Support device, support method, and support program

Abstract

This enables organizations to appropriately support user education. [Solution] The support device 100 stores organizational education data, which is data used for educating users belonging to an organization, in a storage unit 120 for each organization. The support device 100 processes the organizational education data stored in the storage unit 120 for each organization into predetermined tasks used for learning about the organization by users. The support device 100 outputs the processed predetermined tasks to the user.

Description

Technical Field

[0001] The present invention relates to a support device, a support method, and a support program.

Background Art

[0002] Education may be provided to personnel belonging to an organization (hereinafter sometimes referred to as "users") so that they can take appropriate actions within the organization. The materials used in the above education are managed for each educational purpose and content, and are used when a user receives education related to the organization. Then, the user conducts self-study using the materials and content.

[0003] In recent years, when managing materials and content used in education, data management technologies using on-premises or cloud-based server devices are used. For example,a prior art is known in which an application for managing data is constructed without coding, and the constructed application is used to manage materials and content (see, for example, Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the prior art has problems in appropriately supporting user education in an organization. For example, the prior art can create applications for workflow, data management, etc., but has problems in generating problems for using organization-specific educational materials and content in user education and supporting user education. [Means for solving the problem]

[0006] Therefore, in order to solve the above problems and achieve the objective, the support device of the present invention is characterized by comprising: a storage unit that stores the education data, which is data used for educating users belonging to an organization, in a storage unit for each organization; a processing unit that processes the education data of each organization stored in the storage unit into predetermined tasks that the user uses for learning; and an output unit that outputs the processed predetermined tasks to the user. [Effects of the Invention]

[0007] This invention has the effect of enabling appropriate support for user education within an organization. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a diagram illustrating an overview of user education support according to the embodiment. [Figure 2] Figure 2 shows an example of the configuration of the support device according to the embodiment. [Figure 3] Figure 3 is a table diagram showing an example of teaching material information according to the embodiment. [Figure 4] Figure 4 is a table diagram showing an example of user learning information according to the embodiment. [Figure 5] Figure 5 shows an example of a reception screen according to the embodiment. [Figure 6] Figure 6 shows an example of a screen for receiving educational data according to this embodiment. [Figure 7] Figure 7 shows a series of steps in the generation process and output process according to the embodiment. [Figure 8] Figure 8 shows an example of a prompt related to the first example. [Figure 9] Figure 9 shows an example of the output of learning support information in chat format related to the first example. [Figure 10]Figure 10 shows an example of a prompt related to the second example. [Figure 11] Figure 11 shows an example of a prompt related to the second example. [Figure 12] Figure 12 shows an example of learning support information output in chat format related to the second example. [Figure 13] Figure 13 shows an example of a prompt related to the third example. [Figure 14] Figure 14 shows an example of learning support information output in chat format related to the third example. [Figure 15] Figure 15 is a flowchart showing the processing of the support device according to this embodiment. [Figure 16] Figure 16 shows an example of the hardware configuration of a computer that implements the support device according to the embodiment. [Modes for carrying out the invention]

[0009] Hereinafter, embodiments for carrying out the present invention (hereinafter referred to as "embodiments") will be described with reference to the drawings. However, each embodiment is not limited to those described below.

[0010] <Introduction> (background) Traditionally, organizations have provided training to ensure that users can behave appropriately within the organization. This training can be conducted through classroom lectures by the organization's training staff or on-the-job training (OJT) by work instructors. In addition, users may also learn independently using training materials managed for organizational purposes and content.

[0011] As a technology for realizing learning by the user himself / herself, a technology for managing educational materials for each educational purpose or content is known. For example, reference technologies for managing materials and content used for the education of users belonging to an organization using an on-premises or cloud-type server device are known. The above reference technologies can create various applications without coding and manage target data using the applications.

[0012] However, it is difficult to process the data of educational materials managed for each organization to generate predetermined tasks (hereinafter, may be simply referred to as "tasks") such as educational videos for user education and exercise problems. Therefore, it may be difficult to appropriately support user education with the reference technologies.

[0013] (Processing by Support Device 100) Therefore, the support device 100 according to the present embodiment outputs a "task" generated using data for the education of users belonging to an organization, which is data related to the education of the organization managed for each organization (hereinafter, may be referred to as "educational data"), to the educational target users, thereby realizing appropriate support for the education of the organization for the users.

[0014] Here, the overall picture of the processing by the support device 100 will be described. FIG. 1 is a diagram showing an overview of user education support according to the embodiment. The support device 100 shown in FIG. 1 is an example of a computer that provides a technology for realizing the information processing described below.

[0015] First, as shown in (1-1) to (1-3) of FIG. 1, the support device 100 receives the educational data for each organization input by the education staff of each organization and stores it in the storage unit 120.

[0016] The support device 100 processes the educational data for each organization stored in the memory unit 120 into "tasks" that the user 10 can use to learn about the organization (Figure 1 (2-1)). For example, the support device 100 processes the educational data for each organization to generate educational videos (Figure 1 (2-2)) and practice problems (not shown in Figure 1), etc.

[0017] The support device 100 outputs the generated tasks to the user 10. The user 10 then performs tasks such as watching educational videos and solving practice problems output from the support device 100.

[0018] Next, the support device 100 inputs the user 10's learning progress using the task (Figure 1 (3-1)) into a large-scale language model as prior knowledge (Figure 1 (3-2)). Then, based on the user 10's learning progress input as prior knowledge, the support device 100 performs teaching and coaching for the user 10, answers questions from the user 10, and presents problems.

[0019] For example, the support device 100 creates a learning plan to help user 10 learn efficiently, manages the progress of learning based on that plan, and provides encouragement and support to user 10 (as shown in Figure 1 (4-1), "coaching"). The support device 100 also provides practice problems tailored to user 10's academic level and learning progress, and follows up on user 10's answers to the practice problems (as shown in Figure 1 (4-2), "teaching"). The support device 100 also outputs educational data for each organization and answers tailored to user 10's learning status in response to "questions" (as shown in Figure 1 (4-3)) input by user 10 (as shown in Figure 1 (4-4), "answers").

[0020] Furthermore, the support device 100 can implement the above-mentioned "coaching (Figure 1 (4-1))", "teaching (Figure 1 (4-2))", and "answering (Figure 1 (4-4))" through a natural language dialogue format (chat format), as shown in Figure 1 (5).

[0021] In this way, the support device 100 according to this embodiment has the effect of appropriately supporting user education within an organization by generating assignments based on educational data for each organization, and providing teaching and coaching to support user learning using said assignments, answering questions from users, etc.

[0022] <Explanation of support device 100> (Support device 100) Next, the support device 100 according to this embodiment will be described in detail. Figure 2 is a diagram showing an example of the configuration of the support device 100 according to this embodiment. As shown in Figure 2, the support device 100 according to the first embodiment has a communication unit 110, a storage unit 120, and a control unit 130.

[0023] Although not shown in Figure 2, the support device 100 may be equipped with an input unit such as a keyboard or mouse to receive input from an administrator or other user. Furthermore, the support device 100 may be equipped with a display unit such as a screen to show the administrator educational material information, user learning information, etc., stored in the storage unit 120.

[0024] (Communications Department 110) The communications unit 110 performs data communication related to the input of information concerning user learning (hereinafter sometimes referred to as "user learning information"), including educational data for each organization, information such as the learning status of users who learn using the educational data, information such as conversations related to coaching and teaching, answers to assignments, and questions entered by users, and output of generated information that supports user learning (hereinafter sometimes referred to as "learning support information") to terminal devices, etc.

[0025] The communication unit 110 is implemented, for example, by a NIC (Network Interface Card) or a network interface controller. The communication unit 110 is connected to a network (e.g., the Internet) by wire or wireless connection. The communication unit 110 then sends and receives information to and from external devices via the network.

[0026] The communication unit 110 may also transmit and receive information using any communication standard or technology, such as Wi-Fi (registered trademark), Bluetooth (registered trademark), SIM (Subscriber Identity Module), or LPWA (Low Power Wide Area).

[0027] (Storage unit 120) The storage unit 120 stores data and programs used for various processes by the control unit 130, as well as various data acquired through the operation of the control unit 130. The storage unit 120 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or storage devices such as hard disks and optical discs. As shown in Figure 2, the storage unit 120 also includes a teaching material information DB 121, a user learning information DB 122, and a generation model DB 123.

[0028] (Teaching material information DB121) The Educational Materials Information DB121 is a database that stores educational data for each organization and assignments, etc. (educational materials information) generated using that educational data. Specifically, the Educational Materials Information DB121 stores educational materials information such as information that identifies an organization (organization identification information), titles of educational data etc. managed by each organization (titles), link information for accessing educational data stored in a designated storage device (links), information indicating the display order when displaying a list of educational data (order), information indicating the number of contents of a predetermined granularity within the educational data (number of contents), information regarding assignments associated with the educational data (assignments), and information indicating the publication type of the educational materials (type).

[0029] Here, we will explain an example of teaching material information stored in the teaching material information DB121. Figure 3 is a table diagram showing an example of teaching material information according to the embodiment. The teaching material information DB121 stores items such as "organization identification information," "title," "link," "order," "number of content items," "assignment," and "type," as well as information related to those items, in a table format or the like, associated with "No," which is information that identifies individual teaching material information.

[0030] For example, as shown in Figure 3, the teaching material information DB121 stores the organization identification information "1" (information identifying "Organization 1"), identified by No. "1", along with the title "A", link "B", order "C", number of content items "D", task "E", and type "F". The letters "A through F" above are legend information for each item of the individual teaching material information stored in the teaching material information DB121, and the actual information stored is not particularly limited.

[0031] Furthermore, using the teaching material information stored in the teaching material information DB121, the support device 100 can display the information to users and administrators for each organization in a list format that includes items such as "type," "thumbnail," "title," "sequence," "chapter," "section," and "assignment," as shown in Figure 3(1). For example, the support device 100 can display teaching material information at a granularity such as "Company A" and "Company B," or at a granularity such as "sales organization" and "research and development organization" contained within a "company."

[0032] The "Title" above is information that identifies educational data used by users for learning, and includes information such as the title, volume number, and issue number of the educational data expressed in natural language such as text, numbers, and symbols. The "Link" is information for accessing the storage medium that stores the educational data itself or thumbnail information of the educational data, and includes information such as the URL (Uniform Resource Locator) associated with each piece of educational data. The "Order" is information used to define the display order when the educational data is shown in a list.

[0033] "Number of Content Items" refers to the number of content items grouped at a predetermined level of granularity within the educational data, such as "number of chapters" or "number of sections." "Assignments" refers to information about assignments such as educational videos processed from the educational data or practice problems given to users, and includes information such as the content and correct answers of the practice problems, and the number of practice problems. "Type" refers to the status of the educational data's availability, such as "public," which is available to all registered users, or "limited," which is available only to specific users.

[0034] (User learning information DB122) The User Learning Information DB122 is a database that stores information related to user learning (user learning information), including information such as the learning status of users who are learning using the tasks. Specifically, the User Learning Information DB122 stores user learning information such as information that identifies the user (user identification information), tasks that the user uses for learning (learning materials) generated from educational data, the user's learning history (learning history), the user's learning goals (learning goals), and the history of conversations between the user and the support device 100 (conversation history).

[0035] Here, we will explain an example of user learning information stored in the user learning information DB122. Figure 4 is a table diagram showing an example of user learning information according to the embodiment. The user learning information DB122 stores items such as "user identification information," "materials used," "learning history," "learning objectives," and "conversation history," as well as information related to those items, in a table format or the like, associated with "No," which is information that identifies individual user learning information.

[0036] For example, as shown in Figure 4, the user learning information DB122 stores the user identification information "User 10," identified by No. "1," the learning materials used "G," the learning history "H," the learning objectives "I," and the conversation history "J," each associated with the others. The letters "G through J" above are legends for each item of individual user learning information stored in the user learning information DB122, and the actual information stored is not particularly limited.

[0037] Furthermore, using the user learning information stored in the user learning information DB122, the support device 100 can display the information to users and administrators in a list format that includes items such as "chapter," "section," "task," "type," "correct answer rate," and "result," as shown in Figure 4 (1).

[0038] The "user identification information" mentioned above refers to information that identifies an individual user who uses the assignments to learn about the organization. This includes, for example, the user's name or nickname, telephone number, service membership number, and other user-specific identification information belonging to the organization. The user learning information DB122 can store user identification information while protecting personal information by deleting or replacing information that could identify an individual user with masking or other methods based on publicly known technologies.

[0039] "Materials Used" is information used to identify the tasks that the user uses for learning, and includes, for example, information that identifies the "title" of the material information stored in the material information DB121. "Learning History" is history information of how the user has learned using the tasks, and includes, for example, information on the learning date, learning time, and learning progress. "Learning Objectives" is information on learning objectives set by the user who is learning using educational data and tasks generated based on said educational data, and includes, for example, objectives such as "Complete learning content XX by XX / XX" or "Study for XX hours per week." "Conversation History" is history information (past conversation data) of natural language conversations between the learning user and the support device 100.

[0040] (Generative model DB123) The generative model DB123 is a database that stores predetermined generative models used to generate learning support information by the generation unit 134, described later. For example, the generative model DB123 can store large-scale language models as generative models. Specifically, the support device 100 can use "ChatGPT®," a large-scale language model with general-purpose knowledge, as the large-scale language model (see, for example, Reference 1).

[0041] (Reference 1):ChatGPT(OpenAI),<URL:https: / / openai.com / chatgpt> ,<Searched on September 3, 2020>

[0042] (Control unit 130) Now, let's return to Figure 2 and continue the explanation. The control unit 130 is realized when a processor, MPU (Micro Processing Unit), CPU (Central Processing Unit), etc., executes various programs stored in the memory unit 120 using RAM as a working area.

[0043] The control unit 130 is implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). As shown in Figure 2, the control unit 130 includes a receiving unit 131, a storage unit 132, a processing unit 133, a generation unit 134, and an output unit 135.

[0044] (Reception desk 131) The reception unit 131 receives information such as conversations related to teaching and coaching, answers to practice problems, and questions entered by the user in chat format, etc. The reception unit 131 also receives information that the processing unit 133, described later, uses to generate (process) practice problems.

[0045] Here, using Figure 5, an example of a reception screen for information used to generate practice problems and answers to the generated practice problems, as used by the reception unit 131, will be explained. Figure 5 is a diagram showing an example of a reception screen according to the embodiment.

[0046] As shown in Figure 5, the reception unit 131 receives the conditions for generating practice problems from administrators, etc., via the practice problem generation conditions reception screen shown in Figure 5 (1). For example, the reception unit 131 receives conditions for which content of the teaching materials, such as "teaching materials," "chapters," and "sections," should be used to generate practice problems (Figure 5 (1-1)). The reception unit 131 can also display a preview screen of the content identified based on the conditions shown in Figure 5 (1-1) (Figure 5 (1-2)). Furthermore, the reception unit 131 receives conditions such as "number of choices" and "number of problems to create" (Figure 5 (1-3)).

[0047] The reception unit 131 can then receive answers from users to the practice problems presented via the practice problem screen shown in Figure 5 (2). In addition to the practice problem screen (Figure 5 (2)), the reception unit 131 can also, of course, accept answers to the practice problems via chat.

[0048] (Storage section 132) The storage unit 132 stores educational data, which is data used for training users belonging to an organization, in the storage unit 120 (educational material information DB 121) for each organization. Specifically, the storage unit 132 stores educational data entered by training personnel belonging to an organization in the educational material information DB 121, associating it with information that identifies the organization (for example, the "organization identification information" shown in Figure 3).

[0049] Here, an example of a data reception screen for storing educational data will be explained using Figure 6. Figure 6 is a diagram showing an example of an educational data reception screen according to the embodiment. Figure 6 shows a registration screen (Figure 6(1)) used by education personnel in an organization when registering educational data for that organization. The screen shown in Figure 6(1) includes "Title (Figure 6(1-1))", "Document registration screen (Figure 6(1-2))", and "Video registration screen (Figure 6(1-3))".

[0050] In Figure 6 (1-1), the "Title" field is used to input information that identifies the educational data, such as the name of the educational data. The document registration screen shown in Figure 6 (1-2) displays items for selecting the file format of the document data to be registered and the file to be registered, as well as a preview of the document to be registered. The video registration screen shown in Figure 6 (1-3) displays items for selecting the file format of the video data to be registered and the file to be registered, a preview of the document to be registered, and an option to disable video fast-forwarding.

[0051] (Processing section 133) Returning to Figure 2, the explanation continues. The processing unit 133 processes the educational data stored in the storage unit 120 (educational material information DB 121) for each organization into tasks used by users for learning about the organization.

[0052] Specifically, the processing unit 133 generates learning videos and practice problems based on constraints regarding the format of the tasks. Here, constraints regarding the format of the tasks include the type of task, such as "video" or "practice problems."

[0053] For example, if the type of assignment is "a video for practice," the processing unit 133 generates a video file using the text and figures included in the educational data. For example, the processing unit 133 extracts the text data included in the educational data. Next, the processing unit 133 generates audio data that reads the extracted text data aloud. Then, the processing unit 133 generates a video file using the extracted text data and the generated audio data.

[0054] On the other hand, if the type of task is "practice problems," the processing unit 133 generates practice problems that satisfy the above constraints based on a large-scale language model, using the text and figures included in the educational data. Here, constraints regarding the format of the task include the number of practice problems, the number of choices, the difficulty level, constraints in problem generation such as "do not generate the same problem," and character limits. For example, the generation of practice problems by the processing unit 133 will be explained with an example in a later section.

[0055] (Generation unit 134) The generation unit 134 inputs information about the learning of users being educated within the organization into a large-scale language model to generate learning support information tailored to the user's learning status. Specifically, the generation unit 134 inputs user learning information, which includes at least one of the following: the user's learning history, the user's learning goals, and past conversation data, into a large-scale language model to generate user learning support information.

[0056] As an example of the learning support information described above, the generation unit 134 generates messages to support the user's learning (hereinafter referred to as "learning support messages"), answers to questions, tasks to be presented to the user, and information to follow up on the answers to those tasks. The generation process by the generation unit 134 will be explained in the following sections using specific prompt examples.

[0057] (Output section 135) The output unit 135 outputs the tasks processed by the processing unit 133 and the learning support information generated by the generation unit 134 to the user. For example, the output unit 135 outputs the aforementioned tasks and learning support information to the user in chat format. The output processing by the output unit 135 will be explained in the following sections.

[0058] (An example of the overall processing flow) From here, an example of the generation and output of learning support information by the support device 100 will be described. First, Figure 7 will be used to explain the flow of generation and output of tasks and learning support information by the support device 100. Figure 7 is a diagram showing a series of generation and output processes according to the embodiment.

[0059] The support device 100 (processing unit) performs processing using educational data for each organization stored in the educational material information DB 121 (Figure 7 (1-1)). The support device 100 (processing unit) then generates assignments such as educational videos and practice problems (Figure 7 (1-2)).

[0060] The support device 100 (output unit) outputs the generated tasks to the user. The user then uses the tasks, such as educational videos and practice problems, output by the support device 100 to learn.

[0061] The terminal device 200 transmits learning-related input information, such as "watching videos and practicing problems (working on assignments)" and "requests for coaching consultations," received from users who are learning using assignments processed from educational data, to the support device 100 (Figure 7(2)).

[0062] The support device 100 processes the generation of user learning support information in response to learning-related input information received from the terminal device 200. First, the support device 100 (generation unit) generates a prompt that includes a command to assign a predetermined role to the large-scale language model, a command to input the user's learning information, and a command to generate the user's learning support information (Figure 7 (3-1)). Next, the support device 100 (generation unit) inputs the generated prompt to the large-scale language model (Figure 7 (3-2)). Then, the support device 100 (generation unit) generates user learning support information based on the generated prompt.

[0063] Specifically, the support device 100 (generation unit) inputs prompts into the large-scale language model and sets predetermined roles such as "AI instructor who supports the user." The support device 100 (generation unit) also causes the large-scale language model to acquire user learning information from the user learning information DB 122 (Figure 7 (4-1)).

[0064] The support device 100 (generation unit) then uses the acquired user learning information to cause the large-scale language model to generate learning support information (Figure 7 (4-2)). For example, the support device 100 (generation unit) causes the large-scale language model to generate "learning support messages (Figure 7 (4-3))" and "answers to questions (Figure 7 (4-4))" as learning support information related to coaching. The support device 100 (generation unit) also causes the large-scale language model to generate "scoring and follow-up on answers (Figure 7 (4-5))" as learning support information related to teaching.

[0065] The support device 100 (output unit) outputs the generated learning support information to the terminal device 200 (Figure 7(5)). As a result, the terminal device 200 can display information such as "task and explanation information," "learning support messages," and "answers to user questions" to the user as learning support information (Figure 7(6)).

[0066] From here, we will explain an example of the generation and output processing of learning support information by the support device 100, using Figures 8 to 14, with more specific prompts and output screens as examples.

[0067] (Example 1: Generation and output of assignment and explanatory information) First, we will explain an example in which the support device 100 generates "tasks (practice problems)" as learning support information, which is output to the user by processing educational data, and outputs them to the user. In the first example, the support device 100 generates tasks according to the learning status of users belonging to an organization, using educational data for each organization, and outputs them to the user.

[0068] Furthermore, the support device 100 generates scoring results and explanatory information based on those scoring results for the user's answers to the assigned tasks, and outputs them to the user. The "explanatory information" here includes whether the user's answer is correct or incorrect, a detailed explanation of the correct answer, and information on transitions to educational data related to the correct answer.

[0069] Specifically, the support device 100 (processing unit) inputs a command to the large-scale language model to process the organization's educational data into tasks based on constraints regarding the format of tasks such as learning videos and practice problems, and processes the organization's educational data stored in the memory unit 120 for each organization into tasks.

[0070] Furthermore, the support device 100 (reception unit) receives answers to the tasks from the user. The support device 100 (generation unit) inputs the received answers and commands to generate scoring results and explanatory information for the answers into a large-scale language model, and generates the scoring results and explanatory information as learning support information.

[0071] Here, we will explain the input prompt examples and output examples in the processing related to the first example. Figure 8 is a diagram showing an example of prompts related to the first example. Figure 8 shows several examples of prompts (Figure 8(1) and (2)) for processing assignments (exercise problems) into a large-scale language model using educational data for each organization, and an example of the assignment processed and output based on the prompts, as well as the assignment scoring results and explanatory information (Figure 8(3)).

[0072] The first prompt shown in Figure 8(1) includes "#Background information (Figure 8(1-1))", "#Command (Figure 8(1-2))", and "#Constraints (Figure 8(1-3))".

[0073] As shown in Figure 8 (1-1), the support device 100 (processing unit) can assign the roles of "being an AI instructor that presents tasks to users learning about educational data" and "processing educational data into practice problems (# command)" to the large-scale language model.

[0074] "#command" is a directive that causes a large-scale language model to perform a predetermined process. For example, the directive shown in (1-2) of Figure 8 allows the support device 100 (processing unit) to cause the large-scale language model to perform a process such as "processing educational data into a predetermined format of questions, such as multiple-choice questions, while strictly adhering to the set constraints."

[0075] "#Constraints" are instructions used to define (restrict) the execution method of processing by the large-scale language model. For example, the instructions shown in (1-3) of Figure 8 allow the support device 100 (processing unit) to impose constraints on the large-scale language model, such as "limiting the number of questions," "limiting the answer choices," "presenting different questions," and "limiting the number of characters," when processing educational data into practice problems.

[0076] Furthermore, the second prompt shown in (2) of Figure 8 includes "# command". For example, the command shown in (2-1) of Figure 8 allows the support device 100 (processing unit) to cause the large-scale language model to perform tasks such as "the first time, generate a message summarizing the learning theme and confirming whether there are any questions related to the learning theme," and "from the second time onward, generate a message based on the message input by the user and the message generated immediately before."

[0077] The support device 100 (output unit) then outputs the task questions generated based on the prompts shown in Figure 8 (1) and (2), as shown in Figure 8 (3). For example, the support device 100 (output unit) outputs text such as "...(summary of learning theme)...Now, let's try solving the following problem." or "Question: What is the role of ○○?...(omitted)...Which do you think is the correct answer?" as messages displayed when a task is presented (Figure 8 (3-1)).

[0078] Here, we will explain the "example of output in chat format" described in Figure 8 using Figure 9. Figure 9 is a diagram showing an example of output of learning support information in chat format related to the first example. Figure 9 shows a chat-format screen for outputting practice problems generated by a large-scale language model, scoring results and explanatory information for answers to the practice problems, etc., in natural language.

[0079] For example, the support device 100 (output unit) can display a message to the user regarding the assignment generated by the large-scale language model, as shown in Figure 9 (1). The support device 100 (reception unit) can also receive answers and other information entered by the user (Figure 9 (2)).

[0080] Here, the support device 100 (generation unit) inputs a prompt containing commands such as "After receiving an answer from the user, please grade and explain the answer. After grading and explaining the answer, please present a new multiple-choice question" into the large-scale language model, thereby generating grading results and explanation information based on the large-scale language model according to the received answer.

[0081] The support device 100 (output unit) can display scoring results and explanatory information generated in accordance with the received answers to the user (Figure 9 (3)). Furthermore, the support device 100 (output unit) can present a task that has been generated with adjusted difficulty according to the user's answer results as the next problem (Figure 9 (3)).

[0082] As shown in Figures 9(4) and (5), the support device 100 can receive answers from the user and output scoring results and explanatory information generated according to the received answers. Furthermore, by repeatedly executing the above processes, the support device 100 effectively supports the user in completing practice problems and reviewing the explanations.

[0083] (Second example: Generating and outputting learning support messages) Next, we will describe an example in which the support device 100 generates a "learning support message" as learning support information and outputs it to the user. In the second example, the support device 100 generates a "learning support message," which is a message that supports the user's learning to progress effectively, according to the user's learning status, and outputs it to the user.

[0084] Specifically, the support device 100 (generation unit) inputs a command to a large-scale language model to generate a user learning support message that includes at least one of the following: words of encouragement for the user according to the user's learning progress, a summary of the user's learning content, and a message confirming whether or not there are any questions. The device then generates the user learning support message as learning support information.

[0085] Here, we will explain the input prompt examples and output examples in the processing related to the second example using Figures 10 and 11. Figures 10 and 11 are diagrams showing examples of prompts related to the second example.

[0086] Figure 10 shows an example of a prompt (Figure 10(1)) for generating learning support messages from a large-scale language model based on the user's learning progress, and an example of a learning support message generated and output based on that prompt (Figure 10(2)). The prompt shown in Figure 10(1) includes "#Background Information (Figure 10(1-1))", "#Status (Figure 10(1-2))", and "#Command (Figure 10(1-3))".

[0087] "#Background information" refers to instructions for assigning roles desired by educators and others to a large-scale language model. For example, the instructions shown in (1-1) of Figure 10 allow the support device 100 (generation unit) to assign the roles of "being an AI instructor that supports the learning user," "having the role of increasing the user's motivation to continue learning," and "generating information (#command) according to the user's learning status (#status) described in the prompt."

[0088] "#Status" is information that expresses the user's learning status in natural language, and is recorded based on information stored in the user learning information DB122. For example, the information shown in (1-2) of Figure 10 is "User A" and "Theme <1> This means "it has learned up to that point."

[0089] A "#command" is a directive that causes a large-scale language model to perform a predetermined process. For example, the command shown in (1-3) of Figure 10 allows the support device 100 (generation unit) to cause the large-scale language model to perform tasks such as "generating a message for the user to perform the same learning as the previous day, according to the user's learning status," "generating a message to encourage the user," "summarizing the learning content from the previous day," and "generating a message to check whether there are any questions."

[0090] The support device 100 (output unit) then outputs a learning support message generated based on the prompt shown in Figure 10 (1), as shown in Figure 10 (2).

[0091] Next, we will explain an example of generating a learning plan tailored to the user's learning progress using a large-scale language model, with reference to Figure 11. Figure 11 shows an example of a prompt (Figure 11(1)) for generating a learning plan tailored to the user's learning progress using a large-scale language model, and an example of a learning plan generated and output based on that prompt (Figure 11(2)).

[0092] The prompt shown in Figure 11(1) includes "#Background Information", "#Command (Figure 11(1-1))", "#Constraints (Figure 11(1-2))", "#Functions Usable for Learning (Figure 11(1-3))", "#Personal Assignment Data (Figure 11(1-4))", and "#Learning Progress Data (Figure 11(1-5))". Note that "#Background Information" is processed in the same way as in Figure 10 based on the commands described in Figure 11, so its explanation is omitted.

[0093] "#command" is a directive that causes a large-scale language model to perform a predetermined process, similar to the example in Figure 10. For example, the directive shown in (1-1) of Figure 11 allows the support device 100 (generation unit) to cause the large-scale language model to perform tasks such as "comparing the user's personal task data and learning progress data while strictly adhering to the set constraints, and generating advice for the user's learning progress."

[0094] "#Constraints" are instructions that define (restrict) the execution method of processing by a large-scale language model. For example, the instructions shown in (1-2) of Figure 11 allow the support device 100 (generation unit) to impose constraints on the large-scale language model when generating a learning plan, such as "generating a learning plan using only functions that can be used for learning," "character limit," "modification of the learning schedule," and "generating text to communicate the learning plan to the user."

[0095] "#Functions usable for learning" refers to information indicating the functions related to learning that a user can use when learning using the services realized by the support device 100, and is used as a constraint condition when the large-scale language model generates a learning plan. For example, as shown in (1-3) of Figure 11, "#Functions usable for learning" includes information that identifies various functions used for learning, such as "document data viewing function / video viewing function," "problem practice function," "question function," "personal coaching function," and "remote lab function."

[0096] "#Personal Assignment Data" contains information about the user's learning objectives. The information included in "#Personal Assignment Data" may be user learning information stored in the user learning information DB122. For example, as shown in (1-4) of Figure 11, "#Personal Assignment Data" includes information such as "Learning start date," "Learning completion deadline," and "Learning objectives using document data (document data objectives), learning objectives using videos (video objectives), and learning objectives using problems (problem objectives)."

[0097] The "#Learning Progress Data" field contains information about the user's learning progress. The information included in "#Learning Progress Data" may be user learning information stored in the User Learning Information DB122. For example, as shown in (1-5) of Figure 11, "#Learning Progress Data" may include information such as "Current Date," "Progress of learning using document data (Document Data Progress)," "Progress of learning using videos (Video Progress)," and "Progress of learning using problems (Problem Progress)."

[0098] The support device 100 (output unit) then outputs learning support messages generated based on the prompts shown in Figure 11(1), as shown in Figure 11(2). For example, the messages shown in Figure 11(2) include "current situation analysis," "necessary actions," "short-term goals," "suggestions for support functions," and "suggestions for schedule revision" (Figure 11(2-1)), based on a comparison between "#personal assignment data" and "#learning progress data," as well as messages to the user (Figure 11(2-2)).

[0099] For example, the "Current Situation Analysis" section might include messages such as, "As of [Date], there are [number] months remaining until the deadline. Progress on learning using document data is [percentage]%, progress on learning using videos is [percentage]%, and progress on learning using problems is [percentage]%." The "Required Actions" section might include messages such as, "You have [percentage]% remaining to reach your target. You need to progress by [percentage]% each day. We recommend making full use of the document data viewing function and setting a goal to read at least [number] chapters each day."

[0100] "Short-term goals" include things like "advance learning using document data to X%", "advance learning using videos to X%", and "advance learning using problems to X%". "Suggestions for support features" include things like "If you have any questions, use the question function to clarify them. Furthermore, if you face individual challenges, you should also consider using the person-to-person coaching function to receive direct advice." "Suggestions for schedule revision" include things like "Considering the current pace and estimated progress pace, it is highly likely that it will be difficult to meet the deadline, especially for the document data portion. Therefore, please consider extending the deadline to Month Day."

[0101] Furthermore, the message to the user (Figure 11 (2-2)) includes phrases such as, "Follow this plan to continue your daily learning, regularly evaluate your progress, and make adjustments as needed."

[0102] Here, we will explain the "example of output in chat format" described in Figures 10 and 11 using Figure 12. Figure 12 shows an example of output of learning support information in chat format related to the second example.

[0103] Figure 12 shows a chat-style screen for outputting learning support messages and learning plans generated by a large-scale language model. For example, the support device 100 (output unit) can display learning support messages and learning plans generated by the large-scale language model to the user, as shown in (1) of Figure 12.

[0104] Furthermore, the support device 100 (reception unit) can receive replies and questions regarding learning support messages and learning plans entered by the user (Figure 12 (2)). Then, the support device 100 (generation unit) can generate further learning support messages and learning plans in response to the received replies and questions.

[0105] (Third example: Answering a user's question) Next, we will describe an example in which the support device 100 generates "answers to user questions" as learning support information and outputs them to the user. In the third example, the support device 100 generates answers to questions according to the user's learning status and the user's questions, and outputs them to the user.

[0106] Specifically, the support device 100 (reception unit) receives questions from the user in a chat format. The support device 100 (generation unit) inputs the received questions and commands to generate answers to the questions into a large-scale language model, and generates answers to the questions as learning support information.

[0107] Here, we will explain the input prompt examples and output examples in the processing related to the third example using Figure 13. Figure 13 is a diagram showing an example of prompts related to the third example. Figure 13 shows several examples of prompts (Figure 13(1) and (2)) for generating answers in response to the large-scale language model in accordance with the user's learning status and questions from the user, and an example of an answer generated and output based on the prompt (Figure 13(3)).

[0108] The first prompt shown in Figure 13 (1) includes "#Background Information", "#Instruction (Figure 13 (1-1))", and "#Constraints (Figure 13 (1-2))". Note that "#Background Information" is processed in the same way as in Figure 8 based on the instructions described in Figure 13, so its explanation is omitted.

[0109] "#command" is a directive that causes the large-scale language model to perform a predetermined process, similar to the example in Figure 8. For example, the directive shown in (1-1) of Figure 13 allows the support device 100 (generation unit) to cause the large-scale language model to perform tasks such as "generating an answer to a question from a user regarding predetermined content (teaching materials / themes), while strictly adhering to the set constraints."

[0110] "#Constraints" are instructions that define (restrict) the execution method of processing by the large-scale language model, similar to the example in Figure 8. For example, the instruction shown in (1-2) of Figure 13 allows the support device 100 (generation unit) to impose constraints on the large-scale language model when generating an answer, such as "character limit" or "suggest an alternative learning function if an answer is not possible."

[0111] Furthermore, the second prompt shown in (2) of Figure 13 includes "# command (2-1) of Figure 13". For example, the command shown in (2-1) of Figure 13 allows the support device 100 (generation unit) to cause the large-scale language model to perform tasks such as "the first time, generate a message that summarizes the learning theme and confirms whether there are any questions about the learning theme," and "from the second time onward, generate a message that takes into account the message entered by the user and the message generated immediately before."

[0112] The support device 100 (output unit) then outputs the initial message generated based on the prompts shown in (1) and (2) of Figure 13, as well as the answers to the user's questions, as shown in (3) of Figure 13. For example, the support device 100 (output unit) outputs text such as "...(summary of learning theme)...Do you have any questions? (Figure 13 (3-1))" as the initial message.

[0113] Next, the user inputs a question to (3-1) in Figure 13, such as, "I've completely forgotten about XX. Could you explain it to me? (Figure 13 (3-2))." Then, the support device 100 (output unit) outputs a message such as "...(detailed answer)...(Figure 13 (3-3))" in response to the question input by the user.

[0114] Here, we will explain the "example of output in chat format" described in Figure 13 using Figure 14. Figure 14 is a diagram showing an example of output of learning support information in chat format related to the third example. Figure 14 shows a chat-format screen for outputting the initial message and answers to questions generated by a large-scale language model.

[0115] For example, the support device 100 (output unit) can display an initial message generated by a large-scale language model to the user, as shown in Figure 14 (1). The support device 100 (reception unit) can receive questions and other inputs from the user (Figure 14 (2)). The support device 100 (generation unit) can then display the generated answers to the user in response to the received questions and other inputs (Figure 14 (3)). In this way, the support device 100 can output interactive answers to questions from the user.

[0116] (Processing procedure of support device 100) From here, a series of processing procedures implemented by the support device 100 according to the first embodiment will be described. Figure 15 is a flowchart of the processing of the support device 100 according to the embodiment.

[0117] The storage unit 132 stores the educational data for each organization, entered by the organization's education staff, etc., in the teaching material information DB 121 (S101). At this point, the support device 100 waits for processing if the educational data is not to be processed into assignments (No. in S102).

[0118] On the other hand, if the educational data is to be processed into assignments (Yes in S102), the processing unit 133 processes the educational data into assignments (S103). Next, the output unit 135 outputs the assignments processed by the processing unit 133 to the user (S104).

[0119] Next, if learning support is to be provided to a user who is learning about the organization (Yes in S105), educational support information is generated according to the user's input (S106). Next, the output unit 135 outputs the educational support information (S107). On the other hand, if learning support is not to be provided to the user (No in S105), the support device 100 terminates the process.

[0120] Here, if the processing termination condition is not met (No in S108), the support device 100 returns to the previous step and continues processing. On the other hand, if the processing termination condition is met (Yes in S108), the support device 100 terminates the process. The processing termination condition referred to here is an arbitrary condition and includes, for example, input from the user, elapsed time, arrival of a set schedule, instructions from an administrator, etc.

[0121] (effect) The effects of the support device 100 according to this embodiment will now be explained. Conventionally, it has been difficult to process educational data managed by each organization to generate assignments for user education, making it challenging to adequately support user education.

[0122] Therefore, the storage unit 132 of the support device 100 according to this embodiment stores educational data, which is data used for educating users belonging to an organization, in the storage unit 120 for each organization. The processing unit 133 of the support device 100 processes the educational data stored in the storage unit 120 for each organization into tasks used by users for learning about the organization. The output unit 135 of the support device 100 outputs the processed tasks to the user.

[0123] Therefore, the support device 100 has the effect of enabling appropriate learning support according to the user's learning progress. Furthermore, the support device 100 achieves predetermined effects by executing the processes described below.

[0124] The processing unit 133 inputs a command to the large-scale language model to process the educational data into tasks based on constraints regarding the format of the tasks, and processes the educational data stored in the memory unit 120 for each organization into tasks.

[0125] Through the processing described above, the support device 100 can generate appropriate educational videos and exercises tailored to the educational data of each organization and provide them to the user. As a result, the user can appropriately conduct self-study on organizational-related education based on the educational videos and exercises provided by the support device 100, tailored to the educational data of each organization. Therefore, the support device 100 has the effect of enabling more efficient and effective self-study by the user than before.

[0126] The generation unit 134 inputs information about the user's learning, including at least one of the user's learning history, the user's learning goals, and past conversation data, into a large-scale language model to generate learning support information that corresponds to the user's learning status using the task.

[0127] Through the processes described above, for example, the support device 100 can provide teaching, coaching, and answer questions to the user according to the user's learning status and learning goals. As a result, the user can receive educational support from the support device 100 that is tailored to their learning situation, enabling them to resolve areas of confusion and deepen their understanding of the learning content more effectively than before. Therefore, the support device 100 has the effect of enabling the user to progress in their learning efficiently and effectively.

[0128] The reception unit 131 receives answers to exercises and other assignments from the user. The generation unit 134 inputs the received answers and commands to generate scoring results and explanatory information for the answers into a large-scale language model, and generates the scoring results and explanatory information as learning support information.

[0129] Through the processing described above, the support device 100 can provide scoring results, such as correctness, for the user's answers to the practice problems generated by the support device 100. In addition, the support device 100 can provide explanatory information, such as chapters, sections, or individual descriptions of educational data related to the user's answers and scoring results, as well as explanatory videos, in association with the user's answers.

[0130] As a result, users can engage in appropriate self-study, such as reviewing the results of their practice exercises and reviewing content they do not fully understand, based on the scoring results and explanatory information provided by the support device 100. Therefore, the support device 100 has the effect of enabling more efficient and effective self-study for users than before.

[0131] The generation unit 134 inputs a command to the large-scale language model to generate a user learning support message that includes at least one of the following: words of encouragement for the user according to the user's learning progress, a summary of the user's learning content, and a message confirming whether or not there are any questions, and generates the user learning support message as learning support information.

[0132] Through the processing described above, for example, if the user's learning progress is behind, the support device 100 can display encouraging words (words of support) to the user, such as "Your learning is behind, is there anything wrong? It's tough, but let's do our best!" Furthermore, to support the user's learning, the support device 100 can display a "summary of the learning content up to the last lesson" to the user, and then display questions such as "Do you have any questions about the learning content up to the last lesson?" to resolve any points of confusion the user may have and deepen their understanding of the learning content. As a result, the support device 100 has the effect of enabling the user to progress in their learning efficiently and effectively, according to the user's learning progress.

[0133] The reception unit 131 receives questions from users in a chat format. The generation unit 134 inputs the received questions and commands to generate answers to those questions into a large-scale language model, and generates answers to the questions as learning support information.

[0134] Through the process described above, the support device 100 can, for example, display an explanation of "XX" as an answer to a question from a user, such as "What is XX?". As a result, the user can appropriately engage in self-study based on the answers provided by the support device 100 to their questions. Therefore, the support device 100 has the effect of enabling more efficient and effective self-study for users than before.

[0135] The output unit 135 outputs the generated learning support information to the user in chat format. Through the above-described process, the support device 100 enables natural interaction in a conversational format using natural language for coaching, teaching, receiving and answering questions from the user. As a result, the support device 100 has the effect of providing an environment in which the user can naturally consult with or ask questions of the support device 100.

[0136] <Variation> The following describes modifications that can be implemented by the support device 100 according to this embodiment.

[0137] (Another form of assignment generation and grading) The support device 100 described above is said to process educational data based on a large-scale language model to generate assignments and perform scoring and other related tasks, but it is not limited to this.

[0138] For example, the support device 100 presents the user with a task entered by an administrator or the like. Next, the support device 100 receives the user's answer to the task. Then, the support device 100 compares the received answer with a pre-registered correct answer and scores it.

[0139] Through the above processing, the support device 100 can present users with questions that educators or other personnel request, rather than practice questions processed from educational data. As a result, the support device 100 can present users with practice questions that focus on learning topics that educators or other personnel consider particularly important.

[0140] (Data, etc.) The educational data, tasks, coaching, teaching, learning support messages, etc. for each organization, as well as the names of the functional parts of the support device 100, steps, processes, names of steps or processes, etc., used in the description of the above embodiment are merely examples and can be changed at will.

[0141] For example, while it was explained that the teaching material information DB121 stores items such as "organization identification information," "title," "link," "order," "number of content items," "task," and "type," as well as information related to those items, in a table format, etc., associated with "No," which is information that identifies individual teaching material information, the items, content, and storage format are not limited. Similarly, while it was explained that the user learning information DB122 stores items such as "user identification information," "teaching materials used," "learning history," "learning objectives," and "conversation history," as well as information related to those items, in a table format, etc., associated with "No," which is information that identifies individual user learning information, the items, content, and storage format are not limited.

[0142] (Regarding the use of generative models) In this embodiment, the model (large-scale language model) used by the support device 100 is described as being stored in the generated model DB 123 of the storage unit 120, but this is not limited to this. For example, the support device 100 can access an external information processing device (server, etc.) and use a predetermined model.

[0143] (Flowcharts, etc.) In flowcharts, each step may be rearranged as long as it does not create inconsistencies, and some steps may be omitted. Furthermore, conjunctions such as "next," "continue," "in addition," "at this time," and "on this occasion" in flowchart descriptions do not limit the order or timing of the processes in the flowchart.

[0144] (Systems, etc.) Of the processes described in the embodiments and modifications described above, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0145] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily have to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc.

[0146] The aforementioned components include those that are easily conceivable by those skilled in the art, those that are substantially identical, and those that fall within the so-called equivalent range. Furthermore, the embodiments and modifications described above can be combined as appropriate, as long as the processing content is not contradictory.

[0147] Furthermore, the terms "section," "module," and "unit" mentioned above can be replaced with "means" or "circuit," etc. For example, a control unit can be replaced with a control means or a control circuit.

[0148] Although several embodiments have been described in detail above with reference to the drawings, these are merely examples, and it is possible to implement these embodiments in various modified and improved forms based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0149] <Hardware Configuration> The device included in the support device 100 according to this embodiment is implemented by a computer 1000 having a configuration as shown in Figure 16. Figure 16 is a diagram showing an example of the hardware configuration of a computer that implements the support device 100 according to this embodiment.

[0150] The computer 1000 has a configuration in which a CPU 1100, memory 1200, auxiliary storage device 1300, input interface 1400, output interface 1500, and communication interface 1600 are connected by a bus 1700.

[0151] The CPU 1100 operates based on programs stored in the memory 1200 or auxiliary storage device 1300, and controls each functional unit. The memory 1200 consists of, for example, RAM (Random Access Memory) or ROM (Read Only Memory), and stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0152] For example, when the computer 1000 functions as the support device 100 according to this embodiment, the CPU 1100 of the computer 1000 can realize the functions of the control unit 130 by executing a program loaded on the memory 1200.

[0153] The auxiliary storage device 1300 stores programs executed by the CPU 1100, as well as data used by such programs. The CPU 1100 controls input devices 1410, such as keyboards and mice, via the input interface 1400. The CPU 1100 also acquires data from the input devices 1410 via the input interface 1400.

[0154] The CPU 1100 controls output devices 1510, such as displays and printers, via the output interface 1500. The CPU 1100 also outputs generated data to the output devices 1510 via the output interface 1500.

[0155] The communication interface 1600 receives data from other devices via a predetermined network NW and sends it to the CPU 1100, and the CPU 1100 transmits the generated data to other devices via the predetermined network NW. [Explanation of Symbols]

[0156] 100 Support equipment 110 Communications Department 120 Storage section 121 Teaching material information DB 122 User Learning Information Database 123 Generative Model DB 130 Control Unit 131 Reception Department 132 Storage Unit 133 Processing Department 134 Generation part 135 Output section

Claims

1. A storage unit that stores data related to organizational education, which is data used for educating users belonging to an organization, in a storage unit for each organization, A processing unit that processes the data on education of the organization stored in the memory unit for each organization into predetermined tasks used for learning about the organization by the user, An output unit that outputs the processed predetermined task to the user, A support device characterized by having the following features.

2. The aforementioned processing section is Based on constraints regarding the format of a predetermined task, a command is input to the large-scale language model to process the data on the organization's education into the predetermined task, and the data on the organization's education stored in the memory unit for each organization is processed into the predetermined task. The support device according to feature 1.

3. The system further includes a generation unit that inputs information related to the user's learning, including at least one of the user's learning history, the user's learning goals, and past conversation data, into a large-scale language model, and generates information to support the user's learning according to the user's learning status using the predetermined task. The support device according to feature 2.

4. The system further includes a receiving unit that receives answers to the predetermined tasks from the user, The generating unit is The received answer and a command to generate scoring results and explanatory information for the answer are input to the large-scale language model to generate the scoring results and explanatory information as information to support the user's learning. The support device according to feature 3.

5. The generating unit is The large-scale language model is given a command to generate a message that supports the user's learning, which includes at least one of the following: words of encouragement for the user according to the user's learning progress, a summary of the user's learning content, and a message to check whether there are any questions. The message that supports the user's learning is then generated as information that supports the user's learning. The support device according to feature 3.

6. The system further includes a reception unit that receives input from the user in chat format regarding the predetermined task, The generating unit is The received question and the command to generate an answer to the question are input into the large-scale language model, and the answer to the question is generated as information to support the user's learning. The support device according to feature 3.

7. The output unit is, The generated information to support the user's learning is output to the user in a chat format. The support device according to any one of claims 3 to 6.

8. A support method to be executed by a support device, A storage step involves storing data related to organizational education, which is data used for educating users belonging to the organization, in a storage unit for each organization. A processing step of processing the data related to the education of the organization stored in the memory unit for each organization into predetermined tasks used for learning about the organization by the user, An output step of outputting the processed predetermined task to the user, A support method characterized by including

9. A storage step involves storing data related to organizational education, which is data used for educating users belonging to the organization, in a storage unit for each organization, A processing step of processing the data related to the education of the organization stored in the memory unit for each organization into predetermined tasks used for learning about the organization by the user, An output step of outputting the processed predetermined task to the user, A support program characterized by causing a computer to execute a command.

Images