JP2023094022A - Corrected answer retrieval apparatus, corrected answer retrieval method, and program - Google Patents

Abstract

To provide a corrected answer retrieval apparatus which can support correctors in totally improving correction quality while suppressing fluctuation in correction even if variation occurs in answers in a case where there are a plurality of correctors.SOLUTION: The corrected answer retrieval includes: a tag-imparted corrected answer storage unit for storing a corrected answer imparted with a tag specifying a solution written by an answerer and an initial value of evaluation of the correction; a tag/evaluation-imparted corrected answer storage unit for storing the corrected answer imparted with the tag and the evaluation; a corrected answer retrieval unit for acquiring a predetermined tag, retrieving corrected answers imparted with the same tag as the predetermined tag or a tag similar thereto, and sorting the retrieved corrected answers in the descending order of the imparted evaluation for presentation; and an evaluation acquiring unit for acquiring evaluation of the presented corrected answers to impart the acquired evaluation to the corrected answers for storing them in the tag/evaluation-imparted corrected answer storage unit.SELECTED DRAWING: Figure 3

Description

The present invention relates to a corrected answer search device, a corrected answer search method, and a program for searching corrected answers.

For example, Japanese Patent Application Laid-Open No. 2002-200003 discloses a conventional correction system that supports correction based on correction history. The correction server 1 of Patent Document 1 includes answer receiving means 101 for receiving answer information 112 including at least the input content and time information input as an answer to the question of the question information 111; Answer activity analysis means 102 that analyzes activities related to and generates answer activity information 113, answer information 112, answer activity information 113, answer template information 114 that is an answer template for question information 111, and input as correction in the past and correction example creating means 103 for creating correction example information 115, which is an example of correction for the answer information 112 received by the answer receiving means 101, based on the correction history information 117. FIG.

JP 2014-048309 A

Even if the answer is the same, if the corrector is different, there will be variations in the correction. Also, if there are multiple solutions to a question, or if there is a pattern of answering halfway through, variations in answers will occur, and variations in correction will also occur. Such deviations and variations in correction cannot be covered by uniform correction standards, and it has been difficult to improve overall correction quality while suppressing deviations in correction.

The correction system of Patent Document 1 is expected to have a certain effect of preventing blurring in correction by presenting correction examples and supporting correction, but since the assumed problem is dictation of Chinese characters, there are variations in the solution method. do not have. In addition, in writing kanji, it is difficult to think of a case where students comment on halfway answers, so the above problem has not been solved.

Therefore, in the present invention, even if there are variations in answers when there are multiple correctors, it is possible to support the correctors so that the correction quality is improved as a whole while suppressing the deviation of the correction. An object of the present invention is to provide a completed answer retrieval device.

The corrected answer search device of the present invention includes a tagged corrected answer storage unit, a tagged/evaluated corrected answer storage unit, a corrected answer search unit, and an evaluation acquisition unit.

The tagged and corrected answer storage unit stores a corrected answer to which a tag specifying the solution written by the answerer and an initial evaluation value for correction are given. The tagged/evaluated corrected answer storage unit stores corrected answers to which tags and evaluations have been given. The corrected answer search unit acquires a predetermined tag, searches for corrected answers to which a tag identical or similar to the predetermined tag is assigned, and sorts the searched corrected answers in descending order of the given evaluation. Sort and present in order. The evaluation acquisition unit acquires an evaluation for the presented corrected answer, assigns the acquired evaluation to the corrected answer, and stores it in a tag/evaluation-added corrected answer storage unit.

According to the corrected answer retrieval device of the present invention, even if there are variations in answers when there are a plurality of correctors, the correction quality can be improved as a whole while suppressing the blurring of the correction. can support

FIG. 2 is a block diagram showing the functional configuration of the model learning device according to the first embodiment; FIG. 4 is a flow chart showing the operation of the model learning device according to the first embodiment; FIG. 2 is a block diagram showing the functional configuration of the corrected answer search device according to the first embodiment; FIG. 4 is a flowchart showing the tagging operation of the corrected answer search device of the first embodiment; 4 is a flowchart showing an answer presentation operation of the corrected answer search device of the first embodiment; FIG. 11 is a block diagram showing the functional configuration of the corrected answer search device of the second embodiment; 9 is a flowchart showing an answer presentation operation of the corrected answer search device of the second embodiment; The figure which shows the functional structural example of a computer.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail. Components having the same function are given the same number, and redundant description is omitted.

The functional configuration of the model learning device 1 of the first embodiment will be described below with reference to FIG. As shown in the figure, the model learning device 1 of this embodiment includes a model learning section 11 and a model storage section 11A.

The operation of the model learning device will be described below with reference to FIG. The model learning unit 11 learns a model by supervised learning using the corrected answer and the tag corresponding thereto, which is the tag specifying the solution method written by the answerer, as teacher data, and stores the learned model in the model storage unit 11A. (S11). Tags attached to the answers are predicted by reading the answers into the generated model.

As described above, the tag is the content specifying the solution written by the answerer, and must be added manually when prepared as teacher data. Examples of tags include "#property of ratio", "#condition of equal angle size", "#congruence of triangles", "#mathematical induction", and so on. Add descriptions such as "#... success", "#... was able to use", "#... failure", "#... halfway", etc. to indicate how well the solution is handled as a tag operation method, such as " #The condition that the magnitude of the angle is equal can be used.", "#In the middle of the mathematical induction method.", and so on.

Answers used as teacher data may be corrected answers or uncorrected answers, but it is preferable to use uncorrected answers in order to generate a more versatile model. For example, in step S31 of Example 2, which will be described later, uncorrected answers are read into the model and tags are added, so in Example 2, it is preferable to use a model generated by supervised learning using uncorrected answers. is. When using a corrected answer as an uncorrected answer, in the case of electronic answers, correction data is often stored in a separate layer, so remove or hide this layer to make it the same as the uncorrected answer. should be used.

Next, the functional configuration of the corrected answer retrieval device will be described with reference to FIG. As shown in the figure, the corrected answer search device 2 of this embodiment includes a tagging unit 21, a corrected answer storage unit 21A, a tagged and corrected answer storage unit 21B, and a corrected answer search unit 22. , a tagged/evaluated corrected answer storage unit 22A and an evaluation acquisition unit 23. FIG. The corrected answer retrieval device 2 is a device that uses the model learned by the model learning device 1 .

The tagging operation of the corrected answer retrieval device 2 will be described below with reference to FIG.

It is assumed that the corrected answer storage unit 21A stores corrected answers to which tags have not yet been assigned.

Based on the learned model, the tagging unit 21 predicts a tag to be attached to the corrected answer stored in the corrected answer storage unit 21A, and assigns the predicted tag to the corrected answer (S21). The tagging unit 21 stores the tagged corrected answer (hereinafter also referred to as tagged corrected answer) in the tagged corrected answer storage unit 21B. Step S21 is preferably executed each time a fixed amount of corrected answers is stocked. In this way, a tag predicted by a model is added to an answer that has been corrected by a predetermined educational institution, and is accumulated in the tagged and corrected answer storage unit 21B. Since the evaluation is used in step S22, which will be described later, an initial evaluation value is assigned to the tagged and corrected answer stored in the tagged and corrected answer storage unit 21B. For example, 0 is used as an initial value for evaluation. Further, the tagged and corrected answer stored in the tagged and corrected answer storage unit 21B may be given a question ID, which is the ID of the question corresponding to the answer. How to use the question ID will be described later.

It should be noted that the corrector may attach the tag instead of using the tagging unit 21 . In this case, the tagging unit 21 can be omitted.

The answer presentation operation of the corrected answer retrieval device 2 will be described below with reference to FIG. Here, the corrector refers to an uncorrected answer and gives the uncorrected answer an appropriate tag. In the example of Fig. 3, the corrector evaluates an uncorrected answer that "congruence of triangles can be used" and "the conditions for equal angles cannot be derived". In this example, the corrector would use the tag "#congruence of triangles", or the tag "#congruence of triangles is used", in addition to these tags, such as the tag "#Condition of equal magnitude of angles failed", for example. Appropriate tags are given to the corrected answers, and the corrected answers are input to the corrected answer search device 2 .

At this time, the corrector can also input the question ID specifying the question in addition to the tag to the corrected answer retrieval device 2 .

When the corrector inputs only the tag, the corrected answer search unit 22 acquires the tag input by the corrector and attaches the tag to the corrected answer with the same or similar tag as the tag input by the corrector. Both the corrected answer storage unit 21B and the tagged/evaluated corrected answer storage unit 22A are searched, and the searched corrected answers are sorted in descending order of the given evaluation and presented (S22). .

"High evaluation" means, for example, in the case of a favorite evaluation method such as "Like!", "Good!" means When the evaluation is based on the score evaluation method or the grade evaluation method, the higher the average value and the median value of these evaluations, the higher the evaluation. In any evaluation method, the initial value of evaluation corresponds to the lower rank.

When the corrector inputs a question ID together with a tag, the corrected answer search unit 22 acquires the tag and question ID input by the corrector, and is given a tag identical or similar to the tag input by the corrector, and Corrected answers given the same question ID as the input question ID are searched, and the searched corrected answers are sorted in descending order of given evaluation and presented.

The question ID can be used, for example, in situations where a large-scale mock test is graded by a plurality of correctors. For example, if you want to refer only to correction examples for a specific question of the mock exam, by specifying the question ID and searching, you can search for only corrected answers with similar or same tags attached to the question. This is highly convenient for the corrector.

The corrector performs the correction work with reference to the presented corrected answer, and finally inputs the evaluation of the corrected answer that was used as a reference into the corrected answer search device 2 . For example, the simplest evaluation may be a favorite evaluation system in which one input of "Like!", "Good!", or "+" is input. Other formats may be score evaluation based on a 100-point system, grade evaluation such as A/B/C/D/E.

The evaluation acquisition unit 23 acquires an evaluation for the presented corrected answer, assigns the acquired evaluation to the corrected answer, associates it with the question ID if necessary, and stores it in the tag/evaluation-assigned corrected answer storage unit 22A. Store (S23). A corrected answer to which an evaluation and a tag have been assigned is hereinafter also referred to as a tag/corrected answer to which evaluation has been assigned.

According to the corrected answer search device 2 of the present embodiment, by searching based on the tags, it is possible to accurately search for correction examples for answers with similar results in the solution. Among the examples, it is possible to sort so that correction examples with high evaluation among correctors are presented preferentially, so even if there are variations in answers when there are multiple correctors, It is possible to support the corrector so that the correction quality is improved as a whole while suppressing blurring.

Hereinafter, the functional configuration of the corrected answer retrieval device of the second embodiment will be described with reference to FIG. As shown in the figure, the corrected answer search device 3 of this embodiment includes a tagging unit 31, a corrected answer storage unit 21A, a tagged and corrected answer storage unit 21B, and a corrected answer search unit 22. , a tagged/evaluated corrected answer storage unit 22A, and an evaluation acquisition unit 23, and the functional configuration other than the tagging unit 31 is the same as that of the first embodiment.

In the first embodiment, the corrector attaches tags to uncorrected answers and inputs tags into the device. section 31 executes. Specifically, first, an uncorrected answer is input to the corrected answer search device 3 . Next, as shown in FIG. 7, the tagging unit 31 predicts the tag to be given to the uncorrected answer based on the model stored in the model storage unit 11A. (S31). At this time, the tagging section 31 outputs the question ID to the corrected answer searching section 22 if necessary. Since the operations of subsequent steps S22 and S23 are the same as those of the first embodiment, description thereof is omitted.

According to the corrected answer retrieval device 3 of the present embodiment, in addition to the effect of the first embodiment, even if the corrector does not think of appropriate tags for the uncorrected answer, the model predicts the tag instead. The convenience for users is further enhanced.

<Addendum>
The apparatus of the present invention includes, for example, a single hardware entity, which includes an input unit to which a keyboard can be connected, an output unit to which a liquid crystal display can be connected, and a communication device (for example, a communication cable) capable of communicating with the outside of the hardware entity. A communication unit that can be connected to, a CPU (Central Processing Unit, which may include cache memory, registers, etc.), RAM and ROM as memory, an external storage device as a hard disk, and these input units, output units, and communication units , the CPU, the RAM, the ROM, and the external storage device. Also, if necessary, the hardware entity may be provided with a device (drive) capable of reading and writing a recording medium such as a CD-ROM. A physical entity with such hardware resources includes a general purpose computer.

The external storage device of the hardware entity stores the programs necessary for realizing the functions described above and the data required for the processing of these programs (not limited to the external storage device; It may be stored in a ROM, which is a dedicated storage device). Data obtained by processing these programs are appropriately stored in a RAM, an external storage device, or the like.

In the hardware entity, each program stored in an external storage device (or ROM, etc.) and the data necessary for processing each program are read into the memory as needed, and interpreted, executed and processed by the CPU as appropriate. . As a result, the CPU realizes a predetermined function (each component expressed as above, . . . unit, . . . means, etc.).

The present invention is not limited to the above-described embodiments, and can be modified as appropriate without departing from the scope of the present invention. Further, the processes described in the above embodiments are not only executed in chronological order according to the described order, but may also be executed in parallel or individually according to the processing capacity of the device that executes the processes or as necessary. .

As described above, when the processing functions of the hardware entity (apparatus of the present invention) described in the above embodiments are implemented by a computer, the processing contents of the functions that the hardware entity should have are described by a program. By executing this program on a computer, the processing functions of the hardware entity are realized on the computer.

The various processes described above are performed by loading a program for executing each step of the above method into the recording unit 10020 of the computer 10000 shown in FIG. can.

A program describing the contents of this processing can be recorded in a computer-readable recording medium. Any computer-readable recording medium may be used, for example, a magnetic recording device, an optical disk, a magneto-optical recording medium, a semiconductor memory, or the like. Specifically, for example, magnetic recording devices include hard disk devices, flexible discs, and magnetic tapes, and optical discs include DVDs (Digital Versatile Discs), DVD-RAMs (Random Access Memory), CD-ROMs (Compact Disc Read Only). Memory), CD-R (Recordable) / RW (Rewritable), etc. as magneto-optical recording media, MO (Magneto-Optical disc) etc., as semiconductor memory EEP-ROM (Electrically Erasable and Programmable-Read Only Memory) etc. can be used.

Also, the distribution of this program is carried out by selling, assigning, lending, etc. portable recording media such as DVDs and CD-ROMs on which the program is recorded. Further, the program may be distributed by storing the program in the storage device of the server computer and transferring the program from the server computer to other computers via the network.

A computer that executes such a program, for example, first stores the program recorded on a portable recording medium or the program transferred from the server computer once in its own storage device. Then, when executing the process, this computer reads the program stored in its own recording medium and executes the process according to the read program. Also, as another execution form of this program, the computer may read the program directly from a portable recording medium and execute processing according to the program, and the program is transferred from the server computer to this computer. Each time, the processing according to the received program may be executed sequentially. In addition, the above processing is executed by a so-called ASP (Application Service Provider) type service, which does not transfer the program from the server computer to this computer, and realizes the processing function only by the execution instruction and result acquisition. may be It should be noted that the program in this embodiment includes information that is used for processing by a computer and that conforms to the program (data that is not a direct instruction to the computer but has the property of prescribing the processing of the computer, etc.).

Moreover, in this embodiment, the hardware entity is configured by executing a predetermined program on the computer, but at least part of these processing contents may be implemented by hardware.

Claims (7)

a tagged and corrected answer storage unit for storing a corrected answer to which a tag specifying the solution method written by the answerer and an initial evaluation value for correction are given;
a tagged/evaluated corrected answer storage unit for storing the corrected answer to which the tag and the evaluation have been given;
Obtaining a predetermined tag, retrieving corrected answers to which a tag identical or similar to the predetermined tag has been assigned, and presenting the retrieved corrected answers sorted in descending order of assigned evaluation. a corrected answer search unit for
A corrected answer retrieval device including an evaluation acquisition unit that acquires an evaluation for a presented corrected answer, adds the acquired evaluation to the corrected answer, and stores the tag/evaluation added corrected answer storage unit. The corrected answer search device according to claim 1,
The corrected answer search device, wherein the predetermined tag is a tag given to an uncorrected answer by a corrector and input to the corrected answer search device. The corrected answer search device according to claim 1 or 2,
a corrected answer storage unit that stores corrected answers to which tags have not yet been assigned;
A tag to be assigned to a corrected answer stored in the corrected answer storage unit and not yet assigned a tag is predicted based on a model pre-learned using supervised learning data using corrected answers and tags as teacher data. including a tagging part that attaches to
The tagged and corrected answer storage unit
A corrected answer retrieving device that stores corrected answers to which tags are assigned by the tagging unit and to which the initial evaluation values are assigned. The corrected answer search device according to claim 1,
further comprising a tagging unit that predicts and assigns tags to be assigned to uncorrected answers based on a model pre-learned with supervised learning data using the corrected answers and tags as teacher data;
The corrected answer search device, wherein the predetermined tag is a tag predicted and assigned by the model. The corrected answer search device according to any one of claims 1 to 4,
Each storage unit assigns a question ID, which is an ID of a question corresponding to each answer, to each answer and stores the answer;
The corrected answer search unit
Acquiring a predetermined tag and the question ID, searching for a corrected answer to which a tag identical or similar to the predetermined tag and having the same question ID as the question ID is given, and correcting the retrieved answer A corrected answer retrieval device for sorting and presenting corrected answers in descending order of assigned evaluation. A corrected answer search method executed by a corrected answer search device,
a tagged and corrected answer storing step of storing the corrected answer to which a tag specifying the solution written by the answerer and an initial evaluation value for the correction are given;
a tagged/evaluated corrected answer storing step of storing the corrected answer to which the tag and the evaluation have been given;
Obtaining a predetermined tag, retrieving corrected answers to which a tag identical or similar to the predetermined tag has been assigned, and presenting the retrieved corrected answers sorted in descending order of assigned evaluation. a corrected answer search step for
A corrected answer retrieval method, comprising: an evaluation acquisition step of acquiring an evaluation for a presented corrected answer, adding the acquired evaluation to the corrected answer, and storing the tag/evaluation added corrected answer storage unit. A program that causes a computer to function as the corrected answer retrieval device according to any one of claims 1 to 5.

Images