Non-cognitive ability improvement assistance system and non-cognitive ability improvement assistance method

Abstract

Provided is a non-cognitive ability improvement assistance system capable of efficiently assisting in improving non-cognitive ability. A non-cognitive ability improvement assistance system (1) is characterized by being provided with: an emotion information generation unit that generates emotion information on the basis of biological information of a subject; a model input information generation unit that generates model input information for input to a base model (200) on the basis of the emotion information generated by the emotion information generation unit; a message generation unit that generates, on the basis of the model input information generated by the model input information generation unit, a message that resonates with the subject and improves the non-cognitive ability; and an output unit (15) that outputs the message generated by the message generation unit to the subject.

Description

Technical Field

[0001] The present disclosure relates to a non-cognitive ability improvement assisting system and a non-cognitive ability improvement assisting method for assisting in improving non-cognitive abilities. Background Art

[0002] In recent years, research on non-cognitive abilities has been gaining momentum not only in Japan but also worldwide, leading to increased recognition of their importance. While Japan places particular emphasis on passion, interest, and attention, cultivating resilience and a willingness to challenge, key elements of non-cognitive abilities, is less emphasized. There is less recognition that cognitive and non-cognitive abilities grow in tandem.

[0003] If you persevere with passion and focus, you'll naturally engage in deep thinking, research, and creativity, which will enhance your cognitive abilities. This development of cognitive abilities leads to a sense of accomplishment and fulfillment, while strengthening your non-cognitive abilities, such as the drive to "try harder next time." By recognizing this cycle, both cognitive and non-cognitive abilities can be effectively developed.

[0004] Such attitudes and efforts used to be easily considered as temperament and personality, but now they are considered as "skills" to emphasize the possibility of education. For example, the interest and attention of children can be intentionally improved by the environment created by educators. In addition, resilience can be developed through encouragement. By intentionally calling it a "skill", it means that children can do it through specific assistance. For example, a method of assisting in improving non-cognitive abilities by showing the completion of a learning plan has been proposed (see Patent Document 1).

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2016-218103 Summary of the Invention

[0008] Problems to be solved by the invention

[0009] On the other hand, when improving non-cognitive abilities, it is important to appropriately understand the subject's state.

[0010] The present disclosure aims to solve the above-mentioned problems and provides a non-cognitive ability improvement assisting system and a non-cognitive ability improvement assisting method that can efficiently assist in improving non-cognitive abilities.

[0011] Means for solving problems

[0012] The non-cognitive ability improvement assistance system of the first invention is characterized in that it has: an emotion information generating unit, which generates emotion information based on the biological information of the subject; a model input information generating unit, which generates model input information for input to the basic model based on the emotion information generated by the emotion information generating unit; a message generating unit, which generates a message for improving non-cognitive ability based on the model input information generated by the model input information generating unit, the message including a message that resonates with the subject; and an output unit, which outputs the message generated by the message generating unit to the subject.

[0013] The non-cognitive ability improvement support system of the second invention is based on the first invention, characterized in that the basic model is a large-scale language model, and the model input information is a prompt input to the large-scale language model.

[0014] The non-cognitive ability improvement assistance system of the third invention is based on the first invention or the second invention, and is characterized in that the non-cognitive ability improvement assistance system also has an input information acquisition unit, which acquires input information input by the subject into the device, and the model input information generation unit generates the model input information based on the emotion information generated by the emotion information generation unit and the input information acquired by the input information acquisition unit.

[0015] The non-cognitive ability improvement assistance system of the fourth invention is based on the third invention, and is characterized in that the non-cognitive ability improvement assistance system also has an input state analysis unit, which analyzes the input information acquired by the input information acquisition unit to generate input state information including input proficiency or concentration, and the model input information generation unit generates the model input information based on the emotion information generated by the emotion information generation unit and the input state information generated by the input state analysis unit.

[0016] The non-cognitive ability improvement assistance system of the fifth invention is based on the first invention or the second invention, and is characterized in that the message generating unit also generates useful information for improving non-cognitive ability based on the model input information generated by the model input information generating unit.

[0017] The non-cognitive ability improvement assistance system of the sixth invention is based on the first invention or the second invention, characterized in that the emotion information generating unit generates emotion information based on the biological information of the subject without inputting it into the device, and the output unit outputs the message to the subject in voice.

[0018] The non-cognitive ability improvement support system of the seventh invention is characterized in that, according to the first invention or the second invention, the non-cognitive ability improvement support system further includes a reaction evaluation unit that evaluates the reaction of the subject after the message is output.

[0019] The non-cognitive ability improvement support system according to the eighth invention is characterized in that, according to the first invention or the second invention, the output unit outputs a message together with a character corresponding to the emotion information.

[0020] The non-cognitive ability improvement assistance system of the 9th invention is based on the 4th invention, and is characterized in that the non-cognitive ability improvement assistance system also has: an evaluation unit, which evaluates the subject's effort level on the task based on the input status information; and an awarding unit, which provides rewards to the subject based on the evaluation results of the evaluation unit.

[0021] The non-cognitive ability improvement assistance method of the 10th invention is characterized in that it has the following steps: an emotion information generating step, generating emotion information based on the biological information of the subject; a model input information generating step, generating model input information for input to the basic model based on the emotion information generated in the emotion information generating step; a message generating step, generating a message for improving non-cognitive ability based on the model input information generated in the model input information generating step, the message including a message that resonates with the subject; and an output step, outputting the message generated in the message generating step.

[0022] Effects of the Invention

[0023] According to the non-cognitive ability improvement support system and non-cognitive ability improvement support method disclosed herein, a message that resonates with the subject's emotions can be output to the subject, thereby efficiently supporting the improvement of the subject's non-cognitive abilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a diagram for explaining the outline of the non-cognitive ability improvement support system according to the first embodiment.

[0025] Figure 2 This is a diagram for explaining the configuration of the non-cognitive ability improvement support system 1 according to the first embodiment.

[0026] Figure 3 This is a diagram illustrating the configuration of functional blocks of the information processing device 100 according to the first embodiment.

[0027] Figure 4 This is a flowchart illustrating the processing of the information processing device 100 according to the first embodiment.

[0028] Figure 5This is a diagram for explaining the subroutine flow of the reaction confirmation process in the first embodiment.

[0029] Figure 6 This is a diagram for explaining an example of the prompt words and answer sentences in the first embodiment.

[0030] Figure 7 This is a flowchart illustrating the processing of the information processing device 100 according to the second embodiment.

[0031] Figure 8 This is a flowchart for explaining the analysis process of the input state in the second embodiment.

[0032] Figure 9 This is a flowchart for explaining the analysis process of the input state in the third embodiment.

[0033] Figure 10 This is a diagram for explaining an outline of a non-cognitive ability improvement support system according to a third embodiment. DETAILED DESCRIPTION

[0034] Hereinafter, regarding the embodiment of the present disclosure, the non-cognitive ability improvement support system 1 will be described in detail with reference to the drawings. In the following, the same or corresponding parts in the drawings are denoted by the same reference numerals, and their description will not be repeated in principle.

[0035] (Implementation 1: Non-cognitive ability improvement support system 1)

[0036] Figure 1 This is a diagram for explaining the outline of the non-cognitive ability improvement support system 1 according to the first embodiment. Figure 1 The non-cognitive ability improvement support system 1 according to the first embodiment includes a terminal 10 , a network NW, an information processing device 100 , and a basic model 200 .

[0037] The purpose of the non-cognitive ability improvement support system 1 is to improve the non-cognitive abilities of a subject (user U) by communicating with the subject via a terminal 10. The communication performed by the non-cognitive ability improvement support system 1 includes text communication by outputting text data, audio communication by outputting audio data, video communication by outputting image data or moving image data, and video communication by outputting a combination of audio data and image data or moving image data.

[0038] Here, non-cognitive abilities refer to inherent abilities that are difficult to quantify in intelligence tests or academic ability tests. Specifically, they refer to abilities related to a person's personality and sociality, such as drive, endurance, coordination, and self-control. If non-cognitive abilities decrease, it becomes difficult for the user U to sustain their activities, potentially leading to their cessation. Therefore, improving the user U's non-cognitive abilities is desirable because it allows them to sustain their activities and improves their activity outcomes.

[0039] For example, the non-cognitive ability improvement support system 1 generates information for input to the basic model 200 based on information indicating the emotions of user U. The basic model 200 then generates relevant messages for improving user U's non-cognitive abilities, including messages that resonate with user U, and outputs the generated relevant messages to user U. In this case, messages that resonate closely with user U's emotions can be output to user U. More specifically, by providing responses corresponding to the information indicating user U's emotions, continuous communication with user U about content that resonates with user U's emotions can be achieved. This effectively supports improving user U's non-cognitive abilities.

[0040] <User U>

[0041] The user U is a user of the non-cognitive ability improvement assisting system 1. The user U includes a person who explicitly or implicitly desires to improve his or her non-cognitive ability by communicating with the non-cognitive ability improvement assisting system 1.

[0042] For example, when the user U realizes that his or her non-cognitive ability has decreased, the user U can explicitly request the non-cognitive ability improvement support system 1 to communicate for the purpose of improving the non-cognitive ability.

[0043] Furthermore, when user U, for example, is unaware that his or her non-cognitive abilities have decreased, and the non-cognitive ability improvement assistance system 1 believes that user U implicitly requests communication for the purpose of improving non-cognitive abilities based on information related to user U actively or passively obtained by the non-cognitive ability improvement assistance system 1, the user starts communicating with the non-cognitive ability improvement assistance system 1 even without a clear request.

[0044] User U may be, for example, a person who requires a communication partner or a person who requires a personal assistant. User U may also be, for example, a learner, including a student in a school, a student in a cram school, a participant in a seminar / training / e-learning, etc. User U may also be, for example, a trainee who receives self-reliance training (life training) or motor function training such as rehabilitation training. User U may also be, for example, a trainee who receives operation training such as an operation simulator for e-sports or a vehicle, etc. User U may also be, for example, a person who is negotiating with his or her future career or a negotiator who requires work counseling. User U may also be a creator of art works such as graphic art, or music, etc., by operating the terminal 10, or may be a creator of art works such as paintings, sculptures, dancing works, dance works, singing works, etc., without using the terminal 10.

[0045] exist Figure 1In the example of , the user U is a learner who learns programming using the terminal 10. FIG. 1 shows a case where a learning screen 50 for learning programming is displayed on the terminal 10, and a character 52 is provided on the learning screen 50 as an example.

[0046] <Base Model 200>

[0047] The basic model 200 is a well-known AI model that is pre-trained using large-scale learning data and generates appropriate responses when various data are input. Figure 1 As shown, it includes a large language model 200 a (LLM) and a generative AI system that generates data based on input text data.

[0048] The foundation model 200 may, for example, include a large multi modal foundation model (LMM), which generates multimodal data responses in response to input of various data types (multimodal data), such as sound data, image data, dynamic image data, and other sensor data that cannot be captured by the five senses of humans, in addition to text data containing prompt words and queries.

[0049] In addition, text data includes language information such as articles and characters.

[0050] Furthermore, the voice data includes, in addition to the above-mentioned language information, peripheral language information (paralinguistic information) such as the pitch and intonation of the user U's voice that supplements the language information.

[0051] In addition to the above-mentioned language information, the image data also includes: image information such as graphics, body movement information including expressions or gestures, physical characteristic information including the user U's age, gender, physique or chronic diseases, spatial information including the distance or position relationship between the user U and the non-cognitive ability improvement assistance system 1, and environmental information including the usage environment of the non-cognitive ability improvement assistance system 1.

[0052] Furthermore, the dynamic image data is composed of a combination of multiple pieces of information in a continuous relationship among the various types of information included in the aforementioned image data. Dynamic image data includes, for example, change information (elapsed information) having a higher amount of information than the image data, and operation information including the details of operations performed on the non-cognitive ability improvement support system 1. Dynamic image data may include a combination of change information or operation information with various types of information included in the audio data.

[0053] Furthermore, the sensor data includes the temperature, humidity, air pressure, pulse, brainwave, body temperature, facial expression, speed or acceleration of movement of the user U, and the like in the use environment of the non-cognitive ability improvement support system 1 .

[0054] The basic model 200 includes, for example, a question receiving unit 201, an answer generating unit 202, and a database (DB) 204. The question receiving unit 201 receives a question from the information processing device 100 regarding information related to non-cognitive abilities for improving non-cognitive abilities. The answer generating unit 202 analyzes the received message and, based on the analysis results, generates an answer to the question containing information related to non-cognitive abilities by referring to the database 204 and transmits the answer to the information processing device 100. The database 204 contains a large amount of data used to generate the answer.

[0055] Here, the questions received by the base model 200 include not only inquiries to the base model 200 but also user U requests such as instructions or commands to the base model 200. Furthermore, the answers generated by the base model 200 include not only answers to the questions posed to the base model 200 but also answers to input content containing user U requests such as instructions or commands to the base model 200.

[0056] Furthermore, if the basic model 200 pre-stores the results of training (pre-learning) using a large amount of data, it is not necessary to refer to the database 204. In this case, upon receiving an input of a query via the question receiving unit 201, the basic model 200 generates an answer to the query based on the analysis results, including a message related to non-cognitive abilities, without referring to the database 204, and transmits the answer to the query to the information processing device 100.

[0057] Large-Scale Language Model (LLM) 200a

[0058] LLM 200a is a natural language processing system that performs question answering. LLM 200a is a natural language processing model trained using a large amount of text data. It takes text data containing articles or characters as input and outputs text data containing articles or characters. In this case, database 204 uses, for example, an article database containing a large amount of text data.

[0059] When LLM 200a is used in a communication system for answering questions or conducting conversations based on language communication or the like with a user U, when text data is input to LLM 200a, LLM 200a outputs a response to the text data. The information processing device 100 of this embodiment queries LLM 200a for information related to non-cognitive abilities for improving non-cognitive abilities and transmits the response from LLM 200a to terminal 10.

[0060] <<Generate System AI>>

[0061] For example, a generation system AI may be used as the base model 200. The generation system AI includes image generation AI, dynamic image generation AI, and sound generation AI.

[0062] When an image generation AI is used in a communication system for answering questions or conducting conversations with user U, inputting text data to the image generation AI generates image data. Furthermore, when dynamic image generation AI is used, inputting text data to the dynamic image generation AI generates dynamic image data. In this case, database 204 utilizes, for example, a vector database that stores a large amount of text data, image data, and dynamic image data in the form of mathematical expressions.

[0063] Furthermore, when voice-generating AI is used in conjunction with LLM 200a, when text data converted from voice input received from user U is input to LLM 200a, the text data is output from LLM 200a, and the output text data is converted into voice data and then output. In this case, database 204 may use, for example, a voice database containing a large amount of text and voice data. By utilizing voice-generating AI, non-cognitive ability improvement support system 1 enables voice communication between user U and non-cognitive ability improvement support system 1.

[0064] In this case, the information processing device 100 of this embodiment inquires the generation system AI about information related to non-cognitive abilities for improving non-cognitive abilities, and transmits the response from the generation system AI to the terminal 10 .

[0065] <<Multimodal Basic Model>>

[0066] For example, a multimodal base model may be used as the base model 200. When a multimodal base model is applied to a communication system that conducts question-answering or conversation with a user U, the multimodal base model implements at least one of active acquisition of multimodal data from the user U and passive acquisition based on receiving input of multimodal data from the user U, and outputs multimodal data corresponding to the data acquired by these methods.

[0067] In this case, the information processing device 100 of the present embodiment inquires the multimodal base model about information related to non-cognitive abilities for improving non-cognitive abilities, and outputs the answer from the multimodal base model to the terminal 10 .

[0068] Terminal 10

[0069] Terminal 10 is a device for receiving messages related to non-cognitive abilities for improving non-cognitive abilities from base model 200. During program learning, terminal 10 outputs the received messages related to non-cognitive abilities for improving user U's non-cognitive abilities on learning screen 50 of terminal 10 by, for example, outputting voice messages or displaying speech bubbles. For example, terminal 10 may use character 52 to output these messages.

[0070] Terminal 10 is a personal computer (PC) or the like owned by user U and may be portable or stationary. Terminal 10 is configured to communicate with information processing device 100 via network NW. Information processing device 100 is configured to communicate with base model 200 via network NW. Communication may be wireless or wired.

[0071] Figure 2 This is a diagram illustrating the structure of the non-cognitive ability improvement support system 1 according to the first embodiment. Figure 2 As shown, terminal 10 includes a control unit 11, a camera 12, a communication interface 13, an input unit 14, an output unit 15, a microphone 16, a storage unit 17, and an internal bus connecting these components. Control unit 11 includes a CPU (Central Processing Unit), RAM (Random Access Memory), and ROM (Read Only Memory).

[0072] The camera 12 acquires image data (image information) of the user U's facial expression as biometric data (biometric information) of the user U (subject). In this example, image data of the user U's facial expression and other aspects are described, but the present invention is not limited thereto and other information may be acquired as biometric data. For example, an infrared camera may be used to acquire pulse information of the user U.

[0073] The communication I / F 13 is connected to the network NW and transmits and receives data (information) to and from an external device.

[0074] The input unit 14 receives input data (input information) from the user U to the terminal 10. The input unit 14 includes, for example, a well-known mouse and keyboard. Alternatively, the input unit 14 may utilize, for example, a voice recognition module to convert the voice input from the user U into text data before receiving the input. Alternatively, the input unit 14 may receive voice data from the user U as input data via, for example, the microphone 16 or a well-known microphone different from the microphone 16.

[0075] The output unit 15 outputs a message generated by the information processing device 100 to the user U. The output unit 15 includes a display device such as a display that outputs text data, image data, moving image data, etc., and a playback device such as a speaker that outputs audio data, etc.

[0076] For example, the output unit 15 outputs a message together with the character 52 corresponding to the information indicating the emotion of the user U. In this case, it is possible to output information that contains a large amount of information and has a high influence on the emotion of the user U. This can more efficiently assist in improving the non-cognitive abilities of the user U.

[0077] The microphone 16 acquires, for example, voice data (voice information) of the user U as the biological data of the user U. The microphone 16 may acquire, for example, voice data of the user U as the input data of the user U.

[0078] The storage unit 17 includes various application programs, etc. For example, an application program for the user U to learn programming is stored.

[0079] The control unit 11 generates a learning screen 50 for program learning by executing this application program, and displays it on a display device such as a monitor.

[0080] The terminal 10 acquires biometric data (for example, image data and audio data) and transmits it to the information processing device 100 via the communication I / F 13. The terminal 10 acquires input data and transmits it to the information processing device 100 via the communication I / F 13. For example, the input data in this example includes, in addition to data such as the content of operations input by the user U using a keyboard, mouse, etc., used for program learning, data such as the content of the user U's speech input using the microphone 16, and data such as user U's feature values, such as images or dynamic images containing the user U's facial expressions and body movements.

[0081] <Information Processing Device 100>

[0082] The information processing device 100 communicates with the basic model 200 and receives a message related to the non-cognitive ability for improving the non-cognitive ability of the user U.

[0083] For example, Figure 2As shown, the information processing device 100 includes a control unit 101, a storage unit 107, a communication I / F 103, and an internal bus connecting these components. The control unit 101 includes a CPU, RAM, and ROM. The communication I / F 103 is connected to the network NW and performs data transmission and reception with external devices. The storage unit 107 includes various application programs, etc. For example, the storage unit 107 stores application programs for assisting in improving non-cognitive abilities, etc. The control unit 101 implements various processes by executing these application programs. The information processing device 100 receives biometric data, etc. sent from the terminal 10, and generates emotion information of the user U based on the biometric data, etc. The information processing device 100 inquires the basic model 200 about the message related to non-cognitive abilities corresponding to the generated emotion information.

[0084] Information processing device 100 receives messages related to non-cognitive abilities from basic model 200 and transmits them to terminal 10. During program learning, terminal 10 outputs the received messages related to non-cognitive abilities on learning screen 50 of terminal 10. For example, terminal 10 may output these messages using character 52 configured on learning screen 50. While this example describes a configuration in which basic model 200 is provided independently of information processing device 100, it is also possible to provide functionality equivalent to basic model 200 within information processing device 100.

[0085] Figure 3 This is a diagram illustrating the functional block configuration of the information processing device 100 according to the first embodiment. Figure 3 The control unit 101 of the information processing device 100 implements various functional blocks by executing application programs stored in the storage unit 107 .

[0086] Specifically, the information processing device 100 includes a biological information acquisition unit 110, an emotion information generation unit (emotion analysis unit) 112, an input information acquisition unit 114, a model input information generation unit 116, an input state analysis unit 117, a message generation unit (output control unit) 118, an evaluation unit 120, an assignment unit 122 and a reaction evaluation unit 124.

[0087] <<Biological Information Acquisition Unit 110>>

[0088] The biometric information acquisition unit 110 acquires biometric data (image data, audio data, etc.) transmitted from the terminal 10 and received via the communication I / F 103 . The acquired biometric data is stored in the storage unit 107 .

[0089] <<Emotion Information Generating Unit (Emotion Analyzing Unit) 112>>

[0090] The emotion information generation unit 112 infers the user U's emotions (psychology) based on the acquired biometric data (image data, audio data, etc.) and generates emotion information. The emotion information generation unit 112 infers (analyzes) emotions such as joy, anger, sadness, and happiness based on the biometric data. Emotions are not limited to these, and the emotion information generation unit 112 can also infer calmness, surprise, satisfaction, boredom, frustration, fear, peace of mind, uneasiness, etc. Specifically, the emotion information generation unit 112 can use the acquired image data to process the user U's facial expressions, including the eyelid opening, line of sight, eyebrow movement, the presence or absence of nasal wrinkles, mouth movement, mouth opening, pupil opening, etc., to make inferences. The emotion information generation unit 112 can also use audio data, such as the content of sounds, sighs, breathing sounds, etc. emitted by the user U, to infer (analyze) emotions. In addition, the emotion information generation unit 112 can infer emotions using only one type of data or a combination of data.

[0091] <<Input Information Acquisition Unit 114>>

[0092] The input information acquisition unit 114 acquires input data (input information) transmitted from the terminal 10 and received via the communication I / F 103. The input data acquired by the input information acquisition unit 114 is stored in the storage unit 107.

[0093] Here, the input data (input information) includes the content of data input by the user U to the non-cognitive ability improvement support system 1 , input feature quantities such as input speed and the number of input errors.

[0094] The input information acquisition unit 114 can, for example, obtain text data obtained via a keyboard connected to the terminal 10 as input data, or obtain text data or sound data obtained via the microphone 16 as input data, or obtain text data, sound data, image data or dynamic image data obtained via the camera 12 as input data.

[0095] <<Model input information generation unit 116>>

[0096] The model input information generator 116 generates model input information using the non-cognitive ability-related message corresponding to the emotion information generated by the emotion information generator 112 as data (information) input to the base model 200. The model input information generator 116 inputs the generated data to the base model 200, for example.

[0097] The data input to base model 200 includes any of the aforementioned text data, audio data, image data, dynamic image data, and sensor data, or multimodal data. The data input to base model 200 is data containing instructions, commands, inquiries, and the like to base model 200, and may include prompt words or queries as text data.

[0098] Here, when LLM 200a is applied to base model 200, model input information generation unit 116 generates prompt words as model input information. In this case, model input information and messages related to non-cognitive abilities can be reliably obtained using language information, making it easier for user U to understand the messages. This improves convenience for user U.

[0099] The model input information generating unit 116 generates data (information) to be input to the base model 200 in order to inquire about a message related to non-cognitive abilities corresponding to the emotion information generated by the emotion information generating unit 112 , for example.

[0100] Specifically, the model input information generation unit 116 generates model input information for input to the basic model 200 based on the emotion information generated by the emotion information generation unit 112 and the input information acquired by the input information acquisition unit 114. In this case, the output content and timing of messages that resonate with the emotions of user U can be optimized based on the input content and input situation of user U. This can more effectively assist in improving user U's non-cognitive abilities.

[0101] Furthermore, the model input information generation unit 116 can generate model input information for input to the basic model 200 based on the emotion information generated by the emotion information generation unit 112 and the input state information generated by the input state analysis unit 117 based on the input information acquired by the input information acquisition unit 114. In this case, the output content and timing of messages that resonate with the user U's emotions can be optimized based on the user U's input proficiency and concentration. This can more effectively assist in improving the user U's non-cognitive abilities.

[0102] The model input information generating unit 116 may be generated, for example, by referring to a pre-stored database composed of a plurality of data sets that have emotion information as input and output model input information, thereby obtaining data to be input to the basic model 200 for querying the message related to non-cognitive abilities corresponding to the emotion information generated by the emotion information generating unit 112. Furthermore, the model input information generating unit 116 may be generated, for example, by inputting the emotion information generated by the emotion information generating unit 112 to a learning model that has been previously learned using a learning data set that has emotion information as input and output model input information, thereby outputting data to be input to the basic model 200 for querying the message related to non-cognitive abilities corresponding to the emotion information.

[0103] <<Input state analysis unit 117>>

[0104] The input state analysis unit 117 analyzes the input state of the user U based on the acquired input data. Specifically, the input state analysis unit 117 analyzes the input speed of the user U's input data, the number of input errors, and the like to estimate the user U's input state.

[0105] <<Message Generation Unit (Output Control Unit) 118>>

[0106] The message generator 118 generates a message for improving non-cognitive abilities based on the information generated by the model input information generator 116, including a message that resonates with the user U. For example, the message generator 118 generates the message based on information output from the base model 200 to which the information generated by the model input information generator 116 is input. For example, the message generator 118 may input the information generated by the model input information generator 116 into the base model 200 instead of having the model input information generator 116 input the information.

[0107] The message generation unit 118 outputs information such as the answer from the base model 200 to the terminal 10. The message generation unit 118 outputs the message generated by the message generation unit 118 to the terminal 10 based on information such as the answer from the base model 200, for example.

[0108] <<Evaluation Department 120>>

[0109] The evaluation unit 120 evaluates the degree of effort of the user U in performing a task. An example of the task of the user U is program learning. Details of the evaluation will be described later.

[0110] <<Giving section 122>>

[0111] The granting unit 122 provides rewards to user U based on the evaluation results of the evaluation unit 120. For example, if the non-cognitive ability improvement support system 1 incorporates a paid service, the reward is points that can be used for that service. In this case, further improvement of user U's non-cognitive abilities can be promoted independently of interaction with the non-cognitive ability improvement support system 1. This allows for more efficient support in improving user U's non-cognitive abilities.

[0112] <<Reaction Evaluation Unit 124>>

[0113] The reaction evaluation unit 124 evaluates the reaction of the user U to the message output related to the non-cognitive ability for improving the non-cognitive ability of the user U. The reaction evaluation unit 124 adjusts various parameters based on the evaluation result of the user U's reaction.

[0114] The reaction evaluation unit 124 adjusts the parameters based on the emotion information generated by the emotion information generation unit 112. For example, if the reaction evaluation unit 124 infers positive emotions such as joy and happiness based on the emotion information, it increases the output frequency of messages related to non-cognitive abilities. On the other hand, if the reaction evaluation unit 124 infers negative emotions such as anger and sadness based on the emotion information, it increases the interval between predetermined periods and decreases the output frequency of messages related to non-cognitive abilities.

[0115] The reaction evaluation unit 124, for example, evaluates the reaction of user U after the message is output. In this case, a message that resonates with user U's emotions can be output to user U based on the user's emotional changes after interacting with the non-cognitive ability improvement support system 1. This allows for more efficient support in improving user U's non-cognitive abilities.

[0116] (Embodiment 1: Operation Method of Non-cognitive Ability Improvement Support System 1)

[0117] Next, referring to the accompanying drawings, an example of the operation of a non-cognitive ability improvement support system 1 will be described as a non-cognitive ability improvement support method according to Embodiment 1. The operation of the non-cognitive ability improvement support system 1 includes, for example, an emotion information generation step, a model input information generation step, a message generation step, and an output step.

[0118] <Emotional information generation steps>

[0119] Figure 4 This is a flowchart illustrating the processing of the information processing device 100 according to the first embodiment. Figure 4In the emotion information generation step, the information processing device 100 acquires the biometric data (biometric information) of the user U (step S0). Specifically, the biometric information acquisition unit 110 acquires the biometric data transmitted from the terminal 10. Prior to step S0, the biometric data of the user U has been acquired in advance by various components of the terminal 10 (such as the camera 12, the input unit 14, and the microphone 16).

[0120] Next, the information processing device 100 determines whether a predetermined period has elapsed (step S2). If the information processing device 100 determines that the predetermined period has not elapsed ("No" in step S2), it returns to step S0 and repeats the above process. The predetermined period can be set to any period, and as an example, it can be set to 5 minutes.

[0121] When information processing apparatus 100 determines that the predetermined period has elapsed (YES in step S2 ), it proceeds to the next process.

[0122] The information processing device 100 performs emotion analysis (step S4). Specifically, the emotion information generation unit 112 infers the emotions (psychology) of the user U based on the biometric data (image data, audio data, etc.) stored in the storage unit 107 after a predetermined period has elapsed, and generates emotion information. Furthermore, a portion or all of the biometric data (image data, audio data, etc.) stored in the storage unit 107 during the predetermined period may be used.

[0123] For example, the information processing device 100 may end the emotion information generating step when the emotion information is generated.

[0124] <Model input information generation steps>

[0125] Next, in the model input information generation step, information processing device 100 generates model input information to be sent to base model 200 (step S6). Specifically, model input information generation unit 116 generates data for inputting a message related to non-cognitive abilities corresponding to the emotion information generated by emotion information generation unit 112 into base model 200.

[0126] Next, the information processing device 100 transmits the generated model input information to the base model 200 (step S8 ). Specifically, the model input information generating unit 116 transmits the generated model input information, which is data input to the base model 200 .

[0127] The information processing device 100 ends the model input information generation step when, for example, the model input information is generated, the model input information is transmitted to the base model 200 , or the model input information is input to the base model 200 .

[0128] Message Generation Steps

[0129] Next, in the message generation step, the information processing device 100 determines whether a reply message has been received from the base model 200 (step S10 ). Specifically, the message generation unit 118 determines whether a reply message has been received from the base model 200 .

[0130] In step S10 , the information processing device 100 maintains this state until a reply message is received from the base model 200 .

[0131] For example, when the information processing device 100 receives a reply message from the base model 200 ("Yes" in step S10), it can set parameters for outputting the reply message based on the emotion information (step S12). Specifically, the message generation unit 118 can set parameters for outputting the reply message based on the emotion information. The details of setting the message output parameters will be described later.

[0132] The information processing device 100 ends the message generation step when, for example, a reply message is received from the base model 200 or when the output content or form of the reply message received from the base model 200 is adjusted according to a preset format.

[0133] Output steps

[0134] Next, the information processing device 100 outputs the message output information to the terminal 10 (step S14). Specifically, the message generation unit 118 transmits the message output information, including the reply message, to the terminal 10. If the message generation unit 118 receives multiple reply messages from the base model 200, it can select one of them. If the message generation unit 118 receives multiple reply messages, it can also generate a reply message by appropriately combining them or extracting only necessary portions and then processing them appropriately.

[0135] When a message containing a reply is input from the information processing device 100, the terminal 10 outputs the input message to the user U via the output unit 15. In this case, by providing a reply corresponding to information indicating the user U's emotions, continuous communication with the user U can be achieved, which resonates with the user U's emotions. This effectively assists in improving the non-cognitive abilities of the user U.

[0136] Specifically, the operation of the non-cognitive ability improvement support system 1 includes the following steps: an emotion information generation step for generating emotion information based on the biological information of user U; a model input information generation step for generating model input information for input to the basic model 200 based on the emotion information generated in the emotion information generation step; a message generation step for generating a message for improving non-cognitive abilities based on the model input information generated in the model input information generation step, the message including a message that resonates with user U; and an output step for outputting the message generated in the message generation step to user U. In this case, a message that resonates with user U's emotions can be output to user U. This effectively supports the improvement of user U's non-cognitive abilities.

[0137] For example, when the information processing device 100 and the terminal 10 have output a message to the user U, they may end the output step.

[0138] <Post-processing of output step>

[0139] Next, the information processing device 100 may execute a process for confirming the reaction of the user U (reaction confirmation process) (step S16). Specifically, the information processing device 100 may execute a process for confirming the reaction of the user U to the message output, for example. The details of this process are described in "Details of Reaction Confirmation Processing" below.

[0140] Next, the information processing device 100 determines whether the process of assisting in improving non-cognitive abilities has been completed (step S18 ).

[0141] In step S18, if the information processing device 100 determines that the process of assisting in improving non-cognitive abilities has ended, the process ends (End). Specifically, if the information processing device 100 determines that the user U has ended the programming learning application on the terminal 10, the process ends. For example, the information processing device 100 may end the process if it receives a command from the terminal 10 to end the programming learning application.

[0142] On the other hand, in step S18 , when the information processing apparatus 100 determines that the process of assisting improvement of non-cognitive abilities is not to be terminated (“No” in step S18 ), the process returns to step S0 and repeats the above-described process.

[0143] The above steps complete the operation of the non-cognitive ability improvement support system 1. The non-cognitive ability improvement support system 1 may also omit the parameter setting for executing message output (step S12) and the response confirmation process (step S16).

[0144] <Details of Reaction Confirmation Process (Reaction Evaluation Step)>

[0145] Figure 5 This is a diagram illustrating a subroutine flow of the reaction confirmation process (reaction evaluation step) of the first embodiment. Figure 5 The information processing device 100 acquires the biometric data from the terminal 10 after the message is output (step S20). Specifically, the biometric information acquisition unit 110 acquires the biometric data transmitted from the terminal 10 after the message is output.

[0146] Next, the information processing device 100 performs an emotion analysis process (step S22). Specifically, the emotion information generation unit 112 infers the emotions (psychology) of the user U based on the biometric data (image data, audio data, etc.) stored in the storage unit 107 after the message is output, and generates emotion information. The emotion information generation unit 112 can newly acquire biometric data of the user U who reacts to the message after the message is output, or the biometric data of the user U who responds to the message, and then infer the emotions (psychology) of the user U based on this biometric data to generate the emotion information.

[0147] Next, the information processing device 100 performs parameter adjustment processing (step S24). The reaction evaluation unit 124 performs parameter adjustment processing based on the emotion information generated by the emotion information generation unit 112. In this example, the reaction evaluation unit 124 adjusts the interval of the predetermined period in the processing in step S2 based on the generated emotion information. For example, if the reaction evaluation unit 124 infers positive emotions such as joy and happiness based on the emotion information, it shortens the interval of the predetermined period and increases the output frequency of messages related to non-cognitive abilities. On the other hand, if the reaction evaluation unit 124 infers negative emotions such as anger and sadness based on the emotion information, it lengthens the interval of the predetermined period and decreases the output frequency of messages related to non-cognitive abilities. By adjusting the output frequency of messages based on the emotional state of the user U, it is possible to assist in improving non-cognitive abilities. Alternatively, for example, the reaction evaluation unit 124 may only shorten the interval of the predetermined period when the emotion information infers positive emotions such as joy and happiness, and may not change the interval of the predetermined period when the emotion information infers negative emotions such as anger and sadness, or may perform the opposite process.

[0148] Then, the information processing device 100 ends the response confirmation process (return).

[0149] <Example of model input information and responses when using LLM 200a>

[0150] Figure 6This figure illustrates an example of model input information and responses in Implementation 1. In the following description, the case of using LLM 200a is taken as an example, but the same content can also be applied to model input information for generating system AI or multimodal base models. Figure 6 (A) shows the prompt words generated by the model input information generation unit 116 based on the emotion information, and the prompt words serve as text data to be sent to the LLM 200a.

[0151] As an example, the model input information generation unit 116 generates the following prompt words based on the emotional information ("happy"). Specifically, as an example, the model input information generation unit 116 generates a sentence such as "(A) For children in the following emotional states, please think about a conversation that improves the child's non-cognitive abilities. When talking, please use expressions that are specific to the child. I am learning programming. (B) Learning makes me happy. (C) Sometimes I don't understand and I feel a little anxious. (D) Giving up makes me feel regretful."

[0152] Here, the "child" included in (A) in the text is based on, for example, the physical characteristic information of user U pre-stored in the various components of the non-cognitive ability improvement support system 1, or the physical characteristic information of user U obtained by input unit 14. Furthermore, "programming learning" is based on, for example, the user U's intended use of the non-cognitive ability improvement support system 1, pre-stored in the various components of the non-cognitive ability improvement support system 1, or the user U's operation information on the non-cognitive ability improvement support system 1 obtained by input unit 14. Furthermore, "Please think about a conversation that improves non-cognitive abilities" means that a message intended to improve non-cognitive abilities is generated, and this message includes a message that resonates with user U.

[0153] Furthermore, information indicating emotions such as “happy,” “anxious,” and “regretful” included in (B), (C), and (D) in the passage is based on the emotion information of the user U generated by the emotion information generation unit 112 , for example.

[0154] exist Figure 6 (B) shows an example of a reply from LLM 200a to the above-mentioned prompt word. As an example, the information processing device 100 receives a reply such as "Great! Programming is fun, right? However, sometimes there are difficulties, so you might get a little anxious. Even if you don't understand something, it's an opportunity to keep learning! Everyone has a lot of things they don't understand at the beginning. Let's work together to solve that difficulty! If you do this, there may be new and interesting things waiting for you!"

[0155] The information processing device 100 transmits the received reply text to the terminal 10 .

[0156] The terminal 10 outputs a learning screen 50 for program learning, along with the response text sent from the information processing device 100, i.e., a message related to non-cognitive abilities for improving the non-cognitive abilities of the user U. Through this processing, the non-cognitive ability improvement support system 1 analyzes the emotional state of the user U and outputs a resonant message that closely matches the analyzed emotional information, thereby improving and strengthening the non-cognitive abilities. In other words, the non-cognitive ability improvement support system 1 can appropriately understand the state of the user U, effectively assisting in improving non-cognitive abilities.

[0157] <Example of generating prompt words>

[0158] In the above description, the model input information generation unit 116 generates a prompt word based on the emotion information ("happy"). The model input information generation unit 116 can also generate a prompt word for other emotion information in the same manner.

[0159] For example, the model input information generation unit 116 generates a prompt word based on the emotional information ("anger"). For example, the model input information generation unit 116 can change the article (B) among the articles (A) to (D) mentioned above as the prompt word to "angry about learning". In addition, the model input information generation unit 116 can fix the article (A) and appropriately change the articles (B) to (D) based on the emotional information. As an example, the information processing device 100 receives a reply article such as "Anger is an emotion that occurs when learning new things. If you move forward step by step, that anger will gradually disappear. Don't miss out on your growth." for the prompt word. The information processing device 100 sends the received reply article to the terminal 10.

[0160] For example, the model input information generation unit 116 generates a prompt word based on the emotional information ("sadness"). For example, the model input information generation unit 116 can change the article (B) among the articles (A) to (D) mentioned above as the prompt word to "feeling sad about learning". In addition, the model input information generation unit 116 can fix the article (A) and appropriately change the articles (B) to (D) based on the emotional information. As an example, the information processing device 100 receives a reply article such as "Try to regard sad feelings as a step towards growth. It is natural to have things you don't understand, but it is important to work hard to overcome them." The information processing device 100 sends the received reply article to the terminal 10.

[0161] For example, the model input information generation unit 116 generates a prompt word based on the emotional information ("joy"). For example, the model input information generation unit 116 can change the article (B) among the articles (A) to (D) mentioned above as the prompt word to "feeling joy in learning." In addition, the model input information generation unit 116 can fix the article (A) and appropriately change the articles (B) to (D) based on the emotional information. As an example, the information processing device 100 receives a reply article such as "Your enthusiasm for learning is great. You can continue to grow by feeling joy and moving forward." for the prompt word. The information processing device 100 sends the received reply article to the terminal 10.

[0162] <Setting message output parameters>

[0163] The message generation unit 118 sets the parameters of the message output of the reply text according to the emotional information. For example, the message generation unit 118 can set the parameters of the color when the message is output according to the emotional information. For example, the message generation unit 118 can set the parameters of the text color such as orange, red, blue, and yellow according to the emotional information of "joy", "anger", "sadness", and "happiness" to output the message. The message generation unit 118 uses the psychological effect of color to output the message, thereby outputting a resonant message that is close to the analyzed emotional information to the user U with the help of color, thereby improving and strengthening the non-cognitive ability. That is, it is possible to appropriately grasp the status of the user U as the subject and efficiently assist in improving the non-cognitive ability. In addition, the setting of the parameters of the text color is an example and is not limited to this. It can also be set to other colors, or the brightness or other parameters can be changed according to the emotional information to output the message.

[0164] Furthermore, when using the character 52 to output a message, the message generation unit 118 can also adjust the character 52's movements based on the emotional information. For example, multiple movement patterns for the character 52 can be prepared in advance, and the message generation unit 118 can select one of the multiple movement patterns based on the emotional information and display it on the terminal 10. For example, movement patterns corresponding to "joy," "anger," "sadness," and "happiness" can be prepared in advance, and when setting the parameters for message output, the message generation unit 118 sets the multiple movement patterns to the one that matches the emotional information. For example, when outputting a message, the non-cognitive ability improvement assistance system 1 displays the character 52's movements that match the emotional information. This outputs a message that resonates closely with the analyzed emotional information, along with the character 52's movements, to the user U, thereby improving and strengthening non-cognitive abilities. In other words, the non-cognitive ability improvement assistance system 1 can appropriately grasp the state of the user U as the subject and efficiently assist in improving non-cognitive abilities.

[0165] Furthermore, when the terminal 10 uses the character 52 to output a message, a speech bubble display may be used to show that the character 52 is speaking, or the message may be output in the form of sound.

[0166] The message generation unit 118 can adjust the mode of outputting the message in the form of sound based on the emotional information. For example, a plurality of sound output modes may be prepared in advance, and the message generation unit 118 may select one of the plurality of sound output modes for output based on the emotional information. For example, sound output modes corresponding to "joy," "anger," "sadness," and "happiness" may be prepared in advance, and when setting the parameters for outputting the message in the form of sound, the message generation unit 118 may set the mode that matches the emotional information among the plurality of sound output modes. For example, when outputting a message in the form of sound, the non-cognitive ability improvement assistance system 1 outputs it in a sound output mode that matches the emotional information, thereby outputting a resonant sound message that closely matches the analyzed emotional information to the user U, thereby improving and strengthening the non-cognitive ability. That is, the non-cognitive ability improvement assistance system 1 can appropriately grasp the status of the user U as the subject and efficiently assist in improving the non-cognitive ability.

[0167] In addition, the message generating unit 118 may use the above parameter setting method alone or in combination. Figure 4 In the process of the embodiment of the present invention, the method of setting the parameters of the message output based on the emotion information in step S12 is described, but this method is not a necessary structure, and a method without this setting process can also be used. Figure 4 The method of performing the reaction confirmation process in step S16 in the process is described, but this method is not a necessary structure, and a method without the reaction confirmation process can also be used.

[0168] (Embodiment 2: Operation Method of Non-cognitive Ability Improvement Support System 1)

[0169] Figure 7 This is a flowchart for explaining the processing of the information processing device 100 according to the second embodiment. Figure 7 ,and Figure 4 Compared with the flowchart of , the processing of the information processing device 100 of embodiment 2 is different in that: step S1, step S5A and step S5B are added; and step S6 is replaced by step S6#. The other structures are the same as those described above, so their detailed description will not be repeated.

[0170] The operation of the non-cognitive ability improvement support system 1 may further include, for example, an input information acquisition step. The operation of the non-cognitive ability improvement support system 1 may further include, for example, an input state analysis step.

[0171] <Input information acquisition steps>

[0172] After acquiring the biometric data in step S0, the information processing device 100 acquires input data in the input information acquisition step (step S1). Specifically, the input information acquisition unit 114 acquires the input data (input information) of the user U transmitted from the terminal 10. The input data of the user U is acquired in advance by various components of the terminal 10 (such as the camera 12, the input unit 14, and the microphone 16) prior to step S0.

[0173] Next, the information processing device 100 determines whether a predetermined period has elapsed (step S2). If the information processing device 100 determines that the predetermined period has not elapsed ("No" in step S2), it returns to step S0 and repeats the above process. The predetermined period can be set to any period, and as an example, it can be set to 5 minutes.

[0174] When information processing apparatus 100 determines that the predetermined period has elapsed (YES in step S2 ), it proceeds to the next process.

[0175] <Input status analysis steps>

[0176] The information processing device 100 may perform emotion analysis after a predetermined period has elapsed in step S4, and then perform input state analysis in the input state analysis step (step S5A). Specifically, the input state analysis unit 117 analyzes the input state of the user U based on the input data stored in the storage unit 107. Furthermore, the input state analysis unit 117 may utilize a portion of the input data stored in the storage unit 107 during the predetermined period, or may utilize all of the data.

[0177] Figure 8 This is a flowchart for explaining the analysis process of the input state in the second embodiment. Figure 8 The input state analysis unit 117 evaluates the input proficiency based on the acquired input data (step S30). Specifically, the input state analysis unit 117 analyzes the input speed and number of input errors of the user U using the keyboard as input data, and infers the input proficiency of the user U.

[0178] The input state analysis unit 117 can infer whether the input proficiency is high or low and classify it, and can further classify it into multiple levels. In this example, the information on the input proficiency can be used when generating the prompt word.

[0179] Next, the input state analysis unit 117 evaluates the concentration based on the acquired input data (step S32 ). Specifically, the input state analysis unit 117 analyzes the length of the input period of the user U using a keyboard or the like as input data information to evaluate the concentration of the user U.

[0180] For example, the input state analysis unit 117 may infer that the concentration of the user U is high when the input period in a certain period is long. Alternatively, the input state analysis unit 117 may infer that the concentration is high when the input period in a certain period is equal to or greater than a certain threshold, and infer that the concentration is not high in other cases.

[0181] Next, the input state analysis unit 117 evaluates the degree of completion of the task based on the input data obtained (step S34). Specifically, the input state analysis unit 117 evaluates the degree of completion of the programming learning provided as a task based on the input data as information of the input data. Specifically, the input state analysis unit 117 compares the correct answer information of the programming learning provided as a task with the input information input by the current user U to evaluate the degree of completion of the programming learning. The evaluation unit 120 evaluates the degree of effort for the task based on the information on the degree of completion of the programming learning, which will be described later. In addition, the evaluation unit 120 may also evaluate the degree of effort by considering the input period and input proficiency of the programming learning. Then, the input state analysis unit 117 ends the analysis processing of the input state (return).

[0182] Refer again Figure 7 The information processing device 100 determines whether the concentration is high as the analyzed input state information (step S5B). Specifically, the model input information generating unit 116 determines whether the concentration is high as the analysis result of the input state analyzing unit 117.

[0183] In step S5B, when the information processing device 100 determines that the concentration is not high as the analyzed input state (No in step S5B), it ends the input information analysis step and proceeds to the model input information generation step (step S6#).

[0184] <Model input information generation steps>

[0185] Next, the information processing device 100 generates model input information to be sent to the basic model 200 (step S6#). Specifically, when it is determined that the concentration is not high as the analysis result of the input state analysis unit 117, the model input information generation unit 116 generates data related to non-cognitive ability input to the basic model 200 corresponding to the emotion information generated by the emotion information generation unit 112 and the information on input proficiency. Specifically, when generating the model input information based on the emotion information, the model input information generation unit 116 includes articles related to input proficiency. For example, in addition to Figure 6 In addition to the articles (A) to (D) described in the text, information related to the input status of the user U may also include articles such as "(E) Input proficiency is high" or "(E) Input proficiency is low". Figure 4 The processing is the same as that described in the flowchart.

[0186] Specifically, the non-cognitive ability improvement support system 1 includes a model input information generation step that generates model input information for input to the basic model 200 based on the emotion information generated in the emotion information generation step and the input information acquired in the input information acquisition step. In this case, the output content and timing of messages that resonate with the user U's emotions can be optimized based on the user U's input content and input situation. This allows for more efficient support in improving the user U's non-cognitive abilities.

[0187] On the other hand, in step S5B, if the information processing device 100 determines that the analyzed input state is high ("Yes" in step S5B), steps S6# through S16 are skipped and the process proceeds to step S18. Specifically, if the input state analysis unit 117 determines that the concentration is high, the model input information generation unit 116 does not generate prompt words. In other words, if the concentration is high, the information processing device 100 does not output a message to the terminal 10. The information processing device 100 evaluates the concentration based on the acquired input data and does not output a message if it infers a high concentration.

[0188] Message output during periods of high concentration can hinder the thinking of the user U. Therefore, the non-cognitive ability improvement support system 1 stops outputting messages in this state, thereby improving and strengthening non-cognitive abilities. In other words, the non-cognitive ability improvement support system 1 can appropriately grasp the state of the user U, effectively supporting the improvement of non-cognitive abilities.

[0189] In addition, in this example, the input state analysis unit 117 is described as determining the concentration based on the input data. However, this is not limited to this, and the concentration can also be determined in combination with emotional information. In addition, in this example, the non-cognitive ability improvement support system 1 is described as determining whether to output a message based on the concentration of user U. However, the output frequency of messages related to non-cognitive abilities can also be adjusted based on the concentration of user U. For example, if the concentration of user U is high, the non-cognitive ability improvement support system 1 can extend the interval of the specified period of processing in step S2. By adjusting the output frequency of messages based on the concentration of user U, the non-cognitive ability improvement support system 1 can assist in improving non-cognitive abilities.

[0190] After executing the above steps, the operation of the non-cognitive ability improvement support system 1 is completed.

[0191] In addition, the non-cognitive ability improvement support system 1 can also omit the above-mentioned input information analysis step (step S5A, step S5B). In this case, the model input information generation unit 116 may not analyze the input data of the user U in the model input information generation step, but instead generate the content of "happiness", "anxiety", "regret" and so on contained in the input data of the user U. Figure 6 Multimodal data related to (A) to (D) described in .

[0192] <Rewards>

[0193] The evaluation unit 120 evaluates the user U's effort in learning the program as a task. The evaluation unit 120 evaluates the effort based on the task completion degree analyzed by the input state analysis unit 117. The evaluation unit 120 may also evaluate the effort by considering the input period and input proficiency of the programming learning.

[0194] For example, the evaluation unit 120 classifies the user U's effort in learning programming into a plurality of states. For example, the evaluation unit 120 classifies the user U's effort as "excellent" when it is very good. The evaluation unit 120 may classify the task completion as "100%" as "excellent". The evaluation unit 120 classifies the user U's effort as "good" when it is relatively good. The evaluation unit 120 may classify the task completion as "more than 90%" as "good". The evaluation unit 120 classifies the user U's effort as ordinary when it is ordinary. The evaluation unit 120 may classify the task completion as "less than 90%" as "good". In this example, the classification into three levels is described, but it can also be further classified into multiple levels.

[0195] For example, the granting unit 122 grants a reward (for example, points) to the user U whose effort in learning programming is classified and evaluated as “excellent”.

[0196] Regarding the rewards (points), when a charging system is installed and used as a programming learning service, part or all of the rewards (points) can be used to use the service. Alternatively, for example, the rewards (points) can be used to purchase items (clothing, hats, etc.) for the character 52 or to display other characters.

[0197] In the non-cognitive ability improvement support system 1, if the user U's status is classified as "excellent," indicating a very high level of effort in learning, rewards can be given to improve and strengthen their non-cognitive abilities. In other words, the non-cognitive ability improvement support system 1 can appropriately understand the status of the user U and efficiently support the improvement of their non-cognitive abilities.

[0198] In addition, the rewards (points) granted by the non-cognitive ability improvement assistance system are not limited to points, but can also be props, etc., or a mechanism such as rewards provided by affiliated companies (cooperative enterprises, supporting enterprises) of the non-cognitive ability improvement assistance system.

[0199] (Implementation 3: Non-cognitive ability improvement support system 1)

[0200] The non-cognitive ability improvement support system 1 of the third embodiment is different from the above embodiment in that it is applied to users U other than learners. The other configurations are the same as those described above, and therefore detailed descriptions thereof will not be repeated.

[0201] <User U who requests a personal assistant>

[0202] The user U is a person who uses the non-cognitive ability enhancement assistance system 1 as a personal assistant.

[0203] User U chats with the non-cognitive ability improvement support system 1, generating input data via the model input information generation unit 116 and inputting it into the basic model 200. The non-cognitive ability improvement support system 1 then generates a message for improving the non-cognitive ability of user U via the message generation unit 118, and outputs the message to user U via the output unit 15.

[0204] For example, if user U's concentration is low or if user U is determined to be in a negative mood, the non-cognitive ability improvement support system 1 can generate model input information based on user U's emotional information without waiting for user U to speak. It can then generate a message based on the model input information and output the message to user U via the output unit 15. In this case, compared to a case where user U's speech is used as a trigger for message output, bidirectional communication can be achieved. This allows for more efficient support in improving user U's non-cognitive abilities.

[0205] For example, Figure 9 As shown, the non-cognitive ability improvement support system 1 also generates useful information for improving the non-cognitive abilities of user U, via the message generation unit 118, based on the information output from the base model 200, which has been fed with information generated by the model input information generation unit 116. The non-cognitive ability improvement support system 1 then outputs the message and useful information to user U, for example, via the output unit 15 (step S14#). In this case, a more sympathetic response that aligns with the emotions of user U can be provided with a higher level of information than a single message. This allows for more efficient assistance in improving the non-cognitive abilities of user U.

[0206] For example, you can also Figure 10 As shown, the non-cognitive ability improvement assistance system 1 generates emotion information based on the biological information of the user U who does not input the terminal 10 (device) via the emotion information generation unit 112, and outputs a message to the user U via the output unit 15. That is, the user U can use the non-cognitive ability improvement assistance system 1 even when performing operations using the object T at a location far away from the terminal 10, for example. In this case, answers that resonate with the emotions of the user U can be provided to a wider variety of users U. Thus, the non-cognitive ability of the user U can be assisted and improved more efficiently. In addition, since the input of the terminal 10 is not required, one-to-many communication can be achieved between one non-cognitive ability improvement assistance system 1 and multiple users U. Thus, the non-cognitive ability of the user U can be assisted and improved more efficiently.

[0207] Figure 10The example shown in FIG. 1 shows an example in which a user U performs an operation using a device or paper other than the terminal 10 within a space R where the terminal 10 is installed and within a range where the output sound of the non-cognitive ability improvement assistance system 1 can be heard. In addition, as long as the user U can input input to the non-cognitive ability improvement assistance system 1 and output messages and useful information from the non-cognitive ability improvement assistance system 1 via an earphone-microphone, a headset, a wireless camera, etc. that is wirelessly connected to the non-cognitive ability improvement assistance system 1, the user U can use the non-cognitive ability improvement assistance system 1 in a position in the space R where the output sound from the terminal 10 cannot be heard or outside the space R.

[0208] Here, useful information refers to, for example, information that can achieve the purpose of improving the non-cognitive ability of the user U. Specifically, useful information includes content information that can improve the motivation of the user U and put him in a positive mood.

[0209] The method for generating content information can be arbitrarily selected based on, for example, pre-stored preferences of the user U, or emotional information generated through the aforementioned reaction confirmation process for the user U who reacts to the content information. Furthermore, the model input information generating unit 116 may generate model input information and input it to the base model 200. The model input information may include at least one of the following: user U's physical characteristic information (e.g., "children"), emotional information (e.g., "feeling down"), a goal of improving non-cognitive abilities (e.g., "desire to become more positive"), and the content of useful information required to improve non-cognitive abilities (e.g., "movie," "music," "artwork," etc.), thereby outputting content information from the base model 200.

[0210] The useful information may be in any form as long as it can be output to the user U, and may be in any form such as text data (title of work, URL of the web page where the work is posted), sound data, image data, or moving image data.

[0211] Furthermore, the useful information can also be in the form of information that helps user U improve their non-cognitive abilities. Specifically, even if the goal of improving user U's non-cognitive abilities cannot be achieved simply by disclosing useful information such as the URL of a webpage that publishes packaging images, thumbnail images, and commercial clips for a film, music video, or the title and artist name of a work of art, improvement in non-cognitive abilities can be achieved by allowing user U to create opportunities to appreciate the film, music, or art based on the content of the useful information. In this case, a self-treatment method can be provided that allows user U to improve their non-cognitive abilities even when they are not using the non-cognitive ability improvement support system 1. This allows for more efficient assistance in improving user U's non-cognitive abilities.

[0212] <Users of self-reliance training>

[0213] The user U is a person who is undergoing self-reliance training such as rehabilitation training and requests to communicate with the non-cognitive ability improvement support system 1 during or after training. The non-cognitive ability improvement support system 1 outputs a message to the user U who is undergoing self-reliance training or outputs a message together with useful information.

[0214] Useful information in this case refers to, for example, information that can achieve the purpose of improving the non-cognitive ability of the user U who is an independent trainer. Specifically, it includes relaxation technique information (such as rubbing and kneading) that can alleviate the pain associated with the user U's independent training.

[0215] As a method for generating relaxation technique information, for example, it can be arbitrarily selected based on pre-stored chronic diseases, symptoms, rehabilitation training plans, etc. of the user U. Furthermore, the model input information generating unit 116 may generate model input information and input it to the basic model 200. The model input information may include, for example, at least one of the following: physical characteristic information of the user U ("child," "foot injury," etc.), emotional information ("rehabilitation training is painful," etc.), the purpose of improving non-cognitive abilities ("hope to recover to be able to run," etc.), and the content of useful information requested for improving the user's non-cognitive abilities ("methods to relieve foot pain," etc.), thereby outputting relaxation technique information from the basic model 200.

[0216] <Users U who are undergoing operation training>

[0217] The user U is a person who is training for e-sports and requests to communicate with the non-cognitive ability improvement support system 1 during or after training. The non-cognitive ability improvement support system 1 outputs a message to the user U who is training, or outputs a message together with useful information.

[0218] Useful information in this case refers, for example, to information that can achieve the purpose of improving the non-cognitive abilities of user U as an operation trainee. Specifically, it includes entertainment information (information about other trainees, movies and music works that change the mood, etc.) used to eliminate fatigue and stress caused by operation training to user U or promote operation improvement.

[0219] The method for generating entertainment information can be arbitrarily selected based on, for example, the pre-stored preferences of the user U, the content of the operation training, etc. Furthermore, the model input information generating unit 116 may generate model input information and input it to the basic model 200. The model input information may include, for example, at least one of the following: the user U's physical characteristic information (e.g., "20 years old"), emotional information (e.g., "unable to operate the game well"), the user's goal of improving non-cognitive abilities (e.g., "hoping to be able to operate well"), and the content of the information required to improve the user's non-cognitive abilities (e.g., "information about e-sports players undergoing operation training"), thereby outputting entertainment information from the basic model 200.

[0220] <Users participating in training>

[0221] The user U is a person who participates in training such as e-learning and requests to communicate with the non-cognitive ability improvement support system 1 during or after the training. The non-cognitive ability improvement support system 1 outputs a message to the user U who is a training participant, or outputs a message together with useful information.

[0222] Useful information in this case refers to, for example, information that can achieve the purpose of improving the non-cognitive abilities of user U as a training participant. Specifically, it includes career information (career paths and career development blueprints related to training content, feedback related to training results, etc.) used to eliminate user U's anxieties and doubts related to training.

[0223] The method for generating occupational information can be arbitrarily selected based on, for example, the hobbies and training contents of the user U stored in advance. Furthermore, the occupational information can be output from the base model 200 by generating model input information via the model input information generating unit 116 and inputting it into the base model 200. The model input information includes, for example, at least one of the following: the user U's physical characteristic information (e.g., "20 years old"), emotional information (e.g., "worried about whether I can practice the training content"), the user's goal of improving non-cognitive abilities (e.g., "hoping to be able to complete a series of tasks independently as quickly as possible"), and the user's request for useful information for improving their non-cognitive abilities (e.g., "hoping for feedback on training results," "hoping to know the career paths of those participating in the training," etc.).

[0224] <User U who performs creative work>

[0225] The user U is a person who creates artworks and requests communication with the non-cognitive ability improvement support system 1 during or after creation. The non-cognitive ability improvement support system 1 outputs a message to the user U who creates artworks, or outputs a message together with useful information.

[0226] Useful information in this case refers to, for example, information that can achieve the purpose of improving the non-cognitive ability of user U as a creator of artistic works. Specifically, it includes artistic information used to eliminate the reduction in user U's creativity or promote the improvement of user U's creativity (information about other creators related to artistic works, creativity, other artistic works used to change moods, etc.).

[0227] The method for generating art information can be arbitrarily selected based on, for example, the preferences and creative content of the user U stored in advance. Furthermore, the model input information generating unit 116 generates model input information and inputs it to the base model 200. This model input information includes, for example, at least one of the following: the user U's physical characteristic information (e.g., "age 20"), emotional information (e.g., "worry about stopping creativity"), the purpose of improving non-cognitive abilities (e.g., "hope to eliminate the decline in creativity"), and the content of the information requested for useful information to improve the user's non-cognitive abilities (e.g., "hope to know other companies' artworks that serve as inspiration for their creations"), thereby outputting art information from the base model 200.

[0228] As described above, the non-cognitive ability improvement support system 1 can output messages or a combination of messages and useful information that resonate with the emotions of various users U. This can effectively support the improvement of the non-cognitive abilities of users U.

[0229] As mentioned above, although this disclosure was specifically described based on embodiment, this disclosure is not limited to embodiment, and various changes can be made without departing from the scope of the present disclosure.

[0230] Label Description

[0231] 1: Non-cognitive ability improvement assistance system; 10: Terminal; 11, 101: Control unit; 12: Camera; 13, 103: Communication I / F; 14: Input unit; 15: Output unit; 16: Microphone; 17, 107: Storage unit; 50: Learning screen; 52: Character; 100: Information processing device; 110: Biological information acquisition unit; 112: Emotion information generation unit (emotion analysis unit); 114: Input information acquisition unit; 116: Model input information generation unit; 117: Input state analysis unit; 118: Message generation unit (output control unit); 120: Evaluation unit; 122: Assignment unit; 124: Reaction evaluation unit; 200: Basic model; 200a: Large-scale language model; 201: Question receiving unit; 202: Answer generation unit; 204: Database.

Claims

1. A non-cognitive ability improvement assistance system, characterized in that: The non-cognitive ability improvement assistance system has: an emotion information generating unit that generates emotion information based on biological information of a subject; a model input information generating unit configured to generate model input information for input to a basic model based on the emotion information generated by the emotion information generating unit; a message generating unit configured to generate a message for improving non-cognitive ability based on the model input information generated by the model input information generating unit, the message including a message that resonates with the subject; as well as An output unit outputs the message generated by the message generating unit to the target person.

2. The non-cognitive ability improvement assistance system according to claim 1, characterized in that: The base model is a large-scale language model, The model input information is the prompt word input to the large-scale language model.

3. The non-cognitive ability improvement support system according to claim 1 or 2, characterized in that: The non-cognitive ability improvement support system further includes an input information acquisition unit that acquires input information input by the subject to the device. The model input information generating unit generates the model input information based on the emotion information generated by the emotion information generating unit and the input information acquired by the input information acquiring unit.

4. The non-cognitive ability improvement assistance system according to claim 3, characterized in that: The non-cognitive ability improvement support system further includes an input state analysis unit that analyzes the input information acquired by the input information acquisition unit to generate input state information including input proficiency or concentration. The model input information generating unit generates the model input information based on the emotion information generated by the emotion information generating unit and the input state information generated by the input state analyzing unit.

5. The non-cognitive ability improvement support system according to claim 1 or 2, characterized in that: The message generating unit further generates useful information for improving non-cognitive ability based on the model input information generated by the model input information generating unit.

6. The non-cognitive ability improvement support system according to claim 1 or 2, characterized in that: The emotion information generating unit generates emotion information based on the biological information of the subject who does not input the device, The output unit outputs the message to the subject person in voice.

7. The non-cognitive ability improvement support system according to claim 1 or 2, characterized in that: The non-cognitive ability improvement support system further includes a reaction evaluation unit that evaluates a reaction of the subject after the message is output.

8. The non-cognitive ability improvement support system according to claim 1 or 2, characterized in that: The output unit outputs the message together with a character corresponding to the emotion information.

9. The non-cognitive ability improvement assistance system according to claim 4, characterized in that: The non-cognitive ability improvement assistance system further comprises: an evaluation unit that evaluates the subject's effort level for the task based on the input state information; and A granting unit provides an incentive to the subject based on the evaluation result of the evaluation unit.

10. A method for assisting in improving non-cognitive abilities, characterized in that: The non-cognitive ability improvement auxiliary method has the following steps: an emotion information generating step of generating emotion information based on biological information of the subject; a model input information generating step of generating model input information for input to a basic model based on the emotion information generated in the emotion information generating step; a message generating step of generating a message for improving non-cognitive ability based on the model input information generated in the model input information generating step, the message including a message that resonates with the subject; as well as An output step of outputting the message generated in the message generating step.

Images