Information processing device and program

Abstract

Easily obtain learning codes to assign to study materials such as tests and textbooks. [Solution] The information processing device acquires information about learning materials. The information processing device acquires learning code information about learning codes that classify learning materials. The information processing device has the information about learning materials and the learning code information, and creates an assignment request prompt that requests a learning code to be assigned to the learning materials. By inputting the assignment request prompt into the generating AI, the device acquires and outputs assignment information indicating the learning code to be assigned to the learning materials.

Description

Technical Field

[0001] The present invention relates to a technology for assisting the learning of learners.

Background Art

[0002] Conventionally, there is known a system that can save the test results implemented for students in a database and view them on a computer. Patent Document 1 discloses an educational support system having a score aggregation database that aggregates the test scores of learners and configured such that the database can be referred to by a tablet terminal device.

[0003] In the educational field, instructors such as teachers assign learning codes for indicating learning contents to each question of a test. Thereby, it becomes possible to search for questions suitable for the class content and perform a retrospective analysis of test results by referring to the database based on the learning codes. Note that for public education, learning guideline codes are used. The learning guideline codes are codes formulated in accordance with the learning guidelines of the Ministry of Education, Culture, Sports, Science and Technology. For example, corresponding learning guideline codes are assigned and tagged to each question of a test.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Currently, for each question of a test, since learning codes are assigned manually, the burden is large. Also, there is a problem that the results vary depending on the operator.

[0006] The present invention was made, for example, to solve the above-mentioned problems, and aims to provide an information processing device that can easily obtain learning codes to be assigned to learning materials such as tests and reference books. [Means for solving the problem]

[0007] In one aspect of the present invention, the information processing device includes a means for acquiring information about learning materials, a means for acquiring information about learning codes that classify the learning materials, a means for creating an assignment request prompt that has the information about the learning materials and the learning code information and requests a learning code to be assigned to the learning materials, and an assignment information indicating a learning code to be assigned to the learning materials obtained and output by inputting the assignment request prompt to a generating AI. According to this embodiment, the information processing device can utilize the generating AI to obtain a learning code to be assigned to the learning materials as assignment information.

[0008] In one embodiment of the information processing device described above, the learning material is a test, the material information is information relating to the test, the assignment request prompt creation means has the material information and the learning code information, and creates an assignment request prompt requesting a learning code to be assigned to each question constituting the test, and the assignment information acquisition means acquires and outputs assignment information indicating each question constituting the test and the learning code to be assigned to each question. According to this embodiment, the information processing device can utilize a generation AI to acquire a learning code to be assigned to each question constituting the test as assignment information.

[0009] In one embodiment of the information processing device described above, the grant request prompt further includes model answers for each of the questions constituting the test. In this embodiment, the information processing device can obtain more accurate grant information.

[0010] In one embodiment of the above-described information processing device, the device includes a specific request prompt creation means that has the teaching material information and creates a specific request prompt requesting the identification of the subject of the test, and a subject acquisition means that inputs the specific request prompt to a generating AI to identify and acquire the subject of the test, wherein the learning code information acquisition means acquires learning code information corresponding to the identified subject, and the assignment request prompt creation means has the teaching material information and the learning code information corresponding to the subject and requests a learning code to be assigned to each question constituting the test. According to this embodiment, even if the file size of the learning code information is large and exceeds the upper limit of the input token limit of the generating AI, the information processing device can acquire the assignment information using the generating AI by attaching only the learning code information corresponding to the identified subject to the assignment request prompt. Furthermore, the output accuracy can be improved by extracting a part of the learning code information according to the identified subject and providing it to the generating AI.

[0011] In one embodiment of the information processing device described above, the learning code consists of three classifications: major classification, medium classification, and minor classification, the assignment request prompt includes the specification of the classification, and the assignment information acquisition means acquires assignment information indicating the learning code of the specified classification. According to this embodiment, the information processing device can improve the processing efficiency of the generating AI and acquire assignment information more quickly by having the generating AI assign only the learning code of the specified classification.

[0012] In one embodiment of the above-described information processing device, the device further comprises a correlation calculation means for calculating and outputting the correlation between a problem and a learning code assigned to that problem, wherein the assigned information includes each problem, the learning code assigned to each problem, and the correlation. According to this embodiment, the information processing device allows the user to identify learning codes with a high correlation as having a high degree of confidence among multiple learning codes assigned to a single problem. Furthermore, even if only one learning code is assigned to a single problem, the user can identify learning codes with a low correlation as having a low degree of confidence.

[0013] In one embodiment of the above-described information processing device, a screen creation means is provided that, based on the assigned information, separates each problem on the test, creates a screen that displays the problem and the learning code assigned to that problem in association, and outputs the screen. According to this embodiment, the user can easily check and verify the learning code assigned to each problem by looking at the screen. In this way, by checking based on the learning code assigned by the generating AI, the workload of the user can be reduced, and variations in the assignment of learning codes by workers can be eliminated.

[0014] In another aspect of the present invention, a program executed by an information processing device equipped with a computer is provided, which acquires material information relating to learning materials, acquires learning code information relating to learning codes for classifying the learning materials, creates an assignment request prompt having the material information and the learning code information and requesting a learning code to be assigned to the learning materials, and inputs the assignment request prompt to a generating AI, thereby causing the computer to execute a process of acquiring and outputting assignment information indicating a learning code to be assigned to the learning materials. By installing and executing this program on a computer, an information processing device according to the present invention can be configured. [Effects of the Invention]

[0015] According to the information processing device of the present invention, learning codes to be assigned to learning materials such as tests and reference books can be easily obtained. [Brief explanation of the drawing]

[0016] [Figure 1] The configuration of the learning support system is shown. [Figure 2] Block diagram showing the server hardware configuration. [Figure 3] This is an example of learning code information. [Figure 4] This is an example of a curriculum guidelines code. [Figure 5] This is a block diagram showing the functional configuration of the server in the first embodiment. [Figure 6] This is an example of a test. [Figure 7] This is an example of an assignment request prompt and an output result. [Figure 8] This is a presentation example of the learning code assigned to each question. [Figure 9] This is a flowchart of the assignment information acquisition process in the first embodiment. [Figure 10] This is an example of an arithmetic test. [Figure 11] This is a block diagram showing the functional configuration of the server in the second embodiment. [Figure 12] This is an example of a specific request prompt and an output result. [Figure 13] This is a flowchart of the assignment information acquisition process in the second embodiment.

Mode for Carrying Out the Invention

[0017] Hereinafter, embodiments of the present invention will be described with reference to the drawings. <First Embodiment> [Overall Configuration] FIG. 1 shows the configuration of a learning support system to which the server of the present invention is applied. In the first embodiment, the learning support system 100 is a system that acquires assignment information indicating a learning code to be assigned to learning materials such as tests and reference books.

[0018] The learning code is a code for classifying learning materials. In this embodiment, as an example, a learning guideline code is applied. The learning guideline code is a 16-digit code formulated in accordance with the learning guidelines of the Ministry of Education, Culture, Sports, Science and Technology. Although the details will be described later, certain rules are provided for the digits to facilitate searches for school type, subject, grade, etc.

[0019] Learning materials include tests, textbooks, study guides, supplementary materials, and other documents. In this embodiment, as an example, a test consisting of multiple questions is used. The learning support system 100 acquires assignment information indicating each question that makes up the test and the learning code to be assigned to each question. Based on the assignment information, the learning support system 100 presents the learning code to be assigned to each question to the user in an easy-to-understand format.

[0020] The learning support system 100 is configured so that a teacher's terminal 10, a server 20, and a generating AI (Artificial Intelligence) 30 can communicate with each other via a network 5 such as the Internet. The teacher's terminal 10 is used by users such as school teachers who instruct and manage students, and is an information processing device such as a tablet or PC. Specifically, the teacher's terminal 10 sends information about tests to which learning codes should be assigned to the server 20, and receives and displays information from the server 20 that presents the learning codes to be assigned to each question.

[0021] Server 20 is an information processing device that processes, stores, and transmits various types of information, and can be, for example, a server device, a personal computer (PC), or a general-purpose tablet. Specifically, as will be described in detail later, Server 20 receives information related to the test from the teacher terminal 10 and transmits information to the teacher terminal 10 that presents learning codes to be assigned to each question. Server 20 is also connected to a learning code database (hereinafter, "database" will be referred to as "DB") 31 that stores information related to learning codes that classify each question that makes up the test. Server 20 is an example of the information processing device of the present invention.

[0022] The generating AI 30 is a Large Language Model (LLM) capable of understanding multimodal information and is connected to the server 20 via the network 5 for communication. However, the present invention is not limited to this, and the generating AI 30 may be installed on the server 20 without using the network 5.

[0023] [Server hardware configuration] Figure 2 is a block diagram showing the hardware configuration of server 20. Server 20 comprises 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 the learning code DB 31 are interconnected via a bus 19.

[0024] Server 20 can run on a single computer, or it can run in a distributed manner across multiple computers, or it can run in a distributed manner across virtual machines.

[0025] The communication unit 11 is a communication unit for communicating with the teacher terminal 10 and the generating AI 30 via the network 5. Specifically, the communication unit 11 receives information related to tests from the teacher terminal 10 and transmits information to the teacher terminal 10 that presents learning codes to be assigned to questions.

[0026] The control unit 12 includes arithmetic processing units such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), and GPU (Graphics Processing Unit), and performs various information processing and control processing related to the server 20 by reading and executing programs stored in the memory unit 13. The programs can be deployed on a single computer or site, or distributed across multiple sites and executed on multiple computers interconnected by a communication network. Although Figure 2 describes the control unit 12 as a single processor, it may also be a multi-processor system.

[0027] The storage unit 13 includes memory elements such as RAM (Random Access Memory) and ROM (Read Only Memory), and stores programs or data necessary for the control unit 12 to execute processing. The storage unit 13 also temporarily stores data necessary for the control unit 12 to execute arithmetic processing.

[0028] The recording medium 14 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or semiconductor memory, and is configured to be detachable from the server 20. The recording medium 14 stores various programs that the control unit 12 executes. When the server 20 executes the assignment information acquisition process, the programs stored on the recording medium 14 are loaded into the storage unit 13 and executed by the control unit 12.

[0029] The display unit 15 is a liquid crystal display or an organic EL (electroluminescence) display, etc., and displays various information according to the instructions of the control unit 12. The input unit 16 is an input device such as a mouse, keyboard, touch panel, or buttons, and outputs the received operation information to the control unit 12.

[0030] The Learning Code DB31 stores learning code information related to learning codes used to classify learning materials. Figure 3 shows an example of learning code information. As shown in Figure 3, the learning code information is a table that associates subjects, curriculum guideline texts, and curriculum guideline codes. For example, in the subject of "Social Studies," the text corresponding to the "Geography" field is classified under curriculum guideline code "83212A0000000000". Furthermore, according to the learning code information, in the "Geography" field, the text corresponding to "understanding regional phenomena and regional characteristics regarding the land of Japan and various regions of the world, and acquiring the skills to effectively investigate and summarize various geographical information from surveys and various materials" is classified under curriculum guideline code "83212A2100000000".

[0031] Here, we will explain the curriculum guidelines code. The curriculum guidelines code is not a code set independently by a specific textbook manufacturer, but a standard code used by all textbook and teaching material manufacturers. Figure 4 shows an example of a curriculum guidelines code. As shown in Figure 4(a), the curriculum guidelines code is a 16-digit number mechanically assigned to all items of the curriculum guidelines according to classifications such as the date of notification, school type, subject, and grade level. Specifically, certain rules are in place so that the first digit indicates the date of notification, the second digit indicates the school type, the third digit indicates the subject, the fourth digit indicates the field or subject, the fifth digit indicates the major item of objectives or content, the sixth digit indicates the grade level or stage, the seventh digit indicates the minor item of objectives or content, the eighth to fifteenth digits indicate the details, and the sixteenth digit indicates the status of any partial revisions made at the time of notification.

[0032] Furthermore, the curriculum guidelines code may consist of three classifications: a major classification, a medium classification, and a minor classification. For example, as shown in Figure 4(b), the first to second digits may be set as the major classification, the third to eleventh digits as the medium classification, and the twelfth to sixteenth digits as the minor classification, so that the curriculum guidelines code consists of three classifications. Note that the number of digits for each of the major, medium, and minor classifications can be set arbitrarily.

[0033] [Server Functional Configuration] Figure 5 is a block diagram showing the functional configuration of server 20. Server 20 is connected to the learning code DB 31 and functionally comprises a test information acquisition unit 40, a learning code information acquisition unit 41, a grant request prompt creation unit 42, a grant information acquisition unit 43, and a screen creation unit 44. The test information acquisition unit 40, the learning code information acquisition unit 41, the grant request prompt creation unit 42, the grant information acquisition unit 43, and the screen creation unit 44 are realized by the control unit 12 executing a program.

[0034] The process by which the information acquisition unit 43 acquires the information is implemented by the generation AI 30.

[0035] The test information acquisition unit 40 acquires test information related to the test. Figure 6 shows an example of a test. As shown in Figure 6, the test consists of multiple questions and has answer fields corresponding to each question. In this embodiment, as an example, the test information is obtained from the teacher's terminal 10 as image data obtained by scanning the test. However, the test information is not limited to this, and its format can be arbitrarily set, such as data of the test converted to PDF.

[0036] The learning code information acquisition unit 41 acquires learning code information from the learning code DB 31. In this embodiment, the learning code information is acquired from the learning code DB 31 provided by the server 20, but the present invention is not limited to this, and the method of acquiring learning code information can be arbitrarily set.

[0037] The assignment request prompt creation unit 42 has test information and learning code information and generates prompts that request learning codes to be assigned to each question. Here, the "prompt that requests learning codes to be assigned to each question" is also called the "assignment request prompt". Specifically, the assignment request prompt creation unit 42 creates assignment request prompts that request learning codes to be assigned to each question, attaching test information and learning code information to instructional statements such as "Please determine the appropriate learning code from the learning code information for each question that makes up the test and assign it" or "We would like you to take on the task of assigning learning codes to each question of the test".

[0038] Figure 7(a) shows an example of a grant request prompt. As shown in Figure 7(a), the grant request prompt 50 includes an instruction statement 51 requesting a learning code to be assigned to each problem, test image data 52, and learning code information 53.

[0039] The assignment information acquisition unit 43 obtains assignment information indicating the learning code to be assigned to each problem by sending the assignment request prompt created by the assignment request prompt creation unit 42 to the generation AI 30. In other words, when the generation AI 30 receives an assignment request prompt from the server 20, it automatically assigns and outputs a learning code corresponding to the content of each problem that makes up the test. Specifically, the generation AI 30 recognizes the structure of the problems that make up the test, selects an appropriate learning code from the learning code information according to the content of each problem, tags it, and sends the output result to the server 20 as assignment information. At this time, the generation AI 30 may also output information regarding the structure of the problems that make up the test and include it in the assignment information before sending it to the server 20.

[0040] Specifically, the assignment information acquisition unit 43 sends an assignment request prompt 50, as shown in Figure 7(a), to the generation AI 30 as input. Figure 7(b) is an example of the output result of the generation AI 30. When an assignment request prompt 50, as shown in Figure 7(a), is input, the generation AI 30 assigns a learning code to each question that makes up the test and outputs the output result 55 shown in Figure 7(b).

[0041] As shown in Figure 7(b), the output result 55 has the text "Learning code assigned", text 56 corresponding to sub-question 1 which makes up main question 1, text 57 corresponding to sub-question 2, and text 58 corresponding to sub-question 3. Text 56 shows the answers and learning codes for each of the 1 to 7 prefectures on the map, i.e., sub-questions 1 to 7, for sub-question 1, "Name the prefectures 1 to 7 on the map and their respective prefectural capitals." Text 57 shows the answers and learning codes for sub-question 2, "Name the river with the largest drainage basin in Japan, known as Bando Taro." Text 58 shows the answers and learning codes for "Name the coastline with a long stretch of sandy beach in eastern Chiba Prefecture."

[0042] The generating AI 30 sends the output result 55 as annotation information to the server 20. As a result, the annotation information acquisition unit 43 obtains annotation information indicating the learning code to be assigned to each question that makes up the test.

[0043] Furthermore, the assignment request prompt creation unit 42 can include a specification of the learning code classification in the instruction statement, such as "Please assign only the subcategory of the learning code," thereby causing the generating AI to assign only the learning code of the specified classification. In this way, for example, by assigning only the learning code of the subcategory, which varies depending on the content of the problem, to the generating AI 30, excluding the major and minor categories, which are mostly fixed, the processing efficiency of the generating AI 30 can be improved, and the assignment information acquisition unit 43 can acquire the assignment information more quickly.

[0044] The screen creation unit 44 creates and outputs a screen that displays the learning codes assigned to each question in a user-friendly format, based on the assigned information. Specifically, the screen creation unit 44 divides each question on the test based on the assigned information, and creates and outputs a screen that links each question with the learning code assigned to that question. Figure 8(a) is an example of a screen. As shown in Figure 8(a), the screen has dotted lines 61, 62, and 63 that divide the main question 1 that makes up the test. Within dotted line 61, the sub-question 1, the answer field for sub-question 1, and the learning code assigned to sub-question 1 are displayed. Within dotted line 62, the sub-question 2, the answer field for sub-question 2, and the learning code assigned to sub-question 2 are displayed. Within dotted line 63, the sub-question 3, the answer field for sub-question 3, and the learning code assigned to sub-question 3 are displayed. In this way, the screen visualizes and displays the area of ​​each question on the test and the learning code assigned to each question. The screen creation unit 44 transmits screen information related to the screen to the teacher terminal 10, and the teacher terminal 10 displays the screen to present the user with the area of ​​each problem and the learning code to be assigned to each problem.

[0045] In the example shown in Figure 8(a), each sub-question is assigned one learning code, but this is not the only option. For example, each sub-question that makes up sub-question 1 may be assigned a different learning code. In this case, the screen creation unit 44 separates each sub-question on the test and creates and outputs a screen that displays each sub-question linked to the learning code assigned to it.

[0046] In this embodiment, the learning codes to be assigned to each problem are presented to the user on a screen as shown in Figure 8(a). However, the present invention is not limited to this, and for example, as shown in Figure 8(b), a table associating the text of each problem with the learning codes to be assigned to each problem may be presented to the user. In this case, the server 20 transmits the information in the table as shown in Figure 8(b) to the teacher terminal 10. The format of the table can be arbitrarily set to Excel, CSV (Comma Separated Values), etc.

[0047] In the above configuration, the test information acquisition unit 40, learning code information acquisition unit 41, assignment request prompt creation unit 42, assignment information acquisition unit 43, and screen creation unit 44 of the server 20 are examples of the teaching material information acquisition means, learning code information acquisition means, assignment request prompt creation means, assignment information acquisition means, and screen creation means of the present invention.

[0048] [Process to obtain assigned information] Next, we will describe the process of acquiring the assigned information, which involves assigning learning codes to the questions that make up the test and presenting them to the user. Figure 9 is a flowchart of the process of acquiring the assigned information in the first embodiment. This process is mainly achieved by the server 20 executing a pre-prepared program.

[0049] The user uses the teacher terminal 10 to send test information to the server 20 regarding the test for which they wish to be assigned a learning code (step S101). The server 20 receives the test information (step S102). Next, the server 20 retrieves learning code information from the learning code DB 31 (step S103). The server 20, having the test information and learning code information, creates an assignment request prompt requesting a learning code to be assigned to each question and sends it to the generation AI 30 (step S104). The generation AI 30 receives the assignment request prompt and inputs it (step S105). Next, the generation AI 30 assigns a learning code to each question that makes up the test and sends the output result to the server 20 as assignment information (step S106).

[0050] Server 20 obtains assignment information from the generating AI 30 (step S107). Next, based on the assignment information, Server 20 creates a screen that displays the learning codes to be assigned to each problem in a format that is easy for the user to understand, and sends the screen information to the teacher terminal 10 (step S108). The teacher terminal 10 receives the screen information and displays a screen showing the learning codes to be assigned to each problem that makes up the test (step S109). Thus, the assignment information acquisition process is completed.

[0051] This allows users to easily check and verify the learning codes assigned to each problem. By checking based on the learning codes assigned by the generating AI30, the user's workload is reduced, and inconsistencies in learning code assignment by different operators can be eliminated.

[0052] Furthermore, if the user determines, after checking, that there are no problems with the learning code assigned by the generating AI30, the user may present it to students by printing or sharing the screen. In this case, students can easily check which learning code corresponds to each question on the test. Since the learning codes are the same as those in textbooks and reference books, students can review the material appropriately using the learning codes.

[0053] In this embodiment, the information acquisition process is applied to a social studies test for second-year junior high school students, as shown in Figure 6, as an example. However, the present invention is not limited to this and can be applied to tests for any subject in any grade. For example, it can be applied to a mathematics test for first-grade elementary school students, as shown in Figure 10(a). In this case, the server 20 presents the learning code assigned to each question to the user by creating a screen as shown in Figure 10(b) based on the information assigned.

[0054] Furthermore, although learning materials are used as a test in this embodiment, the present invention is not limited to this. By performing similar processing on textbooks, study guides, supplementary materials, data collections, etc., the AI ​​30 can automatically assign learning codes, and the server 20 can acquire the assigned information.

[0055] <Second Embodiment> If the learning code information file size is large, it may exceed the upper limit of the input token (number of characters) limit for the generating AI30. In this case, the file size is reduced by narrowing down some of the learning code information related to the test content by subject or by pre-extracting it through searching, and then inputting it into the generating AI30.

[0056] In the second embodiment, the process for obtaining supplemental information when using RAG (Retrieval-Augmented Generation) will be described. In this embodiment, RAG is a method that uses search technology to extract a portion of the learning code information related to the test content, keeps the file size within the input token limit of the generating AI 30, and improves output accuracy by providing the extracted learning code information to the generating AI.

[0057] In the second embodiment, the learning support system 100x is configured so that the teacher terminal 10, the server 20x, and the generating AI 30 can communicate with each other via a network 5 such as the Internet. The overall configuration and the server hardware configuration are the same as in the first embodiment, so for convenience, a description is omitted.

[0058] [Server Functional Configuration] Figure 11 is a block diagram showing the functional configuration of server 20x. Server 20x is connected to the learning code DB 31 and functionally comprises a test information acquisition unit 70, a specific request prompt creation unit 71, a subject acquisition unit 72, a learning code information acquisition unit 73, an assignment request prompt creation unit 74, an assignment information acquisition unit 75, and a screen creation unit 76. The test information acquisition unit 70, the specific request prompt creation unit 71, the subject acquisition unit 72, the learning code information acquisition unit 73, the assignment request prompt creation unit 74, the assignment information acquisition unit 75, and the screen creation unit 76 are realized by the control unit 12 executing a program.

[0059] Furthermore, the process by which the subject acquisition unit 72 acquires subjects, and the process by which the assigned information acquisition unit 75 acquires assigned information, are implemented by the generation AI 30.

[0060] The test information acquisition unit 70 acquires test information related to the test. Since the test information acquisition unit 70 is the same as the test information acquisition unit 40 in the first embodiment, its explanation will be omitted for convenience.

[0061] The specific request prompt creation unit 71 creates a prompt that has test information and requests the search and identification of a learning code area related to the test. Here, the "prompt that requests the search and identification of a learning code area related to the test" is also called the "specific request prompt". The learning code area related to the test specifically refers to the subject or field corresponding to the test, and in this embodiment, as an example, it identifies the subject of the test. Specifically, the specific request prompt creation unit 71 creates a specific request prompt that requests the search and identification of a learning code area related to the test, which includes instructional statements such as "Please tell me the subject of this test" and "Please search for and identify the subject and field of this test," and to which test information is attached.

[0062] Figure 12(a) shows an example of a specific request prompt. As shown in Figure 12(a), the specific request prompt 80 includes an instruction statement 81 requesting the identification of the subject of the test, and image data 52 of the test.

[0063] The subject acquisition unit 72 acquires the subject of the identified test by sending a specific request prompt to the generation AI 30. In other words, when the generation AI 30 receives a specific request prompt from the server 20x, it identifies and outputs the subject of the test.

[0064] Specifically, the subject acquisition unit 72 sends a specific request prompt 80, as shown in Figure 12(a), to the generating AI as input. Figure 12(b) is an example of the output result of the generating AI 30. When the specific request prompt 80, as shown in Figure 12(a), is input, the generating AI 30 searches for and identifies the subject of the test and outputs it as the output result 85 shown in Figure 12(b). The generating AI 30 sends the output result 85 to the server 20. As a result, the subject acquisition unit 72 acquires the subject of the test.

[0065] The learning code information acquisition unit 73 acquires a portion of the learning code information related to the identified subject from the learning code DB 31. In other words, the learning code information acquisition unit 73 acquires a portion of the learning codes related to the test content, narrowing them down by the subject identified using the generation AI 30.

[0066] The assignment request prompt creation unit 74 has test information and learning code information, and generates assignment request prompts that request learning codes to be assigned to each question. The learning code information attached to the assignment request prompt is a part of the learning code information related to the test content, which is acquired by the learning code information acquisition unit 73. Except for the fact that only a part of the learning code information, not all of it, is attached to the assignment request prompt, this is the same as the assignment request prompt creation unit 42 of the first embodiment, so for convenience, the explanation is omitted.

[0067] The assignment information acquisition unit 75 acquires assignment information indicating the learning code to be assigned to each problem by sending the assignment request prompt created by the assignment request prompt creation unit 74 to the generation AI 30.

[0068] The screen creation unit 76 creates and outputs a screen that displays the learning code to be assigned to each problem in a format that is easy for the user to understand, based on the assigned information.

[0069] Note that the information acquisition unit 75 and the screen creation unit 76 are the same as the information acquisition unit 43 and the screen creation unit 44 of the first embodiment, respectively, so for convenience, their explanation will be omitted.

[0070] In the above configuration, the test information acquisition unit 70, specific request prompt creation unit 71, subject acquisition unit 72, learning code information acquisition unit 73, assignment request prompt creation unit 74, assignment information acquisition unit 75, and screen creation unit 76 of the server 20x are examples of the teaching material information acquisition means, specific request prompt creation means, subject acquisition means, learning code information acquisition means, assignment request prompt creation means, assignment information acquisition means, and screen creation means of the present invention.

[0071] [Process to obtain assigned information] Next, the process for acquiring assigned information in the second embodiment will be described. Figure 13 is a flowchart of the process for acquiring assigned information in the second embodiment. This process is mainly achieved by the server 20x executing a pre-prepared program.

[0072] The user uses the teacher terminal 10 to send test information to the server 20x regarding the test for which they wish to be assigned a learning code (step S201). The server 20x receives the test information (step S202). Next, the server 20x has the test information and creates a specific request prompt that requests the identification of the subject of the test, and sends it to the generating AI 30 (step S203). The generating AI 30 receives the specific request prompt and inputs it (step S204). Next, the generating AI 30 identifies the subject of the test, outputs it, and sends the output result to the server 20x (step S205).

[0073] Server 20x obtains the subject of the identified test from the output results of the generating AI 30 (step S206). Next, Server 20x obtains a portion of the learning code information corresponding to the identified subject from the learning code DB 31 (step S207). Server 20x, possessing the test information and a portion of the learning code information, generates an assignment request prompt requesting the learning code to be assigned to each question and sends it to the generating AI 30 (step S208).

[0074] The generating AI 30 receives and inputs an assignment request prompt (step S209). Next, the generating AI 30 assigns and outputs a learning code corresponding to the content of each question that makes up the test, and sends the output result to the server 20x as assignment information (step S210).

[0075] Server 20x receives assignment information from the generating AI 30 (step S211). Next, based on the assignment information, Server 20x creates a screen that displays the learning codes to be assigned to each problem in a format that is easy for the user to understand, and sends the screen information to the teacher terminal 10 (step S212). The teacher terminal 10 receives the screen information and displays a screen showing the learning codes to be assigned to each problem that makes up the test (step S213). Thus, the assignment information acquisition process is completed.

[0076] This allows server 20x to obtain the assigned information using generation AI 30 even if the file size of the learning code information is large and exceeds the upper limit of the input token limit of generation AI 30. Furthermore, output accuracy can be improved by extracting a portion of the learning code information related to the test and providing it to generation AI 30.

[0077] In this embodiment, learning materials are used as a test, but the present invention is not limited to this. By performing the same process on textbooks, reference books, supplementary materials, data collections, etc., the AI ​​30 can automatically assign learning codes, and the server 20x can acquire the assigned information.

[0078] Furthermore, in this embodiment, the subject of the test is identified using the generated AI 30, but the method of identifying the subject or field of the test is not limited to this. For example, when acquiring test information from the teacher's terminal 10, the subject of the test may be identified by having the user specify it through a predetermined operation.

[0079] Furthermore, in this embodiment, the identification of the test subject and the assignment of learning codes to each question are achieved by a single generating AI 30, but depending on cost and other factors, these may be achieved by different generating AIs.

[0080] <First variation> In the first embodiment described above, one learning code is assigned to one problem, but the present invention is not limited thereto, and multiple learning codes may be assigned to one problem. In this case, the server 20 obtains assignment information from the generating AI 30, including the relationship between the problem and the learning code assigned to the problem, by adding an instruction such as "Please output the degree of relevance between the problem and the learning code" to the assignment request prompt.

[0081] According to this, for example, if the relevance is obtained as a numerical value, the user can identify the learning code with the highest numerical value as having a higher level of confidence among the multiple learning codes assigned to a single problem. Furthermore, even if only one learning code is assigned to a single problem, the user can identify the learning code with the lowest numerical value as having a lower level of confidence by obtaining the assignment information, including the relevance.

[0082] <Second variation> In the first embodiment described above, test information and learning code information are attached to the grant request prompt, but model answers to each problem constituting the test may also be attached. This can improve the accuracy of the output results of the generated AI30.

[0083] <Third variation> In the first embodiment described above, for convenience, the server 20 is assumed to perform the assignment information acquisition process. However, there may be an application that performs the assignment information acquisition process, and the server 20 may be equipped with that application. In this case, the server 20 performs the assignment information acquisition process by executing the application.

[0084] Furthermore, in the first embodiment described above, the user uses a teacher terminal 10, but the present invention is not limited thereto, and the user may use a teacher terminal that has the functionality of a server 20. In this case, the teacher terminal can perform the assignment information acquisition process that was performed by the server 20 and acquire assignment information in which learning codes are assigned to each question that makes up the test.

[0085] Furthermore, the first to third modifications described above can also be applied to the second embodiment, similar to the first embodiment. [Explanation of Symbols]

[0086] 5 Network 10 Teacher's terminals 20, 20x servers 30 Generation AI 31 Learning Code DB 40 Test Information Acquisition Unit 41 Learning Code Information Acquisition Unit 42 Grant Request Prompt Creation Unit 43. Information Acquisition Unit 44 Screen creation section 100, 100x Learning Support System

Claims

1. A means for obtaining information about learning materials, A means for acquiring learning code information that acquires learning code information relating to learning codes for classifying the aforementioned learning materials, A means for creating an assignment request prompt that has the aforementioned teaching material information and the aforementioned learning code information and creates an assignment request prompt that requests a learning code to be assigned to the learning material, An information acquisition means that, by inputting the aforementioned assignment request prompt into the generating AI, acquires and outputs assignment information indicating the learning code to be assigned to the learning material, An information processing device equipped with the following features.

2. The learning material is a test, and the learning material information is information relating to the test, The grant request prompt creation means creates a grant request prompt that has the teaching material information and the learning code information and requests a learning code to be assigned to each question constituting the test, The information processing device according to claim 1, wherein the means for acquiring the assigned information acquires and outputs assigned information indicating each question constituting the test and a learning code to be assigned to each question.

3. The information processing apparatus according to claim 2, wherein the grant request prompt further comprises a model answer for each of the questions constituting the test.

4. A means for creating a specific request prompt that has the aforementioned teaching material information and requests the identification of the subject of the aforementioned test, The system includes a subject acquisition means that identifies and acquires the subject of the test by inputting the aforementioned specific request prompt into a generating AI, The learning code information acquisition means acquires learning code information corresponding to the specified subject, The information processing device according to claim 2, wherein the means for creating the assignment request prompt has the teaching material information and the learning code information corresponding to the subject, and requests a learning code to be assigned to each question constituting the test.

5. The aforementioned learning code consists of three classifications: major classification, medium classification, and minor classification. The grant request prompt includes the specification of the classification, The information processing apparatus according to claim 1, wherein the means for acquiring the assigned information acquires assigned information indicating a learning code for a specified classification.

6. The system further comprises a correlation calculation means that calculates and outputs the correlation between the aforementioned problem and the learning code assigned to the problem, The information processing apparatus according to claim 2, wherein the assigned information includes each problem, a learning code to be assigned to each problem, and the degree of relevance.

7. The information processing apparatus according to claim 2, further comprising a screen creation means for dividing each problem on the test based on the aforementioned assigned information, creating a screen that displays the problem and the learning code assigned to that problem in association, and outputting it.

8. A program executed by an information processing device equipped with a computer, Obtain information about learning materials, Obtain learning code information relating to the learning codes used to classify the aforementioned learning materials, Having the aforementioned teaching material information and the aforementioned learning code information, a grant request prompt is created that requests a learning code to be assigned to the learning material. A program that, by inputting the aforementioned assignment request prompt into the generating AI, causes the computer to execute a process that obtains and outputs assignment information indicating the learning code to be assigned to the learning material.

Images