Information processing device and program

Abstract

To recommend corresponding content of a problem from an assessment item used for rubric assessment.SOLUTION: An information processing apparatus includes a storage unit which stores a problem to be used for rubric assessment and an assessment item indicating a viewpoint of the assessment in association with each other. The rubric assessment is a method for assessing learning achievement of a learner, and can assess "though" or "expressiveness" from a performance, or the like. The information processing apparatus acquires, as input information, information on at least one of an assessment item and a keyword. The information processing apparatus acquires, from the storage unit, a target problem based on the assessment item and the keyword included in the problem from the input information, and outputs the target problem, as a recommendation problem.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a technique for outputting information related to learning. [Background technology]

[0002] In recent years, schools and other educational institutions have placed importance on non-cognitive abilities and active learning. Accordingly, attention has been focused on rubrics, which are indicators for evaluating performance in active learning activities such as "presentations" and "discussions." Patent Document 1 discloses a system that creates test questions based on rubrics, including items for assessing "knowledge and skills," "ability to think, judge, and express," and the like. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Utility Model Registration No. 3227380 Summary of the Invention [Problem to be solved by the invention]

[0004] Typically, the questions used in rubric-based assessments are created from scratch by teachers and other instructors. This places a heavy burden on instructors. In particular, instructors with little experience struggle with creating the questions, taking up a lot of time.

[0005] The present invention has been made to solve the above-mentioned problems, for example, and aims to recommend question contents corresponding to evaluation items used in rubric-based evaluation. [Means for solving the problem]

[0006] In one aspect of the present invention, an information processing device includes a storage unit that stores questions used in rubric evaluation in association with evaluation items that indicate evaluation perspectives, an input information acquisition unit that acquires information related to one or more of the evaluation items and keywords as input information, a recommended question acquisition unit that acquires, from the input information, target questions as recommended questions based on the evaluation items and keywords included in the questions from the storage unit, and a recommended question output unit that outputs the recommended questions. According to this aspect, the information processing device can acquire and output recommended questions for rubric evaluation that correspond to one or more of the evaluation items and keywords based on the input information. This can improve the efficiency of creating questions for rubric evaluation.

[0007] In one aspect of the information processing device, the recommended question acquisition unit acquires, from the storage unit, questions corresponding to evaluation items that completely match the evaluation items included in the input information as recommended questions. According to this aspect, the information processing device can acquire and output recommended questions with completely matching evaluation items.

[0008] In one aspect of the information processing device, the recommended question acquisition unit acquires, from the storage unit, questions corresponding to evaluation items that partially match evaluation items included in the input information as recommended questions. According to this aspect, the information processing device can acquire and output recommended questions that partially match evaluation items.

[0009] In one aspect of the information processing device, the recommended question acquisition unit acquires, from the storage unit, questions that contain a threshold number of words that match keywords included in the input information as recommended questions. According to this aspect, the information processing device can acquire and output recommended questions that correspond to keywords based on the number of matches between the keywords and the words included in the questions.

[0010] In one aspect of the information processing device, the information processing device includes a similarity calculation unit that calculates the similarity between the keyword and the recommended question, and the recommended question acquisition unit acquires, from the storage unit, questions whose similarity is equal to or greater than a threshold value as recommended questions. According to this aspect, the information processing device can acquire and output recommended questions corresponding to the keywords based on the similarity between the keywords and the questions.

[0011] In one aspect of the information processing device, the information processing device includes a constituent word acquisition unit that divides text indicating the content of the question and acquires constituent words that are words that make up the question, and the similarity calculation unit calculates the similarity based on the constituent words and a knowledge graph that represents, in a graph structure, connections of knowledge represented by a plurality of words centered on the evaluation item. According to this aspect, the information processing device can calculate the similarity between keywords and questions using the knowledge graph.

[0012] In one aspect of the information processing device, the information processing device includes a condition information acquisition unit that acquires condition information related to the conditions of the recommended problem, and the recommended problem acquisition unit acquires the recommended problem from the storage unit based on the input information and the condition information. According to this aspect, the information processing device can acquire and output recommended problems that correspond to one or more of the evaluation items and keywords and satisfy the conditions of the recommended problem.

[0013] In one aspect of the information processing device, the condition information includes information regarding whether the evaluation items included in the input information completely or partially match the evaluation items corresponding to the recommended problem. According to this aspect, the information processing device can recognize, based on the condition information, whether the conditions of the recommended problem completely match the evaluation items or partially match the evaluation items.

[0014] In one aspect of the information processing device, the condition information includes information on the number of matches between keywords and words that serves as a threshold, or information on the similarity that serves as a threshold. According to this aspect, the information processing device can recognize the threshold number of matches between keywords and words included in a question, or the threshold similarity between keywords and a question, which are conditions for the recommended question, based on the condition information.

[0015] In one aspect of the information processing device, the recommendation question output unit displays text indicating the content of the recommendation question, and highlights keywords included in the text. According to this aspect, the information processing device can display the text indicating the content of the recommendation question with the keywords highlighted.

[0016] In another aspect of the present invention, there is provided a program executed by an information processing device having a computer, causing the computer to function as a storage unit that stores questions used in rubric evaluation in association with evaluation items that indicate evaluation perspectives, an input information acquisition unit that acquires information related to one or more of the evaluation items and keywords as input information, a recommended question acquisition unit that acquires, from the input information in the storage unit, target questions as recommended questions based on the evaluation items and keywords included in the questions, and a recommended question output unit that outputs the recommended questions. By installing this program on a computer and causing it to execute, the information processing device of the present invention can be configured. [Effects of the Invention]

[0017] According to the information processing device of the present invention, it is possible to recommend question contents corresponding to the evaluation items used in the rubric-based evaluation. [Brief explanation of the drawings]

[0018] [Figure 1] 1 shows the configuration of a learning support system to which a server of the present invention is applied. [Figure 2] FIG. 1 is a diagram illustrating rubric evaluation. [Figure 3] FIG. 2 is a block diagram showing the hardware configuration of a server. [Figure 4] 10 is an example of a data configuration of a question information DB. [Figure 5] FIG. 2 is a block diagram showing the functional configuration of a server. [Figure 6] 10 is an example of a screen. [Figure 7]FIG. 10 is a diagram illustrating a method for specifying an evaluation item based on the degree of coincidence of the evaluation item. [Figure 8] FIG. 10 is a diagram illustrating a method for identifying questions based on the number of matches between questions and keywords. [Figure 9] FIG. 10 is a diagram illustrating a method for dividing the question content into words. [Figure 10] FIG. 1 is a diagram illustrating a method for constructing a knowledge graph. [Figure 11] FIG. 10 is a diagram illustrating calculation of similarity. [Figure 12] FIG. 10 is a diagram illustrating a method for specifying a question based on the similarity between a question and a keyword. [Figure 13] 10 is a flowchart of an output process. [Figure 14] 10 shows an example of the configuration of a learning support system according to a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0019] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. <Embodiment> [Overall configuration] Figure 1 shows the configuration of a learning support system to which the server of the present invention is applied. The learning support system 100 is a system that recommends question contents corresponding to evaluation items used in rubric-based evaluation (hereinafter also referred to as "rubric evaluation"). The learning support system 100 is configured so that a server 10 and a plurality of teacher terminals 20 can communicate with each other via a network 5 such as the Internet.

[0020] Here, we will explain rubric assessment. Rubric assessment is a method of evaluating a learner's learning achievement using a rubric table, and can evaluate "thinking ability" and "expression ability" based on performance in presentations, discussions, etc., rather than test scores.

[0021] Figure 2(a) shows an example of a problem used in rubric evaluation and a learner's response to that problem. The problem shown in Figure 2(a) is a problem for assessing "expressive ability," and the learner's response is oral, such as through a presentation. Figure 2(b) is an example of a rubric table. As shown in Figure 2(b), in the rubric table, the vertical axis represents evaluation items that indicate the evaluation perspectives, such as "thinking ability" and "expressive ability." The horizontal axis represents evaluation levels, which are numerically indicated by the degree of achievement for each evaluation item. The higher the evaluation level, the higher the number, and the lower the degree of achievement, the lower the number. The squares where the evaluation items on the vertical axis intersect with the evaluation levels on the horizontal axis represent the evaluation criteria for each evaluation item corresponding to each evaluation level. The evaluation criteria can be set arbitrarily.

[0022] Generally, the questions and rubrics used in rubric assessment are created by a learning supporter such as a teacher or instructor. The learning supporter then refers to the rubric and evaluates the learner's learning achievement based on the answers. Specifically, the learning supporter evaluates the learner's answers by comparing them with the evaluation criteria for "expressive ability" in the rubric table shown in Figure 2(b) and determining the evaluation level.

[0023] In this embodiment, the evaluation stages are numbers, but the present invention is not limited to this, and the evaluation stages can be set arbitrarily, such as alphabets, as long as they can indicate the degree of achievement. In this embodiment, as an example, the learning supporter who performs the evaluation using the rubric evaluation is the teacher, and the learner who is the subject of evaluation is the student.

[0024] The server 10 is an information processing device that processes, stores, and transmits / receives various types of information, and is, for example, a server device, a personal computer, or a general-purpose tablet PC (personal computer).

[0025] The teacher terminal 20 is used by the teacher and is, for example, an information processing device such as a tablet or personal computer terminal. Specifically, the teacher terminal 20 transmits input information related to evaluation items and keywords used in rubric evaluation, receives and displays recommended questions, etc. Here, recommended questions are rubric evaluation questions recommended to a teacher identified based on the input information, as will be described in detail later.

[0026] [Server Configuration] 3 is a block diagram showing the hardware configuration of server 10. Server 10 includes a communication unit 11, a control unit 12, a storage unit 13, a recording medium 14, a display unit 15, and an input unit 16. These components and a question information database (hereinafter, "database" will be referred to as "DB") 41 are interconnected via a bus 19.

[0027] The server 10 may be executed by a single computer, or may be executed by a plurality of computers in a distributed manner, or may be executed by a virtual machine in a distributed manner.

[0028] The communication unit 11 is a communication unit for communicating with the teacher terminal 20 via the network 5. Specifically, the communication unit 11 receives input information from the teacher terminal 20 and transmits information about recommended questions to the teacher terminal 20.

[0029] The control unit 12 includes a central processing unit (CPU), a micro-processing unit (MPU), a graphics processing unit (GPU), and other arithmetic processing devices, and performs various information processing, control processing, and the like related to the server 10 by reading and executing programs stored in the storage unit 13. The programs can be deployed so that they are executed on a single computer, located at one site, or distributed across multiple sites and on multiple computers interconnected by a communications network. Although the control unit 12 is described in FIG. 3 as being a single processor, it may also be a multi-processor.

[0030] The storage unit 13 includes memory elements such as RAM (Random Access Memory) and ROM (Read Only Memory), and stores programs, data, etc. required for the control unit 12 to execute processing. The storage unit 13 also temporarily stores data, etc. required for the control unit 12 to execute arithmetic processing.

[0031] The recording medium 14 is a non-volatile, non-transitory recording medium such as a disk-shaped recording medium or a semiconductor memory, and is configured to be detachable from the server 10. The recording medium 14 records various programs to be executed by the control unit 12. When the server 10 executes the output process described below, the programs recorded on the recording medium 14 are loaded into the memory unit 13 and executed by the control unit 12.

[0032] The display unit 15 is a liquid crystal display, an organic EL (electroluminescence) display, or the like, and displays various information in accordance with instructions from the control unit 12. The input unit 16 is an input device such as a mouse, keyboard, touch panel, or button, and outputs received operation information to the control unit 12.

[0033] The question information DB41 shows the correspondence between rubric evaluation questions and evaluation items. Rubric evaluation questions are pre-assigned evaluation items corresponding to the content of the question. FIG. 4 shows an example of the data structure of the question information DB41. As shown in the figure, in the question information DB41, the vertical axis represents rubric evaluation questions, and the horizontal axis represents evaluation items. In the question information DB41, a check mark is placed in the square where the question (vertical axis) intersects with the evaluation item (horizontal axis) pre-assigned to the question. Thus, for example, by referring to the question information DB41, it can be seen that the question "Describe the appeal of your favorite food within 30 seconds" corresponds to the evaluation items "thinking ability" and "expression ability."

[0034] The question information DB 41 may be constructed using the contents of questions previously created by teachers and the evaluation items corresponding to those questions. In this case, every time a teacher creates a new question, that question and the evaluation items corresponding to that question are stored in the question information DB 41.

[0035] The storage format of each DB described above is an example, and other storage formats may be used as long as the relationships between the data are maintained. Each DB is realized by a recording medium such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). In this embodiment, the storage unit 13 and the DB may be configured as an integrated storage device, or may be separate storage devices. The DB may also be an external storage device connected to the server 10, and the configuration thereof may be set arbitrarily.

[0036] 5 is a block diagram showing the functional configuration of the server 10. Functionally, the server 10 includes a question information DB 41, an input information acquisition unit 51, a condition information acquisition unit 52, a recommended question acquisition unit 53, and a screen output unit 54.

[0037] The input information acquisition unit 51, the condition information acquisition unit 52, the recommended question acquisition unit 53, and the screen output unit 54 are realized by the control unit 12 executing a program.

[0038] The input information acquisition unit 51 acquires, as input information, information about one or more of the evaluation items and keywords used in the rubric evaluation specified by the teacher. The teacher uses the teacher terminal 20 to input one or more of the evaluation items and keywords on a predetermined screen and transmits them to the server 10.

[0039] The condition information acquisition unit 52 acquires condition information related to the conditions of the recommended questions. Specifically, the conditions of the recommended questions are the criteria for the content of the questions to be acquired as recommended questions from the question information DB 41. For example, the condition information includes information on whether the conditions of the recommended questions are a perfect match, that is, a complete match, or a partial match, that is, a partial match, between the evaluation items included in the input information and the evaluation items corresponding to the recommended questions. The condition information also includes information on whether the conditions of the recommended questions are the number of matches between the keywords and the words included in the recommended questions, and the threshold for the number of matches. The condition information also includes information on whether the conditions of the recommended questions are the similarity between the recommended questions and the keywords, and the threshold for the similarity.

[0040] Note that inputting the conditions for the recommended questions is optional, and if the teacher does not input them, the recommended question acquisition unit 53, which will be described later, may acquire recommended questions based on preset conditions and thresholds. The conditions for the recommended questions may be one condition or a combination of multiple conditions, and can be set arbitrarily.

[0041] The recommended question acquisition unit 53 identifies questions from the question information DB 41 based on one or more of the evaluation items and keywords included in the input information, and acquires them as recommended questions. Methods for identifying recommended questions include a method based on the degree of match of evaluation items, a method based on the number of matches between the question and keywords, and a method based on the similarity between the question and keywords, and the recommended questions are identified by a method depending on the input information and condition information. Recommended questions may be identified by each identification method, or by combining multiple identification methods. The method for identifying recommended questions can be set arbitrarily.

[0042] The screen output unit 54 creates a screen including text showing the content of the recommended question obtained from the question information DB 41, and transmits screen information relating to the screen to the teacher terminal 20. By displaying the screen based on the screen information received by the teacher terminal 20, the teacher can check the content of the recommended question corresponding to the input information.

[0043] The screen output unit 54 may create a screen for the teacher to input input information, condition information, etc., and transmit screen information relating to the screen to the teacher terminal 20. In this embodiment, the details of which will be described later, the screen output unit 54 creates a screen for inputting input information and condition information and outputting recommended questions, and transmits information relating to the screen to the teacher terminal 20 in response to login and various requests.

[0044] In the above configuration, the question information DB 41 of the server 10 is an example of a storage unit of the present invention. Also, in the above configuration, the input information acquisition unit 51, condition information acquisition unit 52, recommended question acquisition unit 53, and screen output unit 54 of the server 10 are examples of an input information acquisition unit, condition information acquisition unit, recommended question acquisition unit, and recommended question output unit of the present invention, respectively. Also, in the above configuration, the recommended question acquisition unit 53 is an example of a similarity calculation unit and constituent word acquisition unit of the present invention.

[0045] [screen] Next, we will explain the screens created and output by the screen output unit 54. The teacher displays the screen by logging in to a predetermined site using the teacher terminal 20 and inputs any evaluation items and keywords. In addition, when the teacher makes a request by pressing a predetermined button on the screen, recommended questions corresponding to the input evaluation items and keywords are displayed on the screen.

[0046] An example of the screen is shown in Fig. 6. As shown in Fig. 6, the screen has an evaluation item input area 61, a keyword input area 62, an exact match checkbox 63, a partial match checkbox 64, a match count checkbox 65, a match count input area 66, a similarity checkbox 67, a similarity selection area 68, a search button 69, and recommended question display areas 70 and 71.

[0047] The evaluation item input area 61 is an area where the teacher inputs any evaluation item. Information about the evaluation item input in the evaluation item input area 61 is acquired by the server 10 as input information. The keyword input area 62 is an area where the teacher inputs any keyword that the teacher wants to include in the rubric evaluation question. Information about the keyword input in the keyword input area 62 is acquired by the server 10 as input information. The teacher only needs to input one or more of the evaluation item and keyword. Furthermore, the number of evaluation items and keywords input is not limited to one, and multiple may be input.

[0048] The exact match checkbox 63, the partial match checkbox 64, the number of matches checkbox 65, and the similarity checkbox 67 are checkboxes for selecting the conditions for the recommendation question.

[0049] The exact match checkbox 63 is a checkbox to be checked when a condition for a recommended question is that the question corresponds to an evaluation item that completely matches the evaluation item entered in the evaluation item input area 61. The partial match checkbox 64 is a checkbox to be checked when a condition for a recommended question is that the question corresponds to an evaluation item that partially matches the evaluation item entered in the evaluation item input area 61.

[0050] The number of matches checkbox 65 is a checkbox to be checked if the condition for the recommended question is that the question contains a threshold number of words that match the keyword entered in the keyword input area 62. The number of matches input area 66 is an area for inputting the number of matches between the keyword that serves as the threshold and the words. The similarity checkbox 67 is a checkbox to be checked if the condition for the recommended question is that the question has a similarity to the keyword entered in the keyword input area 62 that is a threshold number or more. The similarity selection area 68 is an area for selecting the threshold similarity using a radio button. Note that the similarity may be entered by directly entering a number instead of selecting using a radio button. Information regarding the checkbox selected as a condition for the recommended question and the information entered as the threshold are acquired by the server 10 as condition information.

[0051] When the teacher has finished inputting and selecting the evaluation items, keywords, and conditions for the recommended questions using the teacher terminal 20, he or she presses the search button 69 to send an output request for recommended questions to the server 10. Upon receiving the output request, the server 10 identifies and acquires one or more recommended questions that meet the conditions from the question information DB 41 based on the input information and condition information. The server 10 then creates a screen that includes the recommended questions and the evaluation items corresponding to the recommended questions, and sends screen information related to this screen to the teacher terminal 20. In this way, when the teacher presses the search button 69, the screen displays multiple recommended questions that meet the conditions and the evaluation items corresponding to the recommended questions.

[0052] The recommended question display areas 70 and 71 are areas for displaying recommended questions, and include question content areas 70a and 71a that display text indicating the content of the recommended question, and corresponding evaluation item display areas 70b and 71b that display evaluation items corresponding to the recommended question. In the question content areas 70a and 71a, words in the text that match keywords may be highlighted, as shown in Figure 6.

[0053] Furthermore, as shown in Figure 6, when the conditions of a recommended question are a partial match of the evaluation items, evaluation items other than the evaluation items entered by the teacher may be associated with the recommended question. In this case, as shown in the corresponding evaluation item display area 71b, the evaluation items entered by the teacher, "thinking ability" and "expression ability," may be displayed larger, and the other evaluation items, "comprehension ability" and "research ability," may be displayed smaller. In this way, when the conditions of a recommended question are a partial match of the evaluation items, the teacher can easily check whether the evaluation items corresponding to the recommended question are a perfect match, and if so, what other evaluation items are associated with them.

[0054] The teacher can easily use the screen to send input information and condition information to the server 10. The teacher can also easily recognize recommended questions that correspond to the items and keywords that he or she has arbitrarily input. The teacher can also arbitrarily edit the text that shows the content of the recommended questions displayed on the screen.

[0055] When logging in to a specific site, only the range indicated by the input at the top of the screen shown in Fig. 6 is displayed, and when search button 69 is pressed, the range indicated by the output at the bottom of the screen is displayed. The screen shown in Fig. 6 is an example, and the present invention is not limited to this, and the screen configuration can be set as desired.

[0056] [How to identify recommended questions] Next, a method in which the recommended question acquisition unit 53 of the server 10 identifies a recommended question based on input information will be described.

[0057] (Identification method based on the degree of agreement of evaluation items) First, we will explain the method of identifying the evaluation items based on the degree of matching. The recommended question acquisition unit 53 refers to the question information DB 41, and identifies and acquires, as recommended questions, questions corresponding to evaluation items that completely or partially match the evaluation items included in the input information.

[0058] 7(a) is a diagram illustrating a method for identifying a recommended question when the condition information includes an exact match of evaluation items as conditions for the recommended question. The evaluation items included in the input information are assumed to be "thinking ability" and "expressive ability." In this case, the recommended question acquisition unit 53 refers to the question information DB 41 shown in FIG. 7(a), and identifies and acquires the question 73 in which only the evaluation items "thinking ability" and "expressive ability" have check marks as the recommended question.

[0059] FIG. 7B is a diagram illustrating a method for identifying a case in which the condition information includes a partial match of evaluation items as the conditions for a recommended question. Assume that the evaluation items included in the input information are “thinking ability” and “expressive ability.” In this case, the recommended question acquisition unit 53 references the question information DB 41 shown in FIG. 7B and identifies and acquires questions 73, 74, and 75, which have check marks in the evaluation items “thinking ability” and “expressive ability,” as recommended questions. Unlike when the conditions for a recommended question are an exact match, if a question has check marks in the evaluation items “thinking ability” and “expressive ability,” the recommended question acquisition unit 53 identifies the question as a recommended question even if other evaluation items, such as questions 74 and 75, have check marks in the evaluation items “comprehension ability” and “research ability.” At this time, the recommended question acquisition unit 53 may acquire and temporarily store information about other evaluation items. This allows other evaluation items to be displayed together when a recommended question and the evaluation item corresponding to the recommended question are displayed on the screen.

[0060] (Identification method based on the number of matches between the problem and keywords) Next, we will explain the method of identification based on the number of matches between questions and keywords. If the condition information includes a condition for a recommended question that the question contains a threshold or more of words that match the keywords, and that threshold value, the recommended question acquisition unit 53 references the question information DB 41 and identifies and acquires as recommended questions questions that contain a threshold or more of words that match the keywords included in the input information.

[0061] FIG. 8 is a diagram illustrating a method for identifying questions based on the number of matches between questions and keywords. The keywords included in the input information are "attractive," "opinion," "seconds," and "communicate," and the threshold for the number of matches included in the condition information is "2." In this case, the recommended question acquisition unit 53 refers to the question information DB 41 shown in FIG. 8 and counts the number of matches between the words included in the questions and the keywords (hereinafter also referred to as the "number of keyword matches"). The recommended question acquisition unit 53 then identifies and acquires questions 73 and 75 for which the number of keyword matches is equal to or greater than the threshold "2" as recommended questions.

[0062] (Identification method based on the similarity between the problem and keywords) Next, we will explain the method of identification based on the similarity between questions and keywords. If the condition information includes a condition for recommended questions that the similarity between the question and keywords is equal to or greater than a threshold, and that threshold is included, the recommended question acquisition unit 53 references the question information DB 41 and identifies and acquires, as recommended questions, questions whose similarity with the keywords included in the input information is equal to or greater than the threshold.

[0063] The recommended question acquisition unit 53 first divides the text indicating the content of each question stored in the question information DB 41 into words, and acquires the constituent words that make up each question. Figure 9 is a diagram explaining a method for dividing the text indicating the content of the questions. The recommended question acquisition unit 53 uses morphological analysis to divide the text indicating the content of the questions into words, as shown in the figure, and acquires the constituent words.

[0064] The recommended question acquisition unit 53 refers to a knowledge graph created in advance and calculates the similarity between the keywords and the constituent words of each question stored in the question information DB 41. Here, we will explain the knowledge graph. A knowledge graph is a graph structure that represents the connections between knowledge represented by multiple words, and may be generated manually or automatically using machine learning.

[0065] When machine learning is used, a knowledge graph is constructed by performing named entity recognition on a given text and then extracting relationships. FIG. 10 is a diagram illustrating a knowledge graph construction method. As shown in FIG. 10(a), the server 10 classifies each word extracted from the text "Taro arrived at Company X at 9:00 a.m. on March 17th" into classes such as person's name, date, time, and organization. The server 10 then constructs a trained model capable of named entity recognition by training a given learning model using the text and classes shown in FIG. 10(a) as training data. The server 10 also estimates the relationships between each word, such as "Taro, person's name, Company X, affiliation," from the results of the named entity recognition, as shown in FIG. 10(b). The server 10 then constructs a trained model capable of relationship extraction by training a given model using the results of the named entity recognition and the relationships shown in FIG. 10(b) as training data. The server 10 automatically generates a knowledge graph using such a trained model.

[0066] In this embodiment, it is assumed that a knowledge graph centered on all words has been created in advance. The recommended question acquisition unit 53 refers to the knowledge graph centered on keywords and calculates the similarity between the keywords and the constituent words of each question stored in the question information DB 41. The recommended question acquisition unit 53 then calculates the similarity between each question and the keywords by brute-force calculating the similarity between all keywords and all of the constituent words of each question or specific parts of speech such as nouns and adjectives.

[0067] FIG. 11(a) is a diagram illustrating a method for calculating the similarity between constituent words and keywords. Specifically, a method for the recommended question acquisition unit 53 to calculate the similarity between the keyword "attractiveness" and the constituent words of the question shown in FIG. 11(a) will be described. First, the recommended question acquisition unit 53 refers to the knowledge graph 76 centered on the keyword "attractiveness" and determines the distance from the constituent words of the question. For example, the constituent word "friends" of the question is located at 76b via 76a from the keyword "attractiveness" at the center of the knowledge graph 76. In this case, the distance between the constituent word "friends" and the keyword "attractiveness" is calculated to be "2".

[0068] In this embodiment, the similarity is highest at "1," and the similarity is reduced by "0.1" for every "1" increase in distance. According to this, the recommended question acquisition unit 53 calculates the similarity between the constituent word "friends" and the keyword "attractiveness" to be "0.8." Using a similar calculation method, the recommended question acquisition unit 53 calculates the similarity between the constituent word "good points" and the keyword "attractiveness" to be "0.9." Note that if a constituent word does not exist in a knowledge graph centered on a keyword, the similarity between that constituent word and the keyword is "0." Therefore, the recommended question acquisition unit 53 sets the similarity between the constituent word "bad points" and the keyword "attractiveness" to "0."

[0069] 11(b) is a diagram illustrating a method for calculating the similarity between a question and a keyword. Specifically, the recommended question acquisition unit 53 calculates the similarity between all keywords "attractiveness," "opinion," "second," and "communicate" and all constituent words of question 77 "Talk about the good and bad points of your friend in two minutes or less," in a brute-force manner, and calculates the similarity between the question and the keyword by calculating the sum of the similarities. As a result, as shown in the figure, the similarity between question 77 and the keyword is "2.4," and the similarity between question 73 and the keyword is "3.2."

[0070] 12 is a diagram illustrating a method for identifying questions based on the similarity between questions and keywords. The threshold value for similarity included in the condition information is assumed to be "2." In this case, the recommended question acquisition unit 53 refers to the question information DB 41 shown in FIG. 12, and identifies and acquires questions 73, 75, and 77 whose similarity between the questions and keywords is equal to or greater than the threshold value of "2" as recommended questions.

[0071] [Output Processing] Next, we will explain the process of outputting recommended questions that meet the conditions and correspond to the evaluation items and keywords based on the teacher's input. Figure 13 is a flowchart of the output process by the server 10. This process is realized by the server 10 executing a program prepared in advance.

[0072] The teacher logs in to a specific site using the teacher terminal 20 to display the screen, and inputs and selects the evaluation items, keywords, and conditions for the recommended questions. After completing the input and selection, the teacher presses a specific button on the screen to send the input information and condition information to the server 10 and request the output of the recommended questions.

[0073] The server 10 acquires input information and condition information from the teacher terminal 20 (step S101). First, based on the input information and condition information, the server 10 identifies and acquires, as recommended questions, questions from the question information DB 41 that satisfy the conditions included in the condition information and correspond to the evaluation items and keywords included in the input information (step S102). The server 10 then creates a screen including one or more recommended questions and the evaluation items corresponding to the recommended questions, and transmits screen information related to the screen to the teacher terminal 20 (step S103). This completes the output process. The teacher terminal 20 displays a screen based on the screen information received from the server 10. This allows the teacher to easily recognize recommended questions that satisfy the conditions based on the information they have arbitrarily input and the evaluation items corresponding to the recommended questions. The teacher can also modify the content of the recommended questions as necessary.

[0074] In this embodiment, the server 10 outputs screen information including the recommended questions and the evaluation items corresponding to the recommended questions, but the present invention is not limited to this, and it is also possible to output information related to the recommended questions and the evaluation items corresponding to the recommended questions. Also, it is also possible to output only the recommended questions, rather than the recommended questions and the evaluation items corresponding to the recommended questions as a set. The method for outputting the recommended questions and the evaluation items corresponding to the recommended questions can be set arbitrarily.

[0075] The learning support system 100 of this embodiment can recommend to the learning supporter the content of questions that satisfy the conditions included in the arbitrarily input condition information and correspond to the evaluation items and keywords included in the input information. This makes it possible for the learning supporter to create questions for rubric evaluation more efficiently and reduces the workload of the learning supporter.

[0076] <Modification> In the above embodiment, the teacher uses the teacher terminal 20, but the present invention is not limited to this, and the teacher may use a teacher terminal 90 that has the functions of the server 10. Like the server 10, the teacher terminal 90 is, for example, a personal computer or a general-purpose tablet PC (personal computer).

[0077] 14 shows an example of the configuration of a learning support system 200 in this case. A question information DB 91 is connected to the teacher terminal 90, and the teacher terminal 90 executes the output processing that was previously performed by the server 10, and is able to identify and output recommended questions that meet the conditions arbitrarily input by the teacher and correspond to the evaluation items and keywords. In this case, the teacher terminal 90 is an example of the information processing device of the present invention. [Explanation of symbols]

[0078] 5. Network 10 Servers 20, 90 Teacher's terminal 41, 91 Problem information DB 51 Input information acquisition unit 52 Condition information acquisition section 53 Recommendation question acquisition department 54 Screen output section 100, 200 Learning Support System

Claims

1. a storage unit that stores questions used in rubric evaluation and evaluation items that indicate evaluation viewpoints in association with each other; an input information acquisition unit that acquires information regarding one or more of the evaluation items and keywords as input information; a recommended question acquisition unit that acquires, from the storage unit, target questions as recommended questions based on the evaluation items and keywords included in the questions from the input information; a recommendation question output unit that outputs the recommendation question; An information processing device comprising:

2. The information processing device according to claim 1 , wherein the recommended question acquisition unit acquires, from the storage unit, a question corresponding to an evaluation item that completely matches an evaluation item included in the input information, as a recommended question.

3. The information processing device according to claim 1 , wherein the recommended question acquisition unit acquires, from the storage unit, questions corresponding to evaluation items that partially match evaluation items included in the input information, as recommended questions.

4. The information processing device according to claim 1 , wherein the recommended question acquisition unit acquires, from the storage unit, questions that contain a threshold or more of words that match keywords included in the input information as recommended questions.

5. a similarity calculation unit that calculates a similarity between the keyword and the recommended question, The information processing device according to claim 1 , wherein the recommended question acquisition unit acquires, from the storage unit, a question whose similarity is equal to or greater than a threshold value as a recommended question.

6. a constituent word acquisition unit that divides a text indicating the content of the question and acquires constituent words that are words that constitute the question; The information processing device according to claim 5 , wherein the similarity calculation unit calculates the similarity based on a knowledge graph that represents, in a graph structure, connections of knowledge represented by a plurality of words centered on the evaluation item, and the constituent words.

7. a condition information acquisition unit that acquires condition information related to the conditions of the recommendation problem; The information processing device according to claim 5 , wherein the recommended question acquisition unit acquires the recommended questions from the storage unit based on the input information and the condition information.

8. The information processing device according to claim 7 , wherein the condition information includes information regarding whether an evaluation item included in the input information completely matches or partially matches an evaluation item corresponding to the recommended question.

9. The information processing apparatus according to claim 7 , wherein the condition information includes information on the number of matches between a keyword and a word serving as a threshold, or information on a similarity degree serving as a threshold.

10. The information processing device according to claim 1 , wherein the recommendation question output unit displays text indicating the content of the recommendation question, and also displays keywords included in the text in an emphasized manner.

11. A program executed by an information processing device having a computer, a storage unit that stores questions used in rubric evaluation and evaluation items that indicate evaluation viewpoints in association with each other; an input information acquisition unit that acquires information relating to one or more of the evaluation items and keywords as input information; a recommended question acquisition unit that acquires, from the storage unit, target questions based on the evaluation items and keywords included in the questions from the input information, as recommended questions; a recommendation question output unit that outputs the recommendation question; A program that causes the computer to function as a

Images