database to be described later.

  The logical structure extraction unit 12 has a function of extracting a technical term included in the acquired text data and extracting a logical structure among a plurality of technical terms. The reason for extracting the technical terms is that in order to extract the logical structure of the part related to the technical terms, it is necessary to determine whether the word included is a technical term. For extraction of technical terms, for example, a technical term database may be used. The technical term database may be a technical term database concerning sentences prepared and stored in advance. In addition, text data read for the area classified into the index may be used as a technical term database as it is a text that is likely to contain a technical term.

  The extraction process of the logical structure is a process of extracting as a logical structure representing the relationship between the words including the technical term, and specifically, the specialization included in the text data based on a predetermined keyword corresponding to the logical structure Extract the logical structure between terms. For example, for a sentence "A is classified into B and C", when A, B and C are technical terms, from this sentence, the logic "A → (including) → B, C" It is possible to extract the structure and the logical structure “B → (opposition) → C”. This includes in advance the logical structure of inclusion and the regularity that there is a high possibility of being able to extract the logical structure of the opposite when the keywords “are classified into and” are stored in advance. Then, when the keyword "classified" is found, processing for extracting the logical structure of inclusion and the logical structure of substitution is executed. In addition, keywords for extracting the logical structure of inclusion, such as "There is ~," "includes ~," "Does not exist," "is present The logical structure among the technical terms is extracted based on various keywords, such as a keyword for extracting a logical structure that defines the existence or non-existence relationship such as “~”. The extraction of the logical structure is not limited to the case where all the words for which the logical relationship is extracted are technical terms, and the logical structure may be extracted even when only one of the words is a technical term. In addition, as to the definition of words, not only the logical structure extraction between words but also the entire definition part may be extracted as a logical structure. For example, in the case where the keyword "is a ..." is used as the condition for extracting the logical structure of the definition, not the word but the entire explanation of the definition may be extracted. It should be noted that in one sentence where two or more technical terms appear, in most cases with some exceptions, it is likely that some logical structure can be extracted. Therefore, it is desirable to set a possible expression as a keyword expressing the relationship between two or more technical terms as a keyword for extracting the logical structure as much as possible.

  In addition, for each of the logical structures extracted in the logical structure extraction unit 12, a logical structure in the reverse direction is generated. The reverse logical structure refers to a logical structure representing the relationship when the logical structure can be expressed in the reverse direction. Specifically, for example, from the logical structure “A → (included) → B, C”, the logical structure in the reverse direction of “B → (included) → A” and “C → (included) → A” To extract. Extracting the logical structure in the reverse direction has the effect of increasing the choice of the logical structure used to generate the question.

  The question generation unit 13 has a function of generating a question related to a technical term using a logical structure. More specifically, the contents of the question to be generated are determined according to the type of logical structure. For example, from the logical structure “B → (include) → D, E”, the question “B is a concept including D. O or x?” Can be generated. Also, if there are three logical structures: "B → (include) → D, E", "C → (include) → F, G", and "B → (opposite) → C", these three logic structures If all the structures are used, it is possible to generate the question "Select the one included in B. Correct choice: D, E, Incorrect choice: F, G". In addition, it is conceivable to pick up a plurality of logical structures related to the definition, set an element of the content of the definition to be asked as a question as the correct choice, and generate a question having the contents of other definitions as the incorrect choice. These are merely examples, and it becomes possible to automatically generate various questions using relationships of various logical structures. As for the number of questions generated, the number of questions desired by the user may be designated, or all the questions that can be generated from the logical structure may be generated within a range that does not overlap.

  The logical structure table output unit 14 has a function of outputting the relationship of the logical structure between the technical terms expressed by the logical structure extracted by the logical structure extraction unit 12 in the form of a table. Not only are questions generated by the question generation unit 13 based on the logical structure, and questions are given in the test, but by outputting a table representing the relationship of the logical structure between the technical terms, for example, relearning after test It is possible to use a logical structure table as a source of data.

  The storage unit 15 has a function of storing various data necessary for processing of each unit and data obtained as a result of processing in each unit. For example, the text data acquired by the text data acquisition unit 11, the logical structure extracted by the logical structure extraction unit 12, the question generated by the question generation unit 13, the table generated by the logical structure table output unit 14, the technical term database, etc. You may make it memorize | store in this memory | storage part 16. FIG.

  Next, the flow of the question automatic generation process in the question automatic generation device 10 will be described. FIG. 2 is a flow chart showing the flow of the question automatic generation processing in the question automatic generation device 10. In FIG. 2, first, the automatic question generation apparatus 10 acquires text data (S101). Although acquisition of text data may be performed by any method, for example, text contained in image data may be converted into text data by optical character recognition processing (OCR processing). Then, the automatic question generation apparatus 10 classifies the text data acquired according to a predetermined classification such as a headline, a text, a diagram, an index or the like for the text data (S102). In the case of text conversion by OCR processing, classification may be performed at the stage of image data, and text conversion may be performed after classification. Then, with respect to the text data obtained and classified, a portion where a plurality of portions are connected to constitute a sentence, in particular, the text is decomposed into each sentence (S103). This is because the extraction process of the logical structure is executed in units of one sentence.

  Next, the automatic question generation apparatus 10 executes extraction processing of the logical structure one sentence at a time (S104). Extraction of the logical structure tries to extract the logical structure not only from the text but also from the diagram if possible. Also, the automatic question generation apparatus 10 generates a logical structure in the reverse direction for the extracted logical structure (S105). The automatic question generation apparatus 10 automatically generates a predetermined number of questions based on both forward and reverse logical structures (S106). Then, the automatic question generation apparatus 10 outputs the generated question (S107). The output may be displayed on a display screen for confirmation or may be printed out on paper or the like. Also, as a matter of course, processing to be stored in the storage unit 16 is also executed. Further, if necessary, a logical structure table representing the relationship of the logical structure is output together (S108), and the process is ended.

  Next, the operation of the automatic question generation apparatus 10 will be described using a more specific example. FIG. 3 is an explanatory view showing an example of a textbook including contents for which question generation is desired. As shown in FIG. 3, the text book includes a heading part, a text part, a diagram part and the like. Further, FIG. 3 assumes that a textbook is converted into image data by an image scanner. In that case, the process of classifying the contents included in the image of FIG. 3 for each area is performed. The classification process may be classified according to an algorithm that automatically determines from the described content, or may be a method in which the user specifies a classification for each area. After classification, text data is formed for each area by OCR processing.

  FIG. 4 is an explanatory view showing the contents of the page of the textbook shown in FIG. 3 converted into text data. As shown in FIG. 4, the heading portion and the extracted sentence extracted as the main text are distinguished and textified. In addition, when it is converted into text, a plurality of parts are linked and the sentences are divided into sentences and managed. The data in this case may be stored one sentence at a time in different cells of tabular data as shown in FIG.

  FIG. 5 is an explanatory view showing the extraction result when the logical structure is extracted for each of the extracted sentences shown in FIG. As shown in FIG. 5, from the first sentence, four logical structures including a logical structure of “probabilistic influence → (opposition) → determinate influence” are extracted. From the second sentence, the logical structure "Probabilistic effect → (including) → carcinogenesis, genetic effect" is extracted. From the third sentence, the logical structure "determinate effect → (including) → cataract, skin inflammation" is extracted.

  FIG. 6 is an explanatory view showing an example of generation of a logical structure in the reverse direction. When a logical structure of “probabilistic effect → (include) → (impact) → carcinogenesis, genetic effect” is extracted as a logical structure in the forward direction, “carcinogenic → (included)” as a logical structure in the reverse direction. → (Classification of Impact) → Stochastic Impact and “Genetic Impact → (Included) → (Classification of Impact) → Stochastic Impact” Two opposite logical structures are generated. Such a backward logic structure is also used for question generation.

  FIG. 7 is an explanatory view showing an example of questions automatically generated based on the logical structure. For example, the processing logic for automatically generating a question for selecting the correct choice from among a plurality of options includes: (1) selecting a logical structure of a constellation; (2) automatically collecting elements included in each of the logical structures of a constellation (3) Create a question in the following sequence: the element included in the substitution for which you want to answer the correct answer is the correct choice, and the element included in the remainder of the substitution is the incorrect choice. If a question is generated from the logical structure extracted in FIG. 5 according to this processing logic, for example, the question of Example 1 or Example 2 can be generated. In addition, Example 1 and Example 2 are the questions which only reversed the relationship between the correct choice and the incorrect choice, and the logical structure used to create this Example 1 and Example 2 is completely the same.

  FIG. 8 is an explanatory diagram showing an example of a logical structure table showing the relationships of logical structures. When the processing logic of the question automatic generation described above in the explanation of FIG. 7 is executed, the relationship of the discussion structure used for the question generation is extracted. The questions are generated using the relationship of the extracted logical structure, but if the relationship of the logical structure is output as a logical structure table as it is as shown in FIG. 8, it will help the learner to understand It will be possible to use as a supplementary teaching material for Since this logical structure table can be created in accordance with the question generation time, it may be distributed and used as a review material after the test.

  FIG. 9 is an explanatory view showing an example of a chart included in a textbook or the like. Fig. 9 shows that classification of radiation effects includes probabilistic effects and deterministic effects, carcinogenesis as specific examples of probabilistic effects, and genetic effects, and cataracts as specific examples of deterministic effects, It can be read that skin inflammation is included. If the diagram shown in FIG. 9 is shown, for example, in the diagram area of FIG. 3 and the text does not include the contents, the logical structure is extracted from the diagram of FIG. If a logical structure table as shown can be created, the logical structure can be extracted from the contents included only in the diagram and the question can be generated. In the example of FIG. 9, the logical structure of inclusion is extracted by determining that it is most natural to interpret that the logical structure of inclusion exists between the cell on the left and the cell on the right of the diagram. There is. Also, it is possible to extract a logical structure in which the items described in parallel on the vertical axis are in a relation of arrangement. Thus, although the possibility of extracting the logical structure by estimation increases as compared to the case of the sentence, the logical structure can also be extracted from the diagram.

DESCRIPTION OF SYMBOLS 10 Question automatic generation apparatus 11 Text data acquisition part 111 Image area classification part 112 Image text conversion part 12 Logical structure extraction part 13 Question generation part 14 Logical structure table output part 15 Storage part

Claims (11)

A question automatic generation program for causing a computer to realize a process of automatically generating a question regarding the content of a read sentence,
On the computer
A text data acquisition function for acquiring text data of a sentence for which a question is to be generated;
A logical structure extraction function of extracting a technical term included in acquired text data and extracting a logical structure among a plurality of technical terms;
A question automatic generation program for realizing a question generation function of generating a question regarding a technical term using the logical structure. The logical structure extraction function includes a function of classifying logical structure relationships among a plurality of technical terms by logical structure classification including definition, substitution, inclusion, incidental property, etc.,
The question automatic generation program according to claim 1, wherein the question generation function includes a function of generating a question based on an algorithm according to the logical structure classification. The question automatic generation program according to claim 1 or 2, wherein the logical structure extraction function includes a function of automatically generating a reverse logical structure of the logical structure based on the extracted logical structure. The text data acquisition function is an image area classification function of classifying images into areas according to predetermined classifications such as a table of contents, text, figures, indexes, etc. for image data of sentences for which questions are to be generated; The question automatic generation program according to any one of claims 1 to 3, further comprising: an image text conversion function of acquiring text data converted into text data by dynamic character recognition and attached with tag information of a predetermined classification. The automatic question generation program according to any one of claims 1 to 4, wherein the extraction of technical terms in the logical structure extraction function is performed using a technical term database. The automatic question generation program according to claim 5, wherein the technical term database uses, at least in part, text data on an index portion of a book including text data of a sentence for which a question is to be generated. The question according to any one of claims 1 to 6, wherein the logical structure extraction function extracts the logical structure between the technical terms included in the text data based on a predetermined keyword corresponding to the logical structure. Automatic generation program. The automatic question generation program according to any one of claims 1 to 7, wherein the logical structure extraction function decomposes the acquired text data for each sentence and performs extraction of the logical structure for each sentence. The question automatic generation program according to any one of claims 1 to 8, wherein the logical structure extraction function includes a function of extracting a logical structure from a text data portion related to the acquired chart. On the computer
The logical structure table output function is realized, which outputs the logical structure relationship between the technical terms expressed by the logical structure extracted by the logical structure extraction function in the form of a table. Question auto-generation program. It is a question automatic generation device that automatically generates a question regarding the content of the read sentence.
A text data acquisition unit that acquires text data of a sentence for which a question is to be generated;
A logical structure extraction unit that extracts technical terms included in the acquired text data and extracts a logical structure among a plurality of technical terms;
An automatic question generation apparatus comprising: a question generation unit that generates a question regarding a technical term using the logical structure;

Images