Learning support system, learning support method, and learning support program

Abstract

To provide a learning support system capable of answering a question while suppressing dependence on an answerer.SOLUTION: A learning support system 100 comprises a response unit 101, an answerer determination unit 102, and an output unit 103. The response unit 101 determines whether an answerer answers a question from a learner who learns using learning content or whether to answer the question using an answer model for generating answers to questions. When an answerer answers the question, the answerer determination unit 102 determines an answerer who answers the question on the basis of availability information to learning content of answerer candidates. When the response unit answers a question using the answer model, the output unit 103 outputs data of an answer generated by the answer model with the question as an input.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a technique for providing learning content. [Background technology]

[0002] In companies and the like, learning may be conducted using learning content created as electronic media in order to pass on skills or knowledge. In such cases, if a learner using the learning content has a question, he or she can, for example, ask the question via the learning system that provides the learning content and receive a reply from a person set as the person to whom the question should be sent. In such a system, it is desirable to have a system that allows the learner to proceed with the learning without being left with any questions while suppressing dependency on the answerer.

[0003] The answer support system of Patent Document 1 is a system that presents answer candidates to a question. The answer support system of Patent Document 1 predicts an answer to a question using a learning model generated by learning the relationship between questions and answers, and displays the answer candidates on the terminal device of the answerer. The answer support system of Patent Document 1 answers the question by transmitting an answer selected by the answerer from the answer candidates to the terminal device of the person who asked the question. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-71679 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the answer support system of Patent Document 1 assists the answerer in answering by displaying answer candidates predicted by a learning model. Therefore, in order to answer a question, the technology of Patent Document 1 requires an answerer to be present in a state where he or she can respond to the question, and it is difficult to answer the question if there is no answerer.

[0006] In order to solve the above problems, the present invention has an object to provide a learning support system and the like that can answer questions while reducing dependency on answerers. [Means for solving the problem]

[0007] In order to solve the above problems, the learning support system of the present invention includes a response unit, an answerer determination unit, and an output unit. The response unit determines whether to answer questions from learners who are studying using learning content, the answerer will be the answerer or the answer will be made using an answer model that generates answers to questions. When an answerer will be the answerer to the question, the answerer determination unit determines the answerer who will answer the question based on information on whether the answerer candidate can handle the learning content. When an answer is made using the answer model, the output unit outputs answer data generated by the answer model using the question as input.

[0008] The learning support method of the present invention determines whether questions from learners who are studying using learning content will be answered by an answerer or by using an answer model that generates answers to questions. When an answerer will answer a question, the learning support method of the present invention determines the answerer who will answer the question based on information on whether the candidate answerer can handle the learning content. When an answer is to be answered by an answer model, the learning support method of the present invention inputs the question and outputs answer data generated by the answer model.

[0009] The learning support program of the present invention determines whether a question from a learner who is studying using learning content will be answered by an answerer or by using an answer model that generates an answer to the question. When an answerer will answer a question, the learning support program of the present invention determines the answerer who will answer the question based on information on whether the candidate answerer can handle the learning content. When an answerer answers a question, the learning support program of the present invention performs a process of transferring the question and the answer between the terminal device of the determined answerer and the terminal device of the learner. When an answerer is to be answered using an answer model, the learning support program of the present invention outputs answer data generated by the answer model using the question as input. Effect of the Invention

[0010] According to the present invention, it is possible to answer questions while reducing dependency on the answerer. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram showing an outline of a configuration of a first exemplary embodiment of the present invention. [Diagram 2] 1 is a diagram illustrating an example of a configuration of a learning support system according to a first embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram illustrating an example of an operation flow of the learning support system according to the first embodiment of the present invention. [Figure 4] FIG. 2 is a diagram illustrating an example of an operation flow of the learning support system according to the first embodiment of the present invention. [Diagram 5] FIG. 2 is a diagram illustrating an example of an operation flow of the learning support system according to the first embodiment of the present invention. [Figure 6] FIG. 2 is a diagram showing an example of a display screen in the first embodiment of the present invention. [Figure 7] FIG. 2 is a diagram showing an example of a display screen in the first embodiment of the present invention. [Figure 8] FIG. 11 is a diagram illustrating an example of a configuration of a learning support system according to a second embodiment of the present invention. [Figure 9]FIG. 11 is a diagram illustrating an example of an operation flow of the learning support system according to the second exemplary embodiment of the present invention. [Figure 10] FIG. 13 is a diagram illustrating an example of another configuration of the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] (First embodiment) A first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing an outline of the configuration of a learning system of this embodiment. The learning system of this embodiment includes a learning support system 10, a learning terminal 20, and a respondent terminal 30. The learning support system 10 and the learning terminal 20 are connected via a network. The learning support system 10 and the respondent terminal 30 are also connected via a network. A plurality of learning terminals 20 and a plurality of respondent terminals 30 may be provided.

[0013] The learning system of this embodiment is a system that provides learning content, and answers questions from learners who are studying using the provided learning content by providing answers by an answerer or using a previously generated learning model. When providing an answer by an answerer, the learning support system 10 acquires the question from the learner from the learning terminal 20 and transfers it to the answerer terminal 30. The learning support system 10 also acquires the answer to the question from the answerer terminal 30 and transfers it to the learning terminal 20. Furthermore, if a learning model adapted to the learning content and the learner has already been generated, the learning support system outputs an answer generated using the learning model to the learning terminal 20.

[0014] The following describes the configuration of the learning support system 10. Fig. 2 is a diagram showing an example of the configuration of the learning support system 10. The learning support system 10 includes a learning management unit 11, a response unit 12, a model generation unit 13, a respondent determination unit 14, a reward calculation unit 15, an access management unit 16, a storage unit 17, an input unit 18, and an output unit 19.

[0015] The learning management unit 11 outputs learning content in response to a request from the learning terminal 20 to the learning terminal 20 via the output unit 19. The learning content is, for example, content in which the information to be studied is composed of audio, images, and text, and the learner studies by viewing the content. The learning content runs, for example, on a web browser or a dedicated application program on the learning terminal 20.

[0016] The response unit 12 determines whether the answer to the question from the learner is to be made by the answerer or to be made using an answer model, which is a learning model generated by machine learning. The response unit 12 determines whether the answer to the question is to be made by the answerer or to be made using an answer model, based on a criterion set for each learning content. For example, when the learner selects learning content for which an answer model has been generated, the response unit 12 determines that the answer is to be made using the answer model. Also, when the learner selects learning content for which an answer model has not been generated, the response unit 12 determines that the answerer is to be made. Also, when an answer model exists, the response unit 12 may determine whether the answer is to be made by the answerer or to be made using the answer model, based on the knowledge level of the learner. For example, the response unit 12 may determine that the answer is to be made by the answerer when the knowledge level of the learner is higher than a preset criterion, and that the answer is to be made using the answer model when the knowledge level of the learner is lower than the criterion.

[0017] When the respondent answers a question, the response unit 12 sends the respondent's answer to the question from the learner to the questioner. The response unit 12 transfers the question received from the learning terminal 20 to the respondent terminal 30, and transfers the answer received from the respondent terminal 30 to the learning terminal 20. The response unit 12 associates the question with the answer and stores them in the memory unit 17. The response unit 12 may output information indicating that a question has been asked to the respondent terminal 30, and output data on the content of the question to the respondent terminal 30 via the output unit 19 in response to a request from the respondent terminal 30. The response unit 12 may also output information indicating that a question has been asked to the learning terminal 20, and transmit data on the content of the answer to the learning terminal 20 in response to a request from the learning terminal 20.

[0018] The model generation unit 13 learns the relationship between the questions by the learner and the answers by the answerer, and generates an answer model that generates an answer to the question of the learner. The answer model is generated, for example, for each learning content or for each field that the learning content targets. The model generation unit 13 learns the relationship between the questions and the answers by, for example, deep learning using a neural network, and generates an answer model as a learning model. The model generation unit 13 executes machine learning using, for example, data selected by an expert in the field of the learning content from data of questions and answers made when learning the learning content as input data, and generates an answer model that generates an answer from a question. The model generation unit 13 stores the data of the generated answer model in the storage unit 17. The model generation unit 13 may also perform re-learning using data of questions and answers made using the generated answer model, and update the answer model.

[0019] The respondent determination unit 14 determines a person suitable as a respondent according to the learner and the learning content. The respondent determination unit 14 determines a respondent by selecting from the candidate respondents based on a preset criterion using the registered data of the learner and the field of the learning content, and the registered data of the respondent. The registered data of the learner is composed of data on one or more items of the learner's learning history, knowledge level by field, industry, work content, affiliation, work experience, academic background, and specialty field. The registered data of the learner may include data on other items. The registered data of the respondent is composed of data on one or more items of the respondent's specialty field, the respondent's knowledge level by specialty field, past evaluations of the respondent by learners, work history as a respondent, and the respondent's work schedule. The registered data of the respondent may include data on other items.

[0020] The answerer determination unit 14 determines, for example, as an answerer, a person whose knowledge level exceeds a standard among people whose field of expertise matches the field of the learning content. The knowledge level is an index of the amount or level of knowledge in the field of the learning content. The standard for comparing the answerer's knowledge level when determining an answerer is set, for example, so that the answerer's knowledge level is higher than the learner's knowledge level so that the answerer can appropriately answer questions from the learner. The knowledge level may also be set by a numerical value, or may be set by multiple stages.

[0021] The respondent determination unit 14 may determine the respondent using a learning model generated by machine learning. The learning model for determining the respondent is generated by performing machine learning using the learner's registration data, data indicating the contents of the learning content, and the respondent's registration data as input data, and data indicating suitability as a respondent as a label. When determining the respondent using the learning model generated by machine learning, the respondent determination unit 14 predicts a person suitable as a respondent using the learning model, using the learner's registration data and data indicating the contents of the learning content as input data.

[0022] The remuneration calculation unit 15 calculates the remuneration to be paid to the respondent for the work of answering the answer provided by the respondent. The remuneration calculation unit 15 calculates the amount of remuneration to be paid to the respondent using a reference amount that is preset using, for example, information on the learning content and the evaluation from the learner, and the number of answers. The reference amount may be set differently for each field or each respondent. The remuneration calculation unit 15 may also calculate the remuneration using a value other than a monetary amount, such as points.

[0023] The access management unit 16 manages whether or not learners using the learning terminal 20 and respondents using the respondent terminal 30 can access the learning support system 10. The access management unit 16 authenticates learners and respondents, for example, using an identification (ID) number and a password. The access management unit 16 permits authenticated learners and respondents to access the learning support system 10. The access management unit 16 may authenticate learners and respondents using biometric authentication technology.

[0024] The access management unit 16 may manage the range of access for each learner. For example, the access management unit 16 manages the learning content that can be accessed for each service to which the learner subscribes. The access management unit 16 may also manage the range of access for each respondent. For example, the access management unit 16 manages the learning content that can be accessed for each respondent and the range of information registered in the learning support system.

[0025] Memory unit 17 stores learning contents, registered data of learners, registered data of respondents, access management information, question and answer history data, reward standard data, information on reward amounts for respondents, and answer model data.

[0026] The input unit 18 acquires input data that has been input to the learning terminal 20 by the learner from the learning terminal 20. The input unit 18 acquires from the learning terminal 20 the learner's authentication information, the results of the selection of learning content, the answerer's evaluation results, and question data.

[0027] The input unit 18 acquires input data that has been input to the respondent terminal 30 by an operation of the respondent, from the respondent terminal 30. The input unit 18 acquires from the respondent terminal 30 the authentication information of the respondent and answer data.

[0028] The output unit 19 outputs the learning content, data on the answers to the questions, and information requesting an evaluation by the answerer to the learning terminal 20. The output unit 19 outputs the learning content, the registration information of the learner, data on the questions to be answered, and data on the amount of reward to the answerer terminal 30.

[0029] The processes in the learning management unit 11, the response unit 12, the model generation unit 13, the respondent determination unit 14, the reward calculation unit 15, the access management unit 16, the input unit 18, and the output unit 19 can be performed, for example, by executing a computer program on a CPU (Central Processing Unit). The processes in the learning management unit 11, the response unit 12, the model generation unit 13, the respondent determination unit 14, the reward calculation unit 15, the access management unit 16, the input unit 18, and the output unit 19 are performed, for example, in one information processing device.

[0030] The processes in the learning management unit 11, the response unit 12, the model generation unit 13, the respondent determination unit 14, the reward calculation unit 15, the access management unit 16, the input unit 18, and the output unit 19 may be performed in multiple information processing devices connected via a network. For example, the processes in the learning management unit 11, the response unit 12, the reward calculation unit 15, and the access management unit 16 may be performed in the same information processing device, and the generation of the answer model in the model generation unit 13 may be performed in another information processing device. In such a configuration, the memory unit 17, the input unit 18, and the output unit 19 are provided in each information processing device.

[0031] The storage unit 17 is configured using, for example, a hard disk drive. The storage unit 17 may be configured with other types of storage devices such as a non-volatile semiconductor storage device or a combination of multiple types of storage devices. The storage unit 17 may be provided on a storage device connected to the learning support system 10. The storage unit 17 may also be provided on a storage device controlled by an information processing device connected via a network.

[0032] The learning terminal 20 displays on a display device (not shown) the learning content, answers to questions, and other data acquired from the learning support system 10. If the learning content uses audio, the learning terminal 20 outputs audio to an audio output device (not shown).

[0033] The learning terminal 20 outputs to the learning support system 10 as input data the authentication data of the learner, the selection of the learning content, the question content, and the answerer's evaluation, which are input by the learner's operation.

[0034] The respondent terminal 30 outputs the respondent's authentication information and answer data to questions, which are input by the respondent's operation, to the learning support system 10 as input data. The respondent terminal 30 also acquires question data from the learner and reward amount data from the learning support system 10, and displays the question and reward amount on a display device (not shown). The respondent terminal 30 acquires the learning content that the learner is studying from the learning support system 10, and displays the acquired learning content on the display device. If the learning content uses audio, the respondent terminal 30 outputs audio to an audio output device (not shown).

[0035] The learning terminal 20 and the respondent terminal 30 may be, for example, a personal computer, a tablet computer, or a smartphone. The learning terminal 20 and the respondent terminal 30 may also be configured by a combination of multiple information terminal devices. For example, the learning terminal 20 and the respondent terminal 30 may be configured by a combination of a wearable or goggle-type information terminal device and another information terminal device.

[0036] The operation of the learning support system will be described. Figs. 3, 4 and 5 are diagrams showing an example of the operation flow of the learning support system 10. Fig. 3 is a diagram showing an example of the operation flow of the learning support system 10 when a learner uses learning content to study. Fig. 4 is a diagram showing an example of the operation flow of the learning support system 10 when calculating the amount of remuneration to be paid to an answerer. Also, Fig. 5 is a diagram showing an example of the operation flow of the learning support system 10 when generating an answer model.

[0037] When the learner operates the learning terminal 20 to input a request to start learning, the learning terminal 20 outputs a request to start learning to the learning support system 10. When the learning support system 10 receives the request to start learning from the learning terminal 20, it starts processing related to the provision of learning content. When processing related to the provision of learning content starts, the learning support system 10 authenticates the learner who is using the learning terminal 20.

[0038] The access management unit 16 of the learning support system 10 acquires the authentication information of the learner, which is input to the learning terminal 20 by the learner, from the learning terminal 20. When acquiring the authentication information, the access management unit 16 collates it with the registration data of the learner stored in the memory unit 17, and authenticates the learner if it matches.

[0039] When the learner is authenticated, the access management unit 16 outputs, via the output unit 19, to the learning terminal 20, information on the learning content that the learner can study.

[0040] When the selection result of the study content is input to the study terminal 20 by the learner's operation, the study terminal 20 outputs information on the selection result of the study content to the study assistance system 10.

[0041] In FIG. 3, the input unit 18 of the learning support system 10 acquires information on the selection result of the learning content from the learning terminal 20 (step S11).

[0042] When the information on the selection result of the learning content is acquired, the response unit 12 determines whether the question is answered using an answer model or by an answerer. When the answer is to be answered by an answerer (Yes in step S12), the answerer determination unit 14 determines an answerer who will answer the question (step S13). The answerer determination unit 14 determines, for example, as an answerer, a person whose field of expertise matches the field of the learning content and whose knowledge level exceeds the standard set for each learning content.

[0043] When the answerer is determined, the learning management unit 11 outputs data of the learning content according to the selection result to the learning terminal 20 via the output unit 19 (step S14).

[0044] When the answerer is determined in step S13, the access management unit 16 transmits a notification to the answerer terminal 30 carried by the answerer, transmits the answerer's authentication information, and authenticates the answerer. The answerer determination unit 14 may also determine an answerer who will answer the question from among the answerers who are already connected to the learning support system 10.

[0045] When the learning content data is acquired, the learning terminal 20 displays the learning content on a display device. If the learning content includes audio data, the learning terminal 20 outputs the audio to an audio output device such as a speaker. The learner operates the learning terminal 20 to view the learning content and study.

[0046] When a question is input to the learning terminal 20 by the learner's operation, the learning terminal 20 outputs data of the input question to the learning support system 10.

[0047] The input unit 18 of the learning support system 10 acquires the question data from the learning terminal 20. When the question data is acquired, the response unit 12 transfers the question data to the respondent terminal 30 operated by the person determined as the respondent (step S15).

[0048] When the respondent terminal 30 receives the question data, it displays the question on the display device. When the respondent operates the respondent to input answer data to the respondent terminal 30, the respondent terminal 30 outputs the answer data to the learning support system 10.

[0049] The input unit 18 of the learning support system 10 acquires the answer data from the respondent terminal 30. When the answer data is acquired, the response unit 12 transfers the answer data to the learning terminal 20 operated by the learner who asked the question (step S16). After transferring the answer data, the response unit 12 stores the content of the question and answer, i.e., data associating the question and the answer, in the storage unit 17 (step S17).

[0050] When the learning terminal 20 receives the answer data to the question, the learning terminal 20 displays the received answer data on a display device (not shown).

[0051] If the learning continues and questions continue to be asked (No in step S18), the learning support system 10 continues the operations from step S15. If the learning ends (Yes in step S18), the learning support system 10 ends the operations related to providing the learning content and processing the questions and answers.

[0052] In step S12, when the answer is made using the answer model (No in step S12), learning management unit 11 outputs data of the learning content according to the selection result to learning terminal 20 via output unit 19 (step S19).

[0053] When the learning content data is acquired, the learning terminal 20 displays the learning content on the display device. If the learning content includes audio data, the learning terminal 20 outputs the audio to the audio output device. The learner operates the learning terminal 20 to study using the learning content.

[0054] When a question is input to the learning terminal 20 by the learner's operation, the learning terminal 20 outputs data of the input question to the learning support system 10.

[0055] The input unit 18 of the learning support system 10 acquires question data from the learning terminal 20. When the question data is acquired, the response unit 12 generates an answer from the question using an answer model (step S20). When the answer is generated, the response unit 12 outputs the answer data to the learning terminal 20 operated by the learner who asked the question via the output unit 19 (step S21). When the answer data is transferred, the response unit 12 stores the content of the question and answer, i.e., data associating the question with the answer, in the memory unit 17.

[0056] When the learning terminal 20 receives the answer data to the question, the learning terminal 20 displays the received answer data on a display device (not shown).

[0057] If the learning continues and questions continue to be asked (No in step S22), the learning support system 10 continues the operations from step S20. If the learning ends (Yes in step S22), the learning support system 10 ends the operations related to providing the learning content and processing the questions and answers.

[0058] Fig. 6 shows an example of a display screen output by the learning terminal 20 when a learner uses learning content to study. The display screen shown in Fig. 6 is composed of a learning window in which the learning content is displayed, a question input field in which the learner inputs and submits questions, and a field for displaying questions and answers. By studying using a display screen like that shown in Fig. 6, the learner can deepen his / her proficiency in the content of the learning content while asking questions as he / she progresses with his / her study using the learning content.

[0059] Fig. 7 shows an example of a display screen output by the respondent terminal 30 when the respondent answers a question. The display screen shown in Fig. 7 is composed of a learning window that displays the learning content that the learner is studying, a question answer history field that displays the learner's attributes, question content, and question and answer history, and an answer input field for inputting and sending answers. By answering questions using a display screen such as that shown in Fig. 7, the respondent can answer the question while understanding the learner's situation.

[0060] The operation of calculating the reward to be paid to the answerer will be described with reference to Fig. 4. When the learner completes the learning content, the reward calculation unit 15 of the learning support system 10 starts calculating the reward to be paid to the answerer. The reward calculation unit 15 specifies whether the answer to the question was given by the answerer or not in the learning for which the reward is calculated.

[0061] If the question has been answered by an answerer (Yes in step S31), the reward calculation unit 15 acquires, from the learning terminal 20, data on the evaluation result input to the learning terminal 20 by the learner's operation (step S32).

[0062] When the evaluation results are acquired, the reward calculation unit 15 of the learning support system 10 calculates the amount of reward to be paid to the answerer using the reference amount and the number of answers set for each evaluation result (step S33). After calculating the amount of reward, the reward calculation unit 15 associates the amount of reward with the answerer and stores it in the storage unit 17 (step S34).

[0063] In step S31, if the answer is based on an answer model (No in step S31), the reward calculation unit 15 calculates the reward using the usage record of the answer model. The reward calculation unit 15 calculates the amount of reward to be paid to the answerer according to, for example, the contribution degree of each answerer set for each answer model and the number of times the answer model is used. After calculating the amount of reward, the reward calculation unit 15 associates the amount of reward with the answerer and stores it in the storage unit 17 (step S34).

[0064] After calculating the reward, the learning assistance system 10 ends the process related to providing the learning content, and waits until it receives another request.

[0065] The operation of generating an answer model will be described with reference to Fig. 5. The model generation unit 13 acquires data associating questions and answers to be used in generating an answer model (step S41). The model generation unit 13 reads out data designated by the operator from the data associating questions and answers as data to be used for questions and answers to be used in generating an answer model. The model generation unit 13 may acquire data associating questions and answers generated by an expert in the field of learning content.

[0066] When the data associating questions and answers is acquired, the model generation unit 13 analyzes sentences included in the data associating questions and answers by natural language processing and extracts features (step S42). When the features are extracted, the model generation unit 13 executes learning of the relationship between questions and answers and generates an answer model, which is a learning model that generates answers from questions (step S43). The model generation unit 13 generates the answer model by performing machine learning using, for example, a neural network.

[0067] When the answer model is generated, the model generation unit 13 stores data of the generated answer model in the storage unit 17 (step S45).

[0068] When the answer model is generated, the accuracy of the answer prediction is verified using test data. For example, the test data is data that associates questions and answers stored in the storage unit 17 and that was not used to generate the answer model. For example, if an information processing expert or an expert in the field of learning contents determines that the accuracy of the answer prediction is insufficient, the question and answer data is changed, or the conditions of the natural language processing are changed, and the answer model is generated again.

[0069] In the above description, answers to questions from learners are generated using an answer model generated by machine learning, but the response unit 12 may provide a preset answer for each question. A configuration in which a preset answer is provided for each question is also called a rule-based type.

[0070] The learning support system 10 of this embodiment can be used, for example, for sharing information within a company when introducing a new technology or a new system. The learning support system 10 may also be used for passing on knowledge and skills within a company. When sharing information or passing on knowledge within a company, people often receive coaching from people who are proficient in that field during training sessions or the like. In such cases, passing on knowledge and skills becomes easier by bringing together people who are acquiring knowledge or skills with people who already possess them.

[0071] The learning support system 10 of the learning system of this embodiment identifies the presence or absence of an answer model that matches the learner's attributes and the learning content when the learner selects the learning content. If a matching answer model is found, the learning support system 10 uses the answer model to generate an answer to a question sent by the learner who is studying the learning content, and outputs the generated answer to the learning terminal 20 that the learner is using for studying.

[0072] When there is no answer model that matches the attributes of the learner and the learning content, the learning support system 10 selects an answerer suitable for the learner and the learning content, and transfers questions and answers between the learning terminal 20 of the learner and the answerer terminal 30 of the answerer. In addition, the learning support system 10 associates and saves questions and answers transmitted and received between the learner and the answerer, and generates an answer model that generates answers from questions by machine learning that learns the relationship between questions and answers. Therefore, in learning content in a field in which answers by the answerer have been accumulated, the learning support system 10 can answer questions on behalf of the answerer by generating answers to questions using an answer model by machine learning. As a result, the learning system of this embodiment can respond to questions from the learner while suppressing dependency on the answerer. By reducing dependency on the answerer, it is possible to increase the opportunities for the learner to learn using the learning content while asking questions, even when there is no answerer available.

[0073] Second Embodiment The second embodiment of the present invention will be described in detail with reference to the drawings. Fig. 8 is a diagram showing an example of the configuration of a learning support system 100 of this embodiment. The learning support system 100 includes a response unit 101, a respondent determination unit 102, and an output unit 103.

[0074] The response unit 101 determines whether to answer questions from learners who are studying using the learning content by having an answerer answer the question, or to use an answer model that generates answers to the questions. When an answerer answers the question, the answerer determination unit 102 determines an answerer who will answer the question based on information on whether the answerer candidate can handle the learning content. When an answerer answers the question by using an answer model, the output unit 103 outputs answer data generated by the answer model using the question as input.

[0075] The response unit 12 of the learning support system 10 of the first embodiment is an example of the response unit 101. Moreover, the response unit 101 is one aspect of response means. The respondent determination unit 14 is an example of the respondent determination unit 102. Moreover, the respondent determination unit 102 is one aspect of respondent determination means. The output unit 19 is an example of the output unit 103. Moreover, the output unit 103 is one aspect of output means.

[0076] The operation of the learning support system 100 will be described. FIG. 9 is a diagram showing an example of the operation flow of the learning support system 100. The response unit 101 determines whether the answer to the question from the learner who is studying using the learning content will be answered by the answerer or by using an answer model that generates an answer to the question (step S101). If the answerer will answer the question (Yes in step S102), the answerer determination unit 102 determines the answerer who will answer the question based on the information on whether the answerer candidate can handle the learning content (step S103). If the answer is to be answered using the answer model (No in step S102), the output unit 103 outputs the answer data generated by the answer model using the question as input (step S104).

[0077] The learning support system 100 of this embodiment determines whether to use an answerer or an answer model, and if an answerer is to be used, determines the answerer based on information on whether the answerer can handle the learning content. Therefore, the learning support system 100 can answer questions from learners using either an answerer or an answer model, so that questions can be answered while reducing dependency on the answerer.

[0078] Each process in the learning support system 10 of the first embodiment and the learning support system 100 of the second embodiment can be performed by executing a computer program on a computer. Fig. 10 shows an example of the configuration of a computer 200 that executes a computer program that performs each process in the learning support system 10 of the first embodiment and the learning support system 100 of the second embodiment. The computer 200 includes a CPU 201, a memory 202, a storage device 203, an input / output I / F (Interface) 204, and a communication I / F 205.

[0079] CPU 201 reads out and executes computer programs for performing each process from storage device 203. CPU 201 may be configured by a combination of a CPU and a GPU (Graphics Processing Unit). Memory 202 is configured by a DRAM (Dynamic Random Access Memory) or the like, and temporarily stores computer programs executed by CPU 201 and data being processed. Storage device 203 stores computer programs executed by CPU 201. Storage device 203 is configured by, for example, a non-volatile semiconductor storage device. Other storage devices such as a hard disk drive may be used for storage device 203. Input / output I / F 204 is an interface that accepts input from a worker and outputs display data and the like. Communication I / F 205 is an interface that transmits and receives data to and from learning terminal 20. Moreover, learning terminal 20 and respondent terminal 30 may have a similar configuration.

[0080] The computer program used to execute each process may be stored in a recording medium and distributed. As the recording medium, for example, a magnetic disk such as a magnetic tape for recording data or a hard disk may be used. Also, an optical disk such as a CD-ROM (Compact Disc Read Only Memory) may be used as the recording medium. A non-volatile semiconductor memory device may also be used as the recording medium. [Explanation of symbols]

[0081] 10 Learning Support System 11 Learning Management Department 12 Response section 13 Model Generation Unit 14 Respondent Determination Department 15 Remuneration Calculation Department 16 Access Control Department 17 Memory section 18 Input section 19 Output section 20 Learning Devices 30 Respondent terminals 100 Learning Support System 101 Response section 102 Respondent Determination Department 103 Output section 200 Computers 201 CPU 202 Memory 203 Storage device 204 Input / Output Interface 205 Communication I / F

Claims

1. A response means for determining whether a response to a question from a learner who is learning using the learning content should be made by a respondent or by using an answer model that generates an answer to the question; an answerer determination means for determining an answerer who will answer the question based on information on whether the answerer candidate is compatible with the learning content when the answerer answers the question; an output means for outputting answer data generated by the answer model using the question as an input when the answer is given using the answer model; A learning support system equipped with:

2. 2. The learning support system according to claim 1, further comprising a reward calculation means for calculating a reward for the answerer based on at least one of an evaluation from the learner of the answer to the question and a track record of use of the answer model.

3. 3. The learning support system according to claim 1, wherein the output means outputs the same learning content to the terminal device of the learner and the terminal device of the respondent while the learner is studying the learning content.

4. a model generating means for generating the answer model by learning a relationship between a question by the learner and an answer by the answerer, 4. The learning support system according to claim 1, wherein the response means, when the answer model corresponding to the learning content has already been generated, provides an answer using the answer model.

5. 5. The learning support system according to claim 4, wherein the model generating means performs re-learning using data on the questions posed after the generation of the answer model and the answers given by the respondents, and updates the answer model.

6. determining whether to answer questions from a learner who is learning using the learning content by a respondent or by using an answer model that generates answers to the questions; When a respondent answers the question, the respondent who will answer the question is determined based on information on whether the candidate respondent is compatible with the learning content; When the answer is given using an answer model, the question is input and answer data generated by the answer model is output. Learning support methods.

7. The learning support method according to claim 6 , wherein a reward for the answerer is calculated based on at least one of an evaluation from the learner of the answer to the question and a track record of use of the answer model.

8. 8. The method of claim 6, wherein the same learning content is output to a terminal device of the learner and a terminal device of the respondent while the learner is studying the learning content.

9. generating the answer model by learning the relationship between the questions by the learners and the answers by the respondents; 9. The method of claim 6, further comprising the step of: providing an answer based on the answer model when the answer model corresponding to the learning content has already been generated.

10. A process of determining whether a question from a learner who is learning using the learning content is answered by a person who answers the question or by using an answer model that generates an answer to the question; A process of determining an answerer who will answer the question based on information on whether the answerer candidate is capable of responding to the learning content when the answerer answers the question; When the answer is given using an answer model, a process of inputting the question and outputting answer data generated by the answer model is performed. A learning support program that causes a computer to execute the above.

Images