Learning Support System

Abstract

A learning support system suitable for improving learning effectiveness is provided. SOLUTION: The learning assistance server 100 acquires summary information of the learning video and organizational information of the learner, inputs a request to the AI ​​model 50 that includes the acquired summary information and organizational information and includes a request to generate a discussion topic based on the learner's organizational information about the content of the learning video, and acquires a discussion topic that is output from the AI ​​model 50 in response to the request. This results in content that is based on the learner's knowledge and that resonates with the learner, thereby improving learning effectiveness compared to conventional methods.

Description

[Technical Field]

[0001] The present invention relates to a system for supporting a learner's learning, and more particularly to a learning support system suitable for improving learning effectiveness. [Background technology]

[0002] As a technology for supporting learners in their learning, for example, the technology described in Patent Document 1 is known.

[0003] The technology described in Patent Document 1 first inputs a prompt to GPT-3 that instructs it to create a summary from the transcript that includes words included in the answer candidate list obtained in S210, and generates a summary of the lecture content described for the answer candidates included in the answer candidate list. Then, each answer candidate is input to T5 along with its summary, and a system question is generated that will result in the input answer candidate being the correct answer. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2025-82333 A (

[0038] ) Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the invention described in Patent Document 1, although questions based on the content of the lecture can be generated, the content is not suited to the learner, and the learning effect is insufficient. This is because the learning effect is enhanced if the questions are based on the learner's knowledge and resonate with the learner, rather than simply being based on the content of the lecture.

[0006] Therefore, the present invention has been made in consideration of the unresolved problems of the conventional technology, and has an object to provide a learning support system suitable for improving learning effectiveness. [Means for solving the problem]

[0007] [Invention 1] In order to achieve the above object, the learning support system of Invention 1 is a learning support system that supports learners' learning by providing learning content, and is equipped with: content information acquisition means for acquiring content information related to the learning content or a summary thereof; profile information acquisition means for acquiring profile information of the learner; input means for inputting a request to an AI model, the request including the content information acquired by the content information acquisition means and the profile information acquired by the profile information acquisition means, and the request including a request to generate question information related to a question about the content of the learning content based on the learner's profile information; and question information acquisition means for acquiring question information output from the AI ​​model in response to the request.

[0008] With this configuration, the content information acquisition means acquires content information, the profile information acquisition means acquires profile information, and the input means inputs a request including the acquired content information and profile information and a request to generate question information to the AI ​​model, and the question information acquisition means acquires question information output from the AI ​​model in response to the request.

[0009] Here, the input means includes, for example, directly inputting a request to the AI ​​model, or indirectly inputting the request to the AI ​​model via a process, function, device, network, or other means. The same applies hereinafter to the learning support system of Invention 2.

[0010] Furthermore, the means for acquiring question information includes, for example, directly acquiring output information from the AI ​​model, or indirectly acquiring output information from the AI ​​model via processing, functions, devices, networks, or other means. The same applies hereinafter to the learning support system of Invention 2.

[0011] Furthermore, the request or the information contained therein may be configured in any format, such as vector data. The same applies to the learning support system of Invention 2 below.

[0012] Furthermore, the content information acquisition means may, for example, input content information from an input device or the like, acquire or receive content information from an external terminal or the like, read content information from a storage device or storage medium or the like, or generate or calculate content information by information processing or the like. Therefore, acquisition includes at least input, acquisition, reception, reading (including search), generation, and calculation. The same concept of acquisition applies hereinafter.

[0013] Furthermore, content information may be configured, for example, as the content or its summary itself, or as information for identifying the content or its summary (for example, link information such as a name, number, ID, code, or URL), or as feature information relating to statistics or other features of the content or its summary. Content information may also be configured, for example, as characters, numbers, figures, codes, symbols, images, sounds, or other information. Content information may also be configured as keywords relating to the content or its summary (for example, one or more keywords indicating part of the name of the content or its summary). The same applies hereinafter to the learning support system of Invention 2.

[0014] Furthermore, the question information may be, for example, the question itself, or may be configured as information for identifying the question (for example, link information such as a name, number, ID, code, or URL), or as feature information relating to a summary, statistics, or other features of the question. The question information may be configured as, for example, characters, numbers, figures, codes, symbols, images, sounds, or other information. The question information may be configured as keywords relating to the question (for example, one or more keywords indicating part of the name of the question). The same applies hereinafter to the learning support system of Invention 2.

[0015] Furthermore, this system may be realized as a single device, apparatus, terminal, or other device, or as a network system in which multiple devices, apparatus, terminals, or other devices are communicatively connected. In the latter case, each component may belong to any of the multiple devices as long as they are communicatively connected. The same applies to the learning support system of Invention 2 below.

[0016] [Invention 2] Furthermore, the learning support system of Invention 2 is a learning support system that supports learners' learning by providing learning content, and comprises: a registration means for registering content information about the learning content or a summary thereof and profile information about the learner in a knowledge base that can be referenced by an AI model; an input means for inputting a request to the AI ​​model, the input including a request to generate question information about a question based on the learner's profile information about the content of the learning content; and a question information acquisition means for acquiring question information output from the AI ​​model in response to the request by referring to the content information and profile information in the knowledge base.

[0017] With this configuration, the registration means registers the content information and the profile information in the knowledge base, the input means inputs a request including a request to generate question information to the AI ​​model, and the question information acquisition means acquires the question information output from the AI ​​model in response to the request.

[0018] [Invention 3] Furthermore, the learning support system of Invention 3 is a learning support system of either Invention 1 or 2, further comprising: a memo information acquisition means for acquiring the learner's memo information regarding the learning content; a second input means for inputting a request to the AI ​​model, the request including the memo information acquired by the memo information acquisition means and including a request to generate question information regarding a question based on the learner's memo information about the content of the learning content; and a second question information acquisition means for acquiring question information output from the AI ​​model in response to the request.

[0019] With this configuration, the memo information acquisition means acquires memo information of the learner, and the second input means inputs a request including the acquired memo information and a request to generate question information to the AI ​​model, and the second question information acquisition means acquires question information output from the AI ​​model in response to the request.

[0020] Here, the second input means includes, for example, inputting a request directly to the AI ​​model, or inputting it indirectly to the AI ​​model via a process, function, device, network, or other means.

[0021] In addition, the second question information acquisition means includes, for example, directly acquiring the output information of the AI ​​model, or indirectly acquiring the output information of the AI ​​model via processing, function, device, network, or other means.

[0022] Furthermore, memo information may be configured, for example, as the memo itself, or as information for identifying the memo (for example, link information such as a name, number, ID, code, or URL), or as characteristic information relating to statistics or other characteristics of the memo. Furthermore, memo information may be configured, for example, as characters, numbers, figures, codes, symbols, images, sounds, or other information. Furthermore, memo information may be configured as keywords relating to the memo (for example, one or more keywords indicating part of the name of the memo).

[0023] [Invention 4] Furthermore, the learning support system of Invention 4 is the learning support system of either Invention 1 or 2, and further comprises: summary information generation means for generating summary information of the learning content based on content information related to the learning content; first question information generation means for generating question information to the learner regarding the content of the learning content based on the summary information generated by the summary information generation means; second question information generation means for generating question information to the learner regarding the content of the learning content based on the summary information generated by the summary information generation means and the profile information; third question information acquisition means for acquiring question information from the learner; and answer information generation means for generating answer information to a question related to the question information based on the question information acquired by the third question information acquisition means, wherein the second question information generation means has the input means and the question information acquisition means.

[0024] With this configuration, the summary information generating means generates summary information of the study content based on the content information, and the first question information generating means generates question information based on the generated summary information.

[0025] Then, the second question information generating means generates question information based on the generated summary information and profile information. Specifically, the input means inputs a request including the acquired content information and profile information and a request to generate question information to the AI ​​model, and the question information acquiring means acquires question information output from the AI ​​model in response to the request. In this way, question information is generated.

[0026] Furthermore, the third question information acquiring means acquires question information from the learner, and the answer information generating means generates answer information based on the acquired question information. [Effects of the Invention]

[0027] As explained above, according to the learning support system of Invention 1 or 2, questions about the content of the learning content can be obtained based on the learner's profile information, so that the content is based on the learner's knowledge and resonates with the learner, thereby improving learning effectiveness compared to conventional systems.

[0028] Furthermore, according to the learning support system of Invention 3, questions about the content of the learning content can be obtained based on the learner's notes, so the content is based on the learner's knowledge and resonates with the learner, further improving the learning effect.

[0029] Furthermore, according to the learning support system of Invention 4, the learner can use the information obtained by the summary information generation means, the first question information generation means, the second question information generation means, and the answer information generation means for learning, thereby further improving the learning effect. [Brief explanation of the drawings]

[0030] [Figure 1] 1 is a block diagram showing a configuration of a network system according to an embodiment of the present invention; [Figure 2] FIG. 2 is a diagram illustrating a hardware configuration of a learning assistance server 100. [Figure 3] This figure shows the data structures of an account table 400, a video table 402, a keyword table 404, a question table 406, an organizational discussion table 408, a reference site table 410, a question and answer table 412, a memo table 414, and a review table 416. [Figure 4] FIG. 1 is a functional block diagram of a generation AI server 120. [Figure 5] 10 is a flowchart showing video information processing. [Figure 6] 10 is a flowchart showing a learning support process. [Figure 7] This is a screen that displays thumbnails of educational videos. [Figure 8] This is a screen that displays summaries and keywords of educational videos. [Figure 9] This screen displays questions from the educational video, organizational discussion topics, reference sites, and answers to questions for the AI ​​model 50. [Figure 10] This is the screen that displays questions from the educational video. [Figure 11] This is a screen displaying organizational discussion themes for educational videos. [Figure 12] This screen displays reference sites for educational videos and questions for AI model 50. [Figure 13] 10 is a flowchart showing a study note process. [Figure 14] This is a screen that displays study notes. DETAILED DESCRIPTION OF THE INVENTION

[0031] An embodiment of the present invention will be described below, with reference to Figures 1 to 14 showing the embodiment.

[0032] First, the configuration of this embodiment will be described. FIG. 1 is a block diagram showing the configuration of a network system according to this embodiment.

[0033] As shown in Figure 1, the Internet 199 is connected to a learning support server 100 that supports learners' learning by providing learning videos, a generation AI server 120 that generates answer information using an AI (Artificial Intelligence) model in response to requests, and multiple learner terminals 200 used by learners.

[0034] [Learning Support Server 100] Next, the configuration of the learning assistance server 100 will be described. FIG. 2 is a diagram showing the hardware configuration of the learning assistance server 100. As shown in FIG.

[0035] As shown in Figure 2, the learning assistance server 100 is composed of a CPU (Central Processing Unit) 30 that controls calculations and the entire system based on a control program, a ROM (Read Only Memory) 32 that stores the control program and other programs of the CPU 30 in advance in a specified area, a RAM (Random Access Memory) 34 that stores data read from the ROM 32 and other programs and calculation results required in the calculation process of the CPU 30, and an I / F (Interface) 38 that mediates the input and output of data to and from external devices.These components are connected to each other and capable of sending and receiving data via a bus 39, which is a signal line for transferring data.

[0036] The I / F 38 is connected to external devices such as an input device 40 consisting of a keyboard, mouse, etc. that can input data as a human interface, a memory device 42 that stores data, tables, etc. as files, a display device 44 that displays a screen based on an image signal, and a signal line for connecting to the Internet 199.

[0037] [Data Structure] Next, the data structure of the storage device 42 will be described. FIG. 3 shows the data structures of an account table 400, a video table 402, a keyword table 404, a question table 406, an organizational discussion table 408, a reference site table 410, a question and answer table 412, a memo table 414, and a review table 416.

[0038] The storage device 42 stores the video data of the learning video and its thumbnail data for each learning video.

[0039] As shown in FIG. 3, the storage device 42 stores an account table 400, a video table 402, a keyword table 404, a question table 406, an organizational discussion table 408, a reference site table 410, a question and answer table 412, a memo table 414, and a review table 416.

[0040] The account table 400 is a table for registering account information related to learner accounts. As shown in Fig. 3(a), the account table 400 is configured to have a column for registering a user ID for uniquely identifying a learner, a column for registering a password, a column for registering the name of the organization to which the user belongs, a column for registering the URL (Uniform Resource Locator) of the Internet site of the organization to which the user belongs, a column for registering information on the Internet site identified by the URL or its summary, statistics, or other characteristic information related to its characteristics as organization information, a column for registering the classification of the organization (e.g., company or school), and a column for registering other information. The user ID is the primary key.

[0041] The video table 402 is a table for registering video information related to learning videos. As shown in Figure 3(b), the video table 402 is configured to have a column for registering a video ID for uniquely identifying a learning video, a column for registering a path name of the video data of the learning video, a column for registering a path name of thumbnail data of the learning video, a column for registering the title name of the learning video, a column for registering a one-sentence summary of the learning video, a column for registering a bulleted summary of the learning video, a column for registering a full-sentence summary of the learning video, and a column for registering other information. The video ID is the primary key.

[0042] The keyword table 404 is a table for registering keyword information related to keywords of learning videos. As shown in Figure 3(c), the keyword table 404 is configured with a column for registering video IDs, a column for registering keyword IDs for uniquely identifying keywords of learning videos specified by the video IDs, a column for registering keywords, a column for registering keyword explanations, and a column for registering other information. The video ID and keyword ID are primary keys, and multiple keywords can be registered for one learning video.

[0043] The question table 406 is a table for registering question information related to questions posed to learners about the content of learning videos. As shown in FIG. 3(d), the question table 406 is configured with a column for registering video IDs, a column for registering question IDs for uniquely identifying questions in learning videos specified by the video IDs, a column for registering user IDs, a column for registering questions, a column for registering example answers to the questions, and a column for registering other information. The video ID, question ID, and user ID are primary keys, and multiple questions can be registered for each user for one learning video.

[0044] The organizational discussion table 408 is a table that registers organizational discussion theme information related to discussion themes for the content of learning videos based on the organizational information of the organization to which the learner belongs. As shown in FIG. 3(e), the organizational discussion table 408 is configured with a column for registering video IDs, a column for registering discussion theme IDs for uniquely identifying the discussion theme of the learning video specified by the video ID, a column for registering user IDs, a column for registering discussion themes, and a column for registering other information. The video ID, discussion theme ID, and user ID are primary keys, and multiple discussion themes can be registered for each user for one learning video.

[0045] The reference site table 410 is a table for registering reference site information about internet sites (hereinafter referred to as "reference sites") that serve as references for the content of learning videos. As shown in FIG. 3(f), the reference site table 410 is configured with a column for registering a video ID, a column for registering a reference site ID for uniquely identifying a reference site for a learning video specified by the video ID, a column for registering a user ID, a column for registering the name of the reference site, a column for registering the URL of the reference site, and a column for registering other information. The video ID, reference site ID, and user ID are primary keys, and multiple reference sites can be registered for each user for one learning video.

[0046] The question and answer table 412 is a table for registering questions to the AI ​​model 50 about the content of a learning video and question and answer information related to the answers to those questions. As shown in FIG. 3(g), the question and answer table 412 is configured to have a column for registering a video ID, a column for registering a question ID for uniquely identifying a question of a learning video specified by the video ID, a column for registering a user ID, a column for registering a question, a column for registering an answer to the question, and a column for registering other information. The video ID, question ID, and user ID are primary keys, and multiple questions and answers can be registered for each user for one learning video.

[0047] The memo table 414 is a table for registering learner's memo information regarding learning videos. As shown in FIG. 3(h), the memo table 414 is configured with a column for registering a video ID, a column for registering a memo ID for uniquely identifying a memo of a learning video specified by the video ID, a column for registering a user ID, a column for registering a memo, a column for registering a playback point of the learning video if a memo was registered while the learning video was being played, and a column for registering other information. The video ID, memo ID, and user ID are primary keys, and multiple notes can be registered per user for one learning video.

[0048] The review table 416 is a table for registering review information related to learning reviews based on the learner's memo information about the content of the learning video. As shown in FIG. 3(i), the review table 416 is configured with a column for registering a video ID, a column for registering a review ID for uniquely identifying a review of the learning video specified by the video ID, a column for registering a user ID, a column for registering reviews, and a column for registering other information. The video ID, review ID, and user ID are primary keys, and multiple reviews can be registered for each user for one learning video.

[0049] [Generation AI Server] Next, the configuration of the generation AI server 120 will be described. Like the learning assistance server 100, the generation AI server 120 has a hardware configuration similar to that of a general computer in which a CPU, ROM, RAM, I / F, etc. are connected by a bus, and is configured as, for example, a cloud server.

[0050] FIG. 4 is a functional block diagram of the generation AI server 120. As shown in Figure 4, the generation AI server 120 is configured to have multiple AI models 50, an AI model control unit 52 that controls the AI ​​models 50, and a knowledge base 54 that registers data that the AI ​​models 50 refer to for inference.

[0051] The AI ​​model 50 is an AI model trained on a large data set and is a highly versatile model capable of performing a variety of tasks. For example, a large language model can be used as the AI ​​model 50. A large language model is a deep learning model that pre-trains a language model, which models human-spoken language based on its occurrence probability, from a massive amount of data. When a prompt is input, the large language model statistically infers the probability of generating the next word from the sentence included in the input prompt and outputs the inference result. For example, known technologies described on the internet sites "https: / / chatgpt-lab.com / n / n418d3aa56f0b" and "https: / / agirobots.com / chatgpt-mechanism-and-problem / " can be used as the large language model. More specifically, for example, Titan Text G1 - Express, Titan Text G1 - Lite, Titan Image Generator G1, Titan Embeddings G1 - Text, Titan Embeddings Text V2, Titan Multimodal Embeddings G1, Claude, Claude Instant, Claude 3 Sonnet, Claude 3 Haiku, Claude 3 Opus, Jurassic-2 Mid, Jurassic-2 Ultra, Command, Command Light, Command R, Command R+, Embed English, Embed Multilingual, Llama 2 Chat 13B, Llama 2 Chat 70B, Llama 2 13B, Llama 2 70B, Llama 3 8b Instruct, Llama 3 70b Instruct, Mistral 7B Instruct, Mixtral 8X7B Instruct, Mistral Large, and Stable Diffusion XL can be adopted.

[0052] The AI ​​model control unit 52 selects one of the multiple AI models 50 to be used for inference in response to a selection request from the request processing unit 58. Furthermore, when a reference request is input from the request processing unit 58, the AI ​​model control unit 52 causes the selected AI model 50 (hereinafter referred to as the "selected AI model") to refer to the data in the knowledge base 54 in response to the input reference request. Furthermore, when a prompt is input from the request processing unit 58, the input prompt is input to the selected AI model. Then, when an execution request is input from the request processing unit 58, the AI ​​model control unit 52 causes the selected AI model to execute inference in response to the input execution request, obtains an inference result from the selected AI model, and outputs the obtained inference result to the request processing unit 58.

[0053] The knowledge base 54 can register reference information to be referenced for inference. The reference information in the knowledge base 54 is in a data format (for example, vector data) that can be referenced by the AI ​​model 50.

[0054] The generation AI server 120 is further configured to include a request receiving unit 56 that receives requests, a request processing unit 58 that processes the requests received by the request receiving unit 56, and an answer information sending unit 60 that sends answer information to the request received by the request receiving unit 56 to the learning assistance server 100.

[0055] The request receiving unit 56 receives a request for generating answer information from the learning assistance server 100 and outputs the received request to the request processing unit 58. The request includes (1) video data and other parameters, (2) a generation request for generating answer information, (3) a selection request for selecting an AI model 50, and (4) a reference request for referencing reference information in the knowledge base 54. (3) and (4) are not essential but are included additionally.

[0056] If the request received by the request receiving unit 56 includes a selection request or a reference request, the request processing unit 58 outputs the selection request or the reference request to the AI ​​model control unit 52. Furthermore, based on the request received by the request receiving unit 56, the request processing unit 58 generates a prompt that instructs the AI ​​model 50. The prompt, for example, requests the AI ​​model 50 to generate a summary of the learning video by referencing the video data in the knowledge base 54. The request processing unit 58 then outputs the generated prompt and execution request to the AI ​​model control unit 52, and if an inference result is input from the AI ​​model control unit 52 in response to the execution request, the request processing unit 58 outputs the input inference result to the answer information sending unit 60.

[0057] The answer information sending unit 60 sends answer information including the inference result input from the request processing unit 58 to the learning assistance server 100 .

[0058] The generation AI server 120 further comprises a request receiving unit 62 that receives requests, and a reference information registration unit 64 that registers reference information in the knowledge base 54.

[0059] The request receiving unit 62 receives a request for registering reference information from the learning assistance server 100, and outputs the received request to the reference information registration unit 64. The request includes (1) video data and a video ID.

[0060] The reference information registration unit 64 stores the video data included in the request received by the request receiving unit 62 in storage (not shown) and converts it into a data format (e.g., vector data) that can be referenced by the AI ​​model 50. Vector data can be generated using a technology (embedding) that converts data including text, images, audio, etc. into a numerical vector. The video data converted into vector data is then associated with a video ID and registered in the knowledge base 54. The AI ​​model control unit 52 causes the selected AI model to reference the video data in response to a reference request from the request processing unit 58.

[0061] [Student terminal 200] Next, the configuration of the learner terminal 200 will be described. The learner terminal 200 is configured by connecting a CPU, a ROM, a RAM, an I / F, etc. via a bus, similar to the learning support server 100. An input device, a storage device, a display device, etc. are connected to the I / F.

[0062] Next, the operation of this embodiment will be described. First, the operation when processing moving image information will be described.

[0063] FIG. 5 is a flowchart showing video information processing. The CPU 30 is comprised of an MPU (Micro-Processing Unit) or the like, and starts a predetermined program stored in a predetermined area of ​​the ROM 32, and executes the video information processing shown in the flowchart of Fig. 5 in accordance with the program. The video information processing is a process executed based on the account authority of the administrator, and when executed by the CPU 30, the process first proceeds to step S100 as shown in Fig. 5.

[0064] In step S100, based on the video table 402, the video data of the learning video selected for registration by the administrator (hereinafter referred to as the "target learning video" in the video information processing) is obtained from the storage device 42, and a request for registration of reference information is sent to the generation AI server 120. The request includes (1) the video data of the target learning video and its video ID.

[0065] Next, the process proceeds to step S102, where a training video summary generation process is executed. In the training video summary generation process, a request is sent to the generation AI server 120 to generate a one-sentence summary, a bulleted summary, and a full-sentence summary of the target training video. The request includes (1) the video ID of the target training video, (2) a generation request to generate a one-sentence summary, a bulleted summary, and a full-sentence summary of the target training video, (3) a selection request to select a specific AI model 50, and (4) a reference request to reference video data in the knowledge base 54 (video data corresponding to the video ID in (1)). When answer information is received from the generation AI server 120 in response to the request, the one-sentence summary, bulleted summary, and full-sentence summary included in the received answer information are registered in the video table 402.

[0066] Next, the process proceeds to step S104, where the keyword generation process is executed. In the keyword generation process, a request to generate keywords and their explanations for the target learning video is sent to the generation AI server 120. The request includes (1) the video ID of the target learning video, (2) a generation request to generate multiple keywords and their explanations for the target learning video, (3) a selection request to select a specific AI model 50, and (4) a reference request to reference video data in the knowledge base 54 (video data corresponding to the video ID in (1)). When answer information is received from the generation AI server 120 in response to the request, the multiple keywords and their explanations included in the received answer information are associated with the video ID of the target learning video and registered in the keyword table 404.

[0067] Next, the process proceeds to step S106, where the question generation process is executed. In the question generation process, a request is sent to the generation AI server 120 to generate questions to the learner regarding the content of the target learning video. The request includes (1) the video ID of the target learning video, (2) a generation request to generate multiple questions regarding the content of the target learning video, (3) a selection request to select a specific AI model 50, and (4) a reference request to reference video data in the knowledge base 54 (video data corresponding to the video ID in (1)). When answer information is received from the generation AI server 120 in response to the request, the multiple questions included in the received answer information are associated with the video ID of the target learning video and registered in the question table 406.

[0068] When the process of step S106 is completed, the series of processes ends. Next, the operation when providing learning support will be described.

[0069] FIG. 6 is a flowchart showing the learning support process. The CPU 30 starts a predetermined program stored in a predetermined area of ​​the ROM 32, and in accordance with the program, executes the learning support process shown in the flowchart of Fig. 6. The learning support process is executed based on the learner's account authority, and when executed by the CPU 30, the process first proceeds to step S200 as shown in Fig. 5.

[0070] Hereinafter, "display" refers to displaying on the learner terminal 200. Specifically, the learning assistance server 100 transmits display information to the learner terminal 200, and the learner terminal 200 displays information on a display device based on the received display information.

[0071] Figure 7 shows a screen that displays thumbnails of learning videos. In step S200, thumbnail data of the learning video selected by the learner for learning (hereinafter referred to as the "target learning video" in the learning support processing and study note processing) is obtained from storage device 42 based on video table 402, and the title name of the target learning video is obtained from video table 402. Then, a thumbnail of the target learning video is displayed based on the obtained thumbnail data, along with the obtained title name. The screen in Figure 7 displays thumbnail 500, title name 502, and button 504 to display study notes.

[0072] Figure 8 shows a screen that displays the summary and keywords of the learning video. Next, the process proceeds to step S202, where the one-sentence summary, itemized summary, and full-text summary of the target learning video are obtained from video table 402, and the obtained one-sentence summary, itemized summary, and full-text summary are displayed. The screen in Fig. 8 is the screen that is displayed by scrolling down the screen in Fig. 7, with the one-sentence summary displayed in display area 506, the itemized summary displayed in display area 508, and the full-text summary displayed in display area 510. Also displayed is a link 512 that switches between displaying and hiding the full-text summary.

[0073] Next, the process proceeds to step S204, where the keywords and their explanations for the target learning video are obtained from keyword table 404, and the obtained keywords and their explanations are displayed. On the screen of FIG. 8, the keyword "Environmentally Regenerative Agriculture" and its explanation "Environmentally Regenerative Agriculture...", the keyword "Carbon Farming" and its explanation "Carbon Farming...", the keyword "Slurry" and its explanation "Slurry...", the keyword "Digestate" and its explanation "Digestive Fluid...", and the keyword "No-Till Cultivation" and its explanation "No-Till Cultivation..." are displayed in display area 514. Also displayed is a link 516 for switching between displaying and hiding the keywords.

[0074] Figure 9 is a screen that displays questions from the learning video, organizational discussion topics, reference sites, and answers to questions to the AI ​​model 50.

[0075] Next, the process proceeds to step S206, where questions for the target learning video are retrieved from question table 406 and the retrieved questions are displayed. The screen in FIG. 9 is the screen that is displayed by scrolling down the screen in FIG. 8, and the questions "What is carbon farming...", "What is the role of cover crops...", and "Monitoring technology..." are displayed in display area 518. For each question, display area 518 displays button 520 for generating example answers, and button 522 for generating additional questions.

[0076] 9 also displays a display area 526 for displaying organizational discussion themes. In the display area 526, a button 528 for generating organizational discussion themes is displayed.

[0077] 9 also displays a display area 532 for displaying reference sites. In the display area 532, a button 534 for generating a reference site is displayed.

[0078] 9 also displays a display area 540 that displays answers to questions posed to the AI ​​model 50. Display area 540 displays a text box 542 for inputting questions to the AI ​​model 50, a button 544 for sending a request to the generation AI server 120, and a button 546 for resetting the conversation.

[0079] Next, the process proceeds to step S208, where it is determined whether playback of the target learning video has been requested by clicking on the thumbnail 500. If it is determined that playback of the target learning video has been requested (YES), the process proceeds to step S210, where video playback processing is executed. In the video playback processing, video data of the target learning video is obtained from the storage device 42 based on the video table 402, and the obtained video data is transmitted to the learner terminal 200 by streaming.

[0080] Figure 10 shows the screen displaying questions from the learning video. Next, the process proceeds to step S212, where it is determined whether the generation of an example answer to the question has been requested by clicking button 520. If it is determined that the generation of an example answer to the question has been requested (YES), the process proceeds to step S214, where the question / answer example generation process is executed. In the question / answer example generation process, the question corresponding to button 520 is obtained from the question table 406, and a request requesting the generation of an example answer to the question is sent to the generation AI server 120. The request includes (1) the obtained question, (2) a generation request to generate an example answer to the question, and (3) a selection request to select a specific AI model 50. The AI ​​model 50 generates an example answer to the question based on the question. When answer information is received from the generation AI server 120 in response to the request, the answer example included in the received answer information is registered in the question table 406 in association with the video ID of the target learning video and the learner's user ID. The answer example included in the received answer information is also displayed. On the screen of FIG. 10, when button 520 is clicked, an example answer is displayed in display area 524.

[0081] Next, the process proceeds to step S216, where it is determined whether or not the generation of additional questions has been requested by clicking button 522. If it is determined that the generation of additional questions has been requested (YES), the process proceeds to step S218, where the question generation process is executed. The question generation process is the same as step S106. The additional questions included in the answer information from the generation AI server 120 are registered in the question table 406 in association with the video ID of the target learning video and the learner's user ID. On the screen of FIG. 10, when button 522 is clicked, the additional questions are displayed below button 522, and a button 520 is displayed for each question.

[0082] Figure 11 shows a screen displaying the organizational discussion theme of the educational video. Next, the process proceeds to step S220, where it is determined whether the generation of an organizational discussion theme has been requested by clicking button 528. If it is determined that the generation of an organizational discussion theme has been requested (YES), the process proceeds to step S222, where the organizational discussion theme generation process is executed. In the organizational discussion theme generation process, the full-text summary of the target learning video is obtained from the video table 402, the learner's organizational information is obtained from the account table 400, and a request for the generation of a discussion theme is sent to the generation AI server 120. The request includes (1) the obtained full-text summary and organizational information, (2) a generation request to generate a discussion theme, and (3) a selection request to select a specific AI model 50. Based on the full-text summary of the target learning video and the organizational information, the AI ​​model 50 generates a discussion theme, which is a question to the learner about the content of the target learning video. For example, if the learner belongs to organization A, the AI ​​model 50 generates a discussion theme beneficial to organization A as a question to the learner about the content of the target learning video. If the learner belongs to organization B, the AI ​​model 50 generates a discussion theme beneficial to organization B, even if the target learning video is the same. When response information is received from the generation AI server 120 in response to the request, the discussion theme contained in the received response information is registered in the organizational discussion table 408 in association with the video ID of the target learning video and the learner's user ID. The discussion theme contained in the received response information is also displayed. In the screen of Figure 11, when button 528 is clicked in display area 526, the discussion themes "Theme 1...", "Theme 2...", "Theme 3..." are displayed in each display area 530.

[0083] FIG. 12 is a screen that displays reference sites for learning videos and questions to the AI ​​model 50.

[0084] Next, the process proceeds to step S224 to determine whether the generation of a reference site has been requested by clicking button 534. If it is determined that the generation of a reference site has been requested (YES), the process proceeds to step S226 to execute the reference site generation process. In the reference site generation process, a full-text summary of the target learning video is obtained from the video table 402, and a request for the generation of the name and URL of the reference site is sent to the generation AI server 120. The request includes (1) the obtained full-text summary, (2) a generation request for the generation of the name and URL of the reference site, and (3) a selection request for selecting a specific AI model 50. The AI ​​model 50 generates the name and URL of the reference site for the target learning video based on the full-text summary of the target learning video. When answer information is received from the generation AI server 120 in response to the request, the name and URL of the reference site included in the received answer information are associated with the video ID of the target learning video and the learner's user ID and registered in the reference site table 410. The name of the reference site included in the received answer information is also displayed. A link to the URL is set for the name. On the screen of FIG. 12, when a button 534 is clicked in a display area 536, reference sites "Ministry of Agriculture, Forestry and Fisheries..." and "Hokkaido University..." are displayed in each display area 536.

[0085] Next, the process proceeds to step S228, where it is determined whether the generation of an answer to the question has been requested by clicking button 544. If it is determined that the generation of an answer to the question has been requested (YES), the process proceeds to step S230, where the answer generation process is executed. In the answer generation process, a question to the AI ​​model 50 is obtained from the text box 542, and a request to generate an answer to the question is sent to the generation AI server 120. The request includes (1) the obtained question, (2) a generation request to generate an answer to the question, and (3) a selection request to select a specific AI model 50. The AI ​​model 50 generates the answer based on the question. When answer information is received from the generation AI server 120 in response to the request, the obtained question and the answer included in the received answer information are associated with the video ID of the target learning video and the learner's user ID and registered in the question-answer table 412. The answer included in the received answer information is also displayed. In the screen of FIG. 12, when the question “How to manufacture a slurry…” is entered in the text box 542 in the display area 540 and a button 544 is clicked, the answer “How to manufacture a slurry…” is displayed in the display area 548.

[0086] Next, the process proceeds to step S232, where it is determined whether or not the display of the study notes has been requested by clicking button 504. If it is determined that the display of the study notes has been requested (YES), the process proceeds to step S234, where the study note display process is executed, and the process proceeds to step S208.

[0087] On the other hand, if it is determined in step S232 that the display of the study notes is not requested (NO), the process proceeds to step S208.

[0088] On the other hand, if it is determined in step S228 that generation of an answer to the question is not requested (NO), the process proceeds to step S232.

[0089] On the other hand, if it is determined in step S224 that generation of a reference site is not requested (NO), the process proceeds to step S228.

[0090] On the other hand, if it is determined in step S220 that the generation of an organizational discussion topic is not required (NO), the process proceeds to step S224.

[0091] On the other hand, if it is determined in step S216 that additional generation of a question is not requested (NO), the process proceeds to step S220.

[0092] On the other hand, if it is determined in step S212 that generation of example answers to the question is not requested (NO), the process proceeds to step S216.

[0093] On the other hand, if it is determined in step S208 that playback of the target learning video is not requested (NO), the process proceeds to step S212.

[0094] Next, the operation when using the study notebook will be described. FIG. 13 is a flowchart showing the study note processing.

[0095] Figure 14 shows the screen that displays the study notes. The CPU 30 starts a predetermined program stored in a predetermined area of ​​the ROM 32, and in accordance with the program, executes the study notebook process shown in the flowchart of Fig. 13. The study notebook process is executed in step S234 based on the learner's account authority, and first proceeds to step S300 as shown in Fig. 13.

[0096] In step S300, the study notes screen is displayed in a pop-up window, as shown in Fig. 14(a). The screen in Fig. 14(a) displays a text box 550 for entering notes, a button 552 for saving notes, a display area 554 for displaying saved notes and the playback time points of the target study video in association with each other, and a button 556 for starting a review.

[0097] Next, the process proceeds to step S302, where it is determined whether a memo has been entered by clicking button 552. If it is determined that a memo has been entered (YES), the process proceeds to step S304, where a memo registration process is executed. In the memo registration process, the memo is obtained from text box 550, the current playback time of the target learning video is obtained, and the obtained memo and playback time are associated with the video ID of the target learning video and the learner's user ID and registered in memo table 414. For example, if "carbon farming" is registered as a memo 1 minute 33 seconds after playback begins, the memo "carbon farming" and playback time "01:33" are registered in memo table 414, and "01:33 carbon farming" is displayed in display area 554.

[0098] Next, the process proceeds to step S306, where it is determined whether the generation of a review has been requested by clicking button 556. If it is determined that the generation of a review has been requested (YES), the process proceeds to step S308, where the review generation process is executed. In the review generation process, a full-text summary of the target learning video is obtained from video table 402, notes registered by the learner about the target learning video are obtained from note table 414, and a request for generating a review is sent to generation AI server 120. The request includes (1) the obtained full-text summary and one or more notes, (2) a generation request to generate a review, and (3) a selection request to select a specific AI model 50. Based on the full-text summary of the target learning video and one or more notes, the AI ​​model 50 generates a review, which is a question to the learner to reflect on the content of the target learning video. For example, if the note "carbon farming" is registered, the AI ​​model 50 generates a review related to the note "carbon farming" as a question to the learner about the content of the target learning video. If two notes, "Carbon Farming" and "Environmental Regeneration," are registered, a review related to the notes, "Carbon Farming" and "Environmental Regeneration," is generated. When response information is received from the generation AI server 120 in response to the request, the review included in the received response information is associated with the video ID of the target learning video and the learner's user ID and registered in the review table 416. The review included in the received response information is also displayed. On the screen of Figure 14(b), when button 556 is clicked after the notes, "Carbon Farming" and "Environmental Regeneration," have been registered, the review "Carbon is..." is displayed in the display area 558.

[0099] Below the display area 558, a text box 560 for inputting a question to the AI ​​model 50 and a button 562 for sending a request to the generation AI server 120 are displayed. The operation of the text box 560 and the button 562 is the same as in steps S228 and S230. The request may include a review of the display area 558.

[0100] Next, the process proceeds to step S310, where it is determined whether or not the end of the study notes has been requested by clicking the "X" button in the pop-up window. If it is determined that the end of the study notes has been requested (YES), the process ends and returns to the original process.

[0101] On the other hand, if it is determined in step S310 that the end of the study notes is not requested (NO), the process proceeds to step S302.

[0102] On the other hand, if it is determined in step S306 that the generation of a review is not requested (NO), the process proceeds to step S310.

[0103] On the other hand, if it is determined in step S302 that no memo has been input (NO), the process proceeds to step S306.

[0104] Next, the effects of this embodiment will be described. In this embodiment, the learning assistance server 100 acquires summary information of the learning video and organizational information of the learner, inputs a request to the AI ​​model 50 that includes the acquired summary information and organizational information and includes a request to generate a discussion topic based on the organizational information of the learner regarding the content of the learning video, and acquires a discussion topic output from the AI ​​model 50 in response to the request.

[0105] This allows the content to be based on the learner's knowledge and to resonate with the learner, thereby improving the learning effect compared to the past.

[0106] Furthermore, in this embodiment, the learning assistance server 100 acquires summary information of the learning video and the learner's note information, inputs a request to the AI ​​model 50 that includes the acquired note information and a request to generate a reflection on the content of the learning video based on the learner's note information, and acquires the reflection output from the AI ​​model 50 in response to the request.

[0107] This allows the content to be based on the learner's knowledge and to resonate with the learner, further improving the learning effect.

[0108] Furthermore, in this embodiment, the learning assistance server 100 includes: (1) a process for generating summary information of the learning video based on the video data of the learning video (step S102); (2) a process for generating keywords for the learning video based on the summary information of the learning video (step S104); (3) a process for generating questions based on the summary information of the learning video (step S106); (4) a process for generating discussion topics based on the summary information of the learning video and organizational information of the learner (step S222); (5) a process for generating reference links for the learning video based on the summary information of the learning video (step S226); (6) a process for acquiring questions from the learner and generating answers to the acquired questions (step S230); and a process for generating a review based on the summary information of the learning video and the learner's note information (step S308).

[0109] This allows the learner to use the information obtained through each process in their studies, further improving the effectiveness of their studies.

[0110] In this embodiment, step S102 corresponds to the summary information generation means of invention 4, step S106 corresponds to the first question information generation means of invention 4, step S222 corresponds to the content information acquisition means of invention 1, the profile information acquisition means of invention 1, the input means of invention 1 or 4, the question information acquisition means of invention 1 or 4, or the second question information generation means of invention 4. Also, step S230 corresponds to the third question information acquisition means of invention 4 or the answer information generation means of invention 4, and step S308 corresponds to the memo information acquisition means of invention 3, the second input means of invention 3, or the second question information acquisition means of invention 3.

[0111] In this embodiment, the learning video corresponds to the learning content of the first, third or fourth invention, and the organization information corresponds to the profile information of the first or fourth invention.

[0112] [Modification] In the above embodiment and its variations, the processing of steps S104, S106, S222, S226, and S308 was generated based on summary information of the learning video, but this is not limited to this, and a configuration can be adopted in which generation is performed based on video data in the storage device 42 or knowledge base 54.

[0113] Furthermore, in the above embodiment and its variations, video data is registered in the knowledge base 54 and used, but this is not limited to this, and video data posted on external video distribution sites can also be used as video data for learning videos.

[0114] Furthermore, in the above embodiment and its modifications, the generation AI server 120 is used, but this is not limiting, and a configuration can be adopted in which an AI model is constructed in the storage device 42 and the AI ​​model in the storage device 42 is used. For example, the following configuration can be adopted for the processing of steps S222 and S308. The processing of steps S104, S106, and S226 can also be configured in a similar manner.

[0115] The first configuration acquires characteristic information of the learning video and organizational information of the learner, and estimates the discussion topic from the acquired characteristic information and organizational information using a trained model trained based on learning data including the characteristic information of the learning video, organizational information of the learner, and the discussion topic.

[0116] The second configuration acquires feature information of the learning video and learner's note information, and estimates reflections from the acquired feature information and note information using a trained model trained based on learning data including the feature information of the learning video, learner's note information, and discussion topics.

[0117] Furthermore, in the above embodiment and its variations, discussion topics are generated based on organizational information about the organization to which the learner belongs, but this is not limited to this. Discussion topics and other information can also be generated based on profile information about the learner's other profiles (e.g., name, user ID, password, email address, department, study group, position, job title, department, employee number, student number, authority level, course history, progress, grades, last access date).

[0118] In the above embodiment and its variations, the processing in step S220 generates a discussion topic based on one piece of organizational information, but this is not limiting and a configuration in which a discussion topic is generated based on multiple pieces of organizational information can be adopted. For example, if a learner belongs to multiple organizations A and B, a discussion topic that is beneficial to both organization A and organization B, or a discussion topic that is beneficial to either organization A or organization B, is generated.

[0119] In the above embodiment and its variations, the processing in step S220 uses information on an Internet site identified by the URL of the organization as organizational information, but this is not limiting and the URL can also be used as organizational information. In this case, for example, the AI ​​model 50 can perform Retrieval-Augmented Generation (RAG), which collects necessary information from the URL and generates discussion topics based on the collected information.

[0120] Furthermore, in the above embodiment and its variations, the processing in step S308 generates a review based on the summary information of the learning video and the learner's note information, but this is not limited to this, and a configuration can be adopted in which a review is generated based on the learner's note information.

[0121] Furthermore, in the above embodiment and its variations, the processing of step S308 generates a reflection based on multiple notes when multiple notes are registered, but this is not limited to this, and a configuration can be adopted in which a reflection is generated based on a portion of the multiple notes (one or multiple notes).

[0122] Furthermore, in the above embodiment and its modified examples, the generation AI server 120 is configured with the functions 56 to 64 integrated as in the above embodiment, but this is not limited to this, and some functions can be configured by separate servers, etc. The same applies to the learning assistance server 100.

[0123] Furthermore, in the above embodiment and its modifications, the system is realized as a network system, but the present invention is not limited to this and can be realized as a single device or application.

[0124] Furthermore, in the above embodiment and its modifications, the case where the present invention is applied to a network system consisting of the Internet 199 has been described, but the present invention is not limited to this, and may be applied, for example, to a so-called intranet that communicates using the same method as the Internet 199. Of course, the present invention is not limited to a network that communicates using the same method as the Internet 199, and may be applied to a network of any communication method.

[0125] Furthermore, in the above embodiment and its modifications, the learning assistance server 100 is configured to use the storage device 42, but this is not limiting and the learning assistance server 100 can also be configured to use an external storage device such as a database server.

[0126] Furthermore, in the above-described embodiment and its variations, the processes shown in the flowcharts of Figures 5, 6 and 12 are executed by executing a program that is pre-stored in ROM 32. However, this is not limiting, and the program showing these procedures may be read into RAM 34 from a storage medium on which the program is stored and executed.

[0127] In the above embodiment and its modifications, the present invention is applied to a case where a learner's learning is supported by providing learning videos, but the present invention is not limited to this and can be applied to other cases without departing from the spirit of the present invention. For example, the following configuration can be adopted.

[0128] The first configuration is a configuration in which the AI ​​model 50 generates questions and a predetermined number of options (including correct and incorrect options) regarding the learning video based on the video data or summary information of the learning video, displays the generated questions and options, allows the user to select an answer option, and determines whether the answer is correct or incorrect based on the selection result.

[0129] The second configuration is the first configuration in which the reinforcement learning agent analyzes the learner's behavior log and determines the optimal problem to be generated next.

[0130] The third configuration is a configuration in which the AI ​​model 50 generates questions in accordance with Bloom's taxonomy (understand → apply → evaluate → create) in the first configuration, which can raise the learner's cognitive level.

[0131] The fourth configuration employs LCA (Learning Contents Automation). LCA is a concept that software-izes the process of creating learning materials. First, in the input stage, diverse sources—text, slides, existing videos, and internal knowledge bases—are uploaded in natural language. An AI model performs structural analysis and converts them into metadata tagged with topic, difficulty, and media attributes. Next, a logical chapter structure is automatically designed by referencing a knowledge graph, and a three-layered sequence of "essential → applied → advanced" is generated in conjunction with the learning objectives. A speech synthesis engine outputs the narration script with tone, speed, and emotional parameters. Images and B-roll video are automatically extracted and retimed in sync with this from stock materials or an internal digital asset management system. Subtitles are generated in multiple languages ​​via an automatic translation API (Application Programming Interface) and authored into a package compliant with SCORM (Sharable Content Object Reference Model) and xAPI (Experience API). Furthermore, AI model 50 detects key phrases scattered throughout the learning materials and generates knowledge cards, glossaries, and reference links in parallel, allowing learners to jump to additional information in a pop-up format while watching the video. After release, the system learns quality indicators (viewing completion rate, playback speed change rate, error reporting, etc.) based on viewing logs and quiz results, creating a "self-improvement loop" that automatically adjusts the scenario and length the next time the video is rebuilt. While a process that previously required several weeks and involved multiple professionals, including project managers, scriptwriters, video editors, and LMS (Learning Management System) implementers, LCA can be completed in minutes to hours and continues to evolve even after release. As a result, the system achieves (1) significant reductions in production costs, (2) shorter update cycles, and (3) a new knowledge supply chain in which learning data and learning material revisions circulate in real time.

[0132] The fifth configuration employs Personal Learning Optimization (PLO). PLO is an algorithmic system that "rewrites the learning process to suit individual needs." The learner's profile (job type, years of experience, key performance indicators (KPIs), and areas of interest) collected during registration and behavioral logs (playback operations, rewind position, speech content, and quiz answer patterns) are vectorized and used to estimate comprehension using Bayesian inference. Furthermore, a reinforcement learning agent calculates the "expected gradient toward goal achievement" and determines the next learning piece and question to present in real time. Questions are generated according to Bloom's taxonomy and are designed to automatically progress through the stages of memory, comprehension, application, evaluation, and creation. When the learner types or responds via voice, the interactive AI responds with personalized, probing questions, supplemental case studies, and comparative analysis, providing metacognitive feedback rather than simple "correct / incorrect" judgments. During the evaluation phase, AI Model 50 dynamically creates a scoring rubric and automatically issues a digital badge or internal qualification if the criteria set by the administrator are met. The performance data generated in this way can be linked via API to the HR information system and internal talent management system, allowing it to be reflected in promotion requirements and project assignments. The introduction of PLO has been confirmed to have three effects: (1) improved knowledge retention rate per learning hour, (2) reduced dropout rates due to enhanced self-efficacy, and (3) improved quality of personnel evaluation data. Ultimately, AI Model 50 visualizes the skill portfolio of the entire organization, developing it into a "learning intelligence platform" that quantitatively optimizes reskilling investments.

[0133] The sixth configuration is a cloud-based multi-tenant design that allows for instant horizontal deployment across entire corporate groups and university federations while complying with security infrastructure (SAML (Security Assertion Markup Language) / OIDC (OpenID Connect), CSR (Corporate Social Responsibility) audit logs) and local regulations (GDPR (General Data Protection Regulation), Personal Information Protection Act). It automates training integration across multiple corporations and localization for overseas bases via API, achieving centralized optimization of global HR-D.

[0134] The seventh configuration is one in which the teaching materials generated by the LCA are fed to the PLO via xAPI / Caliper, and the PLO analysis results again update the LCA generation conditions, forming a "self-evolving loop."

[0135] The eighth configuration is a configuration in which, instead of a learning video, learning images, text, audio, or a combination of these are provided. [Explanation of symbols]

[0136] 100...learning support server, 30...CPU, 32...ROM, 34...RAM, 38...I / F, 39...bus, 40...input device, 42...storage device, 44...display device, 120...generation AI server, 50...AI model, 52...AI model control unit, 54...knowledge base, 56, 62...request receiving unit, 58...request processing unit, 60...answer information sending unit, 64...reference information registration unit, 200...learner terminal, 199...Internet, 400...account table, 402...video table, 404...keyword table, 406...quest table, 408...organizational discussion table, 410...reference site table, 412...quest and answer table, 414...memo table, 416...reflection table, 500...thumbnail, 502...title name, 504, 520, 522, 528, 534, 544, 546, 552, 556, 562... Buttons; 506, 508, 510, 514, 518, 524, 526, 530, 532, 536, 540, 548, 554, 558... Display areas; 512, 516... Links; 542, 550, 560... Text boxes

Claims

1. A learning support system that supports a learner's learning by providing learning content, a summary information generating means for generating summary information of the study content based on content information relating to the study content; a first question generation means for generating question information to be posed to the learner regarding the content of the study content based on the summary information generated by the summary information generation means; a second question generation means for generating question information for the learner regarding the content of the study content based on the summary information generated by the summary information generation means and the learner's profile information; a reference site information generating means for generating reference site information about Internet sites that are useful for the content of the study content based on the summary information generated by the summary information generating means; The reference site information generating means a first input means for inputting a request including the summary information generated by the summary information generating means and a request for generating the reference site information to an AI model; A learning support system characterized by comprising a reference site information acquisition means for acquiring reference site information output from the AI ​​model in response to the request.

2. In claim 1, the learner's profile information is organizational information related to an organization to which the learner belongs; A learning support system characterized in that the second question information generation means generates organizational discussion topic information regarding discussion topics based on the learner's organizational information regarding the content of the learning content based on the summary information and the organizational information.

3. In claim 2, The second query information generating means a second input means for inputting a request including the summary information generated by the summary information generating means and the organizational information, and including a request for generating the organizational discussion topic information, into the AI ​​model; A learning support system characterized by comprising an organizational discussion topic information acquisition means for acquiring organizational discussion topic information output from the AI ​​model in response to the request.

4. In any one of claims 1 to 3, a question information acquisition means for acquiring question information from the learner; a response information generating means for generating response information to a question related to the question information based on the question information acquired by the question information acquiring means;

5. In any one of claims 1 to 3, a memo information acquisition means for acquiring memo information of the learner regarding the study content; A learning support system characterized by comprising a third question information generation means for generating question information regarding a question based on the learner's note information about the content of the learning content based on the note information acquired by the note information acquisition means.

Images