System

Abstract

An object of a system according to an embodiment is to improve consistency and reliability of a response to generated AI.SOLUTION: A system includes a prompt generation unit, an answer collection unit, a question generation unit, and a stabilization unit. The prompt generation unit receives an input from a user. The answer collection unit makes a question to the plurality of generated AI based on the prompt generated by the prompt generation unit, and collects an answer. The question generation unit generates an advanced question based on the answer collected by the answer collection unit. The stabilization unit asks a question to the plurality of generation AI again based on the question generated by the question generation unit, and repeats this operation until the answer reaches a predetermined reference.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] With conventional technology, the answers given by the generating AI were sometimes inconsistent, making it difficult to obtain reliable information.

[0005] The system according to the embodiment aims to improve the consistency and reliability of the answers of the generation AI. [Means for solving the problem]

[0006] The system according to the embodiment includes a prompt generation unit, an answer collection unit, a question generation unit, and a stabilization unit. The prompt generation unit receives input from a user. The answer collection unit asks questions to multiple generation AIs based on prompts generated by the prompt generation unit and collects answers. The question generation unit generates advanced questions based on the answers collected by the answer collection unit. The stabilization unit again asks questions to the multiple generation AIs based on the questions generated by the question generation unit, and repeats this process until the answers reach a certain standard. [Effects of the Invention]

[0007] The system according to the embodiment can improve the consistency and reliability of the answers of the generation AI. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE INVENTION

[0009] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), and Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A system according to an embodiment of the present invention combines multi-turn and multi-LLM to maximize the capabilities of a generative AI. This system accepts user input, extracts prerequisite knowledge from the generative AI, asks the same question to multiple generative AIs, collects responses, generates more advanced questions, and asks them again, repeating this process until the answers stabilize. For example, when a user inputs a prompt such as "Please tell me some basic information about a specific technology," the generative AI generates an answer. The system then asks multiple generative AIs a question such as "Please tell me some application examples of this technology," each of which generates an answer and collects the answers. Based on the collected answers, the system generates a more advanced question such as "What is the most effective application example of this technology?" and asks the question again to multiple generative AIs, collecting their answers. This process is repeated until multiple generative AIs repeatedly give the same answer to the same question, at which point the system determines that the answer has stabilized. This allows the system to maximize the capabilities of the generative AI and obtain reliable information. For example, generative AI can be used to efficiently collect and analyze information in the research and development of new technologies.

[0029] An information processing system according to an embodiment includes a prompt generation unit, an answer collection unit, a question generation unit, and a stabilization unit. The prompt generation unit receives input from a user. For example, the prompt generation unit can receive text or voice input from a user and generate it as a prompt. The prompt generation unit can also use a generation AI to generate prompts to elicit knowledge that is the basis of a conversation. For example, the generation AI generates a prompt such as, "Please tell me some basic information about a specific technology." The answer collection unit asks multiple generation AIs questions based on the prompt generated by the prompt generation unit and collects their answers. For example, the answer collection unit asks multiple generation AIs a question such as, "Please tell me some application examples of this technology," and each generation AI generates an answer and collects the answers. The question generation unit generates an advanced question based on the collected answers. For example, the question generation unit generates an advanced question such as, "Among the application examples of this technology, what is the most effective one?" based on the collected answers. The stabilization unit again asks multiple generation AIs questions based on the question generated by the question generation unit until the answers are stabilized. For example, the stabilization unit repeats a question until multiple generation AIs repeatedly give the same answer to the same question. This allows the information processing system according to the embodiment to maximize the capabilities of the generation AI and obtain highly reliable information. For example, the stabilization unit determines that the generation AI's capabilities have been fully utilized when the answers become stable.

[0030] The prompt generation unit can generate prompts to elicit knowledge that is a prerequisite for the conversation. The prompt generation unit generates prompts to elicit knowledge that is a prerequisite for the conversation, for example, by using a generation AI. For example, the prompt generation unit generates a prompt such as, "Please tell me some basic information about a specific technology." The prompt generation unit can also generate an appropriate prompt based on user input. For example, if the user inputs, "I would like to know about application examples of a new technology," the prompt generation unit generates a prompt such as, "Please tell me some application examples of this technology." In this way, by generating a prompt that elicits knowledge that is a prerequisite for the conversation, preparations for the dialogue are completed.

[0031] The answer collection unit can ask the same question to multiple generation AIs and collect their answers. For example, the answer collection unit asks the same question to multiple generation AIs and collects their answers. For example, the answer collection unit asks multiple generation AIs a question such as "Please tell me some application examples of this technology," and each generation AI generates an answer, which is then collected. The answer collection unit can also organize the collected answers and use them to generate the next question. For example, the answer collection unit organizes the collected answers by category and provides them to the question generation unit. In this way, by collecting answers from multiple generation AIs, information from various perspectives can be obtained.

[0032] The question generation unit can generate advanced questions based on the collected answers. For example, the question generation unit generates advanced questions based on the collected answers. For example, the question generation unit generates advanced questions such as "What is the most effective application example of this technology?" based on the collected answers. The question generation unit can also analyze the collected answers using a generation AI to generate appropriate questions. For example, the question generation unit generates advanced questions by analyzing the collected answers using a generation AI and extracting important information. In this way, by generating advanced questions based on the collected answers, deeper information can be elicited.

[0033] The stabilization unit can repeatedly ask questions to multiple generation AIs until the answers become stable. For example, the stabilization unit repeats questions to multiple generation AIs until the answers become stable. For example, the stabilization unit repeats questions until multiple generation AIs repeatedly give the same answer to the same question. The stabilization unit can also use the generation AI to evaluate the stability of the answers. For example, the stabilization unit evaluates the consistency and reliability of the answers and repeats the questions until a stable answer is obtained by the generation AI. In this way, by repeating the questions until the answers become stable, highly reliable information can be obtained.

[0034] The stabilization unit can determine that the generation AI's capabilities have been fully utilized when the answer becomes stable. For example, the stabilization unit determines that the generation AI's capabilities have been fully utilized when the answer becomes stable. For example, the stabilization unit determines that the answer has stabilized if multiple generation AIs begin to repeatedly give the same answer to the same question. The stabilization unit can also use the generation AI to evaluate the stability of the answer. For example, the stabilization unit evaluates the consistency and reliability of the answer and determines that the generation AI's capabilities have been fully utilized when a stable answer is obtained. In this way, optimal information can be obtained by determining that the generation AI's capabilities have been fully utilized when the answer becomes stable.

[0035] The prompt generation unit can analyze the user's past input history and generate an appropriate prompt. The prompt generation unit, for example, analyzes the user's past input history and generates an appropriate prompt. For example, the prompt generation unit generates a relevant prompt based on questions or themes that the user has frequently input in the past. The prompt generation unit can also extract specific patterns from the user's past input history and customize prompts based on those patterns. For example, the prompt generation unit generates a familiar prompt by referring to the language and expressions used by the user in the past. In this way, by analyzing the user's past input history, it is possible to generate more relevant prompts.

[0036] The prompt generation unit can customize prompts based on the user's current areas of interest or projects. The prompt generation unit customizes prompts based on, for example, the user's current areas of interest or projects. For example, the prompt generation unit generates specific prompts based on information related to a project the user is currently working on. The prompt generation unit can also analyze the user's recent search history or browsing history to generate prompts that are aligned with the user's themes of interest. For example, the prompt generation unit generates highly relevant prompts by referring to topics in communities or forums in which the user participates. This enables more specific interactions by customizing prompts based on the user's current areas of interest or projects.

[0037] The prompt generation unit can select an appropriate prompt format depending on the user's input method. For example, the prompt generation unit selects an appropriate prompt format depending on the user's input method. For example, if the user is using voice input, the prompt generation unit generates a prompt suitable for voice recognition. Furthermore, if the user is using text input, the prompt generation unit can also generate a concise and easy-to-read prompt. Furthermore, if the user is using image input, the prompt generation unit can also generate a prompt based on image analysis. In this way, by selecting the optimal prompt format depending on the user's input method, more user-friendly interactions are possible.

[0038] The prompt generation unit can generate a highly relevant prompt based on the user's geographical location information. The prompt generation unit generates a highly relevant prompt based on, for example, the user's geographical location information. For example, if the user is in a specific area, the prompt generation unit generates a prompt including information related to the area. Furthermore, if the user is traveling, the prompt generation unit can generate a prompt related to the travel destination. Furthermore, if the user is participating in a specific event, the prompt generation unit can generate a prompt related to the event. In this way, by taking the user's geographical location information into consideration, more relevant prompts can be generated.

[0039] The prompt generation unit can analyze the user's social media activity and generate relevant prompts. The prompt generation unit, for example, analyzes the user's social media activity and generates relevant prompts. For example, the prompt generation unit generates relevant prompts based on content shared by the user on social media. The prompt generation unit can also generate interesting prompts by referring to the activities of the user's friends on social media. The prompt generation unit can also analyze the content posted by accounts the user follows and generate relevant prompts. In this way, more relevant prompts can be generated by analyzing the user's social media activity.

[0040] The prompt generation unit can customize the prompt content by reflecting the user's past feedback. The prompt generation unit customizes the prompt content by reflecting the user's past feedback, for example. For example, the prompt generation unit generates an improved prompt based on feedback provided by the user in the past. Furthermore, if the user gives a high rating to a specific prompt, the prompt generation unit can generate a similar prompt. Furthermore, the prompt generation unit can avoid prompts that the user has previously expressed dissatisfaction with and generate more appropriate prompts. In this way, more appropriate prompts can be generated by reflecting the user's past feedback.

[0041] The answer collection unit can analyze the generation AI's past answer history and select an appropriate collection method. The answer collection unit, for example, analyzes the generation AI's past answer history and selects an appropriate collection method. For example, the answer collection unit prioritizes collecting highly reliable answers from the generation AI's past answer history. The answer collection unit can also analyze the generation AI's past answer patterns and select an efficient collection method. The answer collection unit can also prioritize collecting answers related to specific topics based on the generation AI's past answer history. In this way, more reliable answers can be collected by analyzing the generation AI's past answer history.

[0042] The answer collection unit can customize the collection means based on the characteristics and performance of the generation AI. The answer collection unit customizes the collection means based on, for example, the characteristics and performance of the generation AI. For example, the answer collection unit preferentially collects answers from high-performance generation AI. The answer collection unit can also preferentially collect answers from generation AIs that are strong in a particular field. The answer collection unit can also adjust the collection means according to the characteristics of the generation AI to obtain the optimal answer. In this way, by customizing the collection means based on the characteristics and performance of the generation AI, it is possible to obtain a more optimal answer.

[0043] The answer collection unit can evaluate the reliability of the collected answers and preferentially collect highly reliable answers. The answer collection unit, for example, evaluates the reliability of the collected answers and preferentially collects highly reliable answers. For example, the answer collection unit evaluates the reliability of the collected answers and preferentially collects highly rated answers. The answer collection unit can also check the source of the collected answers and preferentially collect answers from highly reliable sources. The answer collection unit can also analyze the content of the collected answers and preferentially collect answers that include highly reliable information. In this way, by evaluating the reliability of the collected answers, more reliable information can be obtained.

[0044] The answer collection unit can prioritize collecting highly relevant answers by taking into account the geographical distribution of the generation AI. The answer collection unit, for example, prioritizes collecting highly relevant answers by taking into account the geographical distribution of the generation AI. For example, if the generation AI has information related to a specific region, the answer collection unit prioritizes collecting answers related to that region. The answer collection unit can also analyze the geographical distribution of the generation AI and prioritize collecting answers from generation AIs that are strong in a specific region. The answer collection unit can also prioritize collecting answers that are in line with the characteristics of each region by taking into account the geographical distribution of the generation AI. In this way, by taking into account the geographical distribution of the generation AI, more relevant answers can be collected.

[0045] The answer collection unit can improve the accuracy of collection by referring to literature related to the generation AI. The answer collection unit can improve the accuracy of collection by, for example, referring to literature related to the generation AI. For example, the answer collection unit collects highly accurate answers based on the related literature referred to by the generation AI. The answer collection unit can also analyze literature related to the generation AI and preferentially collect answers that include highly reliable information. The answer collection unit can also collect detailed answers on a specific topic by referring to literature related to the generation AI. In this way, by referring to literature related to the generation AI, more accurate answers can be collected.

[0046] The response collection unit can perform collection based on the market value of the generating AI. The response collection unit performs collection based on, for example, the market value of the generating AI. For example, the response collection unit preferentially collects responses from generating AIs with high market value. The response collection unit can also analyze the market value of the generating AI and preferentially collect responses from generating AIs that are strong in a particular field. The response collection unit can also adjust the collection means, taking into account the market value of the generating AI, to obtain optimal responses. In this way, by taking into account the market value of the generating AI, it is possible to collect more valuable responses.

[0047] The question generation unit can adjust the level of detail of the question based on the importance of the collected answers. The question generation unit adjusts the level of detail of the question based on, for example, the importance of the collected answers. For example, the question generation unit generates a detailed question based on an answer with a high level of importance. The question generation unit can also generate a concise question based on an answer with a low level of importance. The question generation unit can also evaluate the importance of the collected answers and generate a question with an appropriate level of detail. In this way, by adjusting the level of detail of the question based on the importance of the collected answers, it is possible to generate a more appropriate question.

[0048] The question generation unit can apply different question algorithms depending on the category of the collected answers. For example, the question generation unit applies different question algorithms depending on the category of the collected answers. For example, the question generation unit applies a specialized question algorithm to technical answers. The question generation unit can also apply a simple question algorithm to general answers. The question generation unit can also analyze the category of the collected answers and apply the most appropriate question algorithm. In this way, by applying different question algorithms depending on the category of the collected answers, more appropriate questions can be generated.

[0049] The question generation unit can improve the accuracy of questions by referring to the user's past question history. The question generation unit improves the accuracy of questions by referring to the user's past question history, for example. For example, the question generation unit generates related questions based on questions the user has previously asked. The question generation unit can also extract specific patterns from the user's past question history and customize questions based on those patterns. The question generation unit can also generate familiar questions by referring to the language and expressions the user has used in the past. In this way, more accurate questions can be generated by referring to the user's past question history.

[0050] The question generation unit can determine the priority of questions based on the submission time of collected answers. The question generation unit determines the priority of questions based on, for example, the submission time of collected answers. For example, the question generation unit generates questions preferentially based on the most recently submitted answers. The question generation unit can also generate questions to be postponed based on older answers. The question generation unit can also analyze the submission time of collected answers and generate questions with appropriate priority. In this way, by determining the priority of questions based on the submission time of collected answers, more appropriate questions can be generated.

[0051] The question generation unit can adjust the order of questions based on the relevance of the collected answers. The question generation unit adjusts the order of questions based on, for example, the relevance of the collected answers. For example, the question generation unit generates questions preferentially based on answers with high relevance. The question generation unit can also generate questions to be postponed based on answers with low relevance. The question generation unit can also evaluate the relevance of the collected answers and generate questions in an appropriate order. In this way, by adjusting the order of questions based on the relevance of the collected answers, more appropriate questions can be generated.

[0052] The question generation unit can appropriately use technical terms in questions depending on the user's level of expertise. The question generation unit appropriately uses technical terms in questions depending on, for example, the user's level of expertise. For example, if the user is an expert, the question generation unit generates questions that use a lot of technical terms. Furthermore, if the user is a beginner, the question generation unit can also generate questions that explain things in simple terms. Furthermore, the question generation unit can evaluate the user's level of expertise and generate questions that use appropriate technical terms. This allows for more appropriate dialogue by adjusting the use of technical terms in questions depending on the user's level of expertise.

[0053] The stabilization unit can select the optimal stabilization method by analyzing past fluctuations in the collected answers. For example, the stabilization unit analyzes past fluctuations in the collected answers to select the optimal stabilization method. For example, the stabilization unit analyzes past fluctuations in the collected answers and preferentially selects stable answers. The stabilization unit can also evaluate fluctuation patterns in the collected answers to select the optimal stabilization method. The stabilization unit can also select a highly reliable stabilization method based on past fluctuations in the collected answers. In this way, a more reliable stabilization method can be selected by analyzing past fluctuations in the collected answers.

[0054] The stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly reliable answers. The stabilization unit, for example, evaluates the reliability of the collected answers and preferentially stabilizes highly reliable answers. For example, the stabilization unit evaluates the reliability of the collected answers and preferentially stabilizes highly rated answers. The stabilization unit can also check the source of the collected answers and preferentially stabilize answers from highly reliable sources. The stabilization unit can also analyze the content of the collected answers and preferentially stabilize answers that include highly reliable information. In this way, by evaluating the reliability of the collected answers, more reliable information can be obtained.

[0055] The stabilization unit can improve the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method based on user feedback. The stabilization unit can also apply a more effective stabilization method by reflecting feedback provided by the user. The stabilization unit can also analyze user feedback and select an optimal stabilization method. In this way, a more effective stabilization method can be applied by reflecting user feedback.

[0056] The stabilization unit can select the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit selects the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit analyzes the geographic distribution of the collected responses and selects a stabilization method according to the characteristics of each region. The stabilization unit can also prioritize stabilization of responses related to a specific region. The stabilization unit can also select the optimal stabilization method by taking into account the geographic distribution of the collected responses. In this way, a more appropriate stabilization method can be selected by taking into account the geographic distribution of the collected responses.

[0057] The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers. The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers, for example. For example, the stabilization unit can perform highly accurate stabilization by referring to literature related to the collected answers. The stabilization unit can also analyze literature related to the collected answers and preferentially stabilize answers containing highly reliable information. The stabilization unit can also perform detailed stabilization on a specific topic by referring to literature related to the collected answers. In this way, by referring to literature related to the collected answers, more accurate stabilization can be performed.

[0058] The stabilization unit can perform stabilization taking into account the market value of the collected answers. The stabilization unit, for example, performs stabilization taking into account the market value of the collected answers. For example, the stabilization unit prioritizes stabilizing answers with high market value. The stabilization unit can also analyze the market value of the collected answers and prioritize stabilizing answers that are strong in a particular field. The stabilization unit can also select an optimal stabilization method taking into account the market value of the collected answers. In this way, by taking into account the market value of the collected answers, more valuable information can be stabilized.

[0059] The question generation unit can adjust the level of detail of the question based on the importance of the collected answers. The question generation unit adjusts the level of detail of the question based on, for example, the importance of the collected answers. For example, the question generation unit generates a detailed question based on an answer with a high level of importance. The question generation unit can also generate a concise question based on an answer with a low level of importance. The question generation unit can also evaluate the importance of the collected answers and generate a question with an appropriate level of detail. In this way, by adjusting the level of detail of the question based on the importance of the collected answers, it is possible to generate a more appropriate question.

[0060] The question generation unit can apply different question algorithms depending on the category of the collected answers. For example, the question generation unit applies different question algorithms depending on the category of the collected answers. For example, the question generation unit applies a specialized question algorithm to technical answers. The question generation unit can also apply a simple question algorithm to general answers. The question generation unit can also analyze the category of the collected answers and apply the most appropriate question algorithm. In this way, by applying different question algorithms depending on the category of the collected answers, more appropriate questions can be generated.

[0061] The question generation unit can improve the accuracy of questions by referring to the user's past question history. The question generation unit improves the accuracy of questions by referring to the user's past question history, for example. For example, the question generation unit generates related questions based on questions the user has previously asked. The question generation unit can also extract specific patterns from the user's past question history and customize questions based on those patterns. The question generation unit can also generate familiar questions by referring to the language and expressions the user has used in the past. In this way, more accurate questions can be generated by referring to the user's past question history.

[0062] The question generation unit can determine the priority of questions based on the submission time of collected answers. The question generation unit determines the priority of questions based on, for example, the submission time of collected answers. For example, the question generation unit generates questions preferentially based on the most recently submitted answers. The question generation unit can also generate questions to be postponed based on older answers. The question generation unit can also analyze the submission time of collected answers and generate questions with appropriate priority. In this way, by determining the priority of questions based on the submission time of collected answers, more appropriate questions can be generated.

[0063] The question generation unit can adjust the order of questions based on the relevance of the collected answers. The question generation unit adjusts the order of questions based on, for example, the relevance of the collected answers. For example, the question generation unit generates questions preferentially based on answers with high relevance. The question generation unit can also generate questions to be postponed based on answers with low relevance. The question generation unit can also evaluate the relevance of the collected answers and generate questions in an appropriate order. In this way, by adjusting the order of questions based on the relevance of the collected answers, more appropriate questions can be generated.

[0064] The question generation unit can adjust the use of technical terms in questions according to the user's level of expertise. The question generation unit adjusts the use of technical terms in questions according to the user's level of expertise, for example. For example, if the user is an expert, the question generation unit generates questions that use a lot of technical terms. Furthermore, if the user is a beginner, the question generation unit can also generate questions that explain things in simple terms. Furthermore, the question generation unit can evaluate the user's level of expertise and generate questions that use appropriate technical terms. In this way, adjusting the use of technical terms in questions according to the user's level of expertise enables more appropriate dialogue.

[0065] The stabilization unit can select the optimal stabilization method by analyzing past fluctuations in the collected answers. For example, the stabilization unit analyzes past fluctuations in the collected answers to select the optimal stabilization method. For example, the stabilization unit analyzes past fluctuations in the collected answers and preferentially selects stable answers. The stabilization unit can also evaluate fluctuation patterns in the collected answers to select the optimal stabilization method. The stabilization unit can also select a highly reliable stabilization method based on past fluctuations in the collected answers. In this way, a more reliable stabilization method can be selected by analyzing past fluctuations in the collected answers.

[0066] The stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly reliable answers. The stabilization unit, for example, evaluates the reliability of the collected answers and preferentially stabilizes highly reliable answers. For example, the stabilization unit evaluates the reliability of the collected answers and preferentially stabilizes highly rated answers. The stabilization unit can also check the source of the collected answers and preferentially stabilize answers from highly reliable sources. The stabilization unit can also analyze the content of the collected answers and preferentially stabilize answers that include highly reliable information. In this way, by evaluating the reliability of the collected answers, more reliable information can be obtained.

[0067] The stabilization unit can improve the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method based on user feedback. The stabilization unit can also apply a more effective stabilization method by reflecting feedback provided by the user. The stabilization unit can also analyze user feedback and select an optimal stabilization method. In this way, a more effective stabilization method can be applied by reflecting user feedback.

[0068] The stabilization unit can select the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit selects the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit analyzes the geographic distribution of the collected responses and selects a stabilization method according to the characteristics of each region. The stabilization unit can also prioritize stabilization of responses related to a specific region. The stabilization unit can also select the optimal stabilization method by taking into account the geographic distribution of the collected responses. In this way, a more appropriate stabilization method can be selected by taking into account the geographic distribution of the collected responses.

[0069] The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers. The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers, for example. For example, the stabilization unit can perform highly accurate stabilization by referring to literature related to the collected answers. The stabilization unit can also analyze literature related to the collected answers and preferentially stabilize answers containing highly reliable information. The stabilization unit can also perform detailed stabilization on a specific topic by referring to literature related to the collected answers. In this way, by referring to literature related to the collected answers, more accurate stabilization can be performed.

[0070] The stabilization unit can perform stabilization taking into account the market value of the collected answers. The stabilization unit, for example, performs stabilization taking into account the market value of the collected answers. For example, the stabilization unit prioritizes stabilizing answers with high market value. The stabilization unit can also analyze the market value of the collected answers and prioritize stabilizing answers that are strong in a particular field. The stabilization unit can also select an optimal stabilization method taking into account the market value of the collected answers. In this way, by taking into account the market value of the collected answers, more valuable information can be stabilized.

[0071] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0072] The information processing system may further include a history analysis unit that analyzes the user's past behavioral history. The history analysis unit analyzes patterns of questions and answers asked by the user in the past to identify the user's interests. For example, the history analysis unit may identify topics about which the user has frequently asked questions in the past and generate prompts related to those topics. The history analysis unit may also generate similar questions based on answers that the user has previously rated highly. Furthermore, the history analysis unit may analyze the user's behavioral patterns and provide prompts at optimal times. This enables more personalized interactions by utilizing the user's past behavioral history.

[0073] The answer collection unit can analyze the generation AI's past answer history and select an appropriate collection method. For example, the answer collection unit prioritizes collecting highly reliable answers from the generation AI's past answer history. The answer collection unit can also analyze the generation AI's past answer patterns and select an efficient collection method. Furthermore, the answer collection unit can prioritize collecting answers related to specific topics based on the generation AI's past answer history. In this way, by analyzing the generation AI's past answer history, more reliable answers can be collected.

[0074] The question generation unit can adjust the level of detail of the question based on the importance of the collected answers. For example, the question generation unit generates a detailed question based on answers with high importance. The question generation unit can also generate a concise question based on answers with low importance. Furthermore, the question generation unit can evaluate the importance of the collected answers and generate a question with an appropriate level of detail. In this way, by adjusting the level of detail of the question based on the importance of the collected answers, more appropriate questions can be generated.

[0075] The prompt generation unit can generate highly relevant prompts based on the user's geographical location information. For example, if the user is in a specific area, the prompt generation unit can generate a prompt including information related to that area. If the user is traveling, the prompt generation unit can also generate a prompt related to the user's travel destination. Furthermore, if the user is participating in a specific event, the prompt generation unit can also generate a prompt related to the event. In this way, more relevant prompts can be generated by taking the user's geographical location information into consideration.

[0076] The question generator can adjust the use of technical terms in questions according to the user's level of expertise. For example, if the user is an expert, the question generator can generate questions that use a lot of technical terms. If the user is a beginner, the question generator can also generate questions that explain things in simple terms. Furthermore, the question generator can evaluate the user's level of expertise and generate questions that use appropriate technical terms. This allows for more appropriate dialogue by adjusting the use of technical terms in questions according to the user's level of expertise.

[0077] The stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly reliable answers. For example, the stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly rated answers. The stabilization unit can also check the source of the collected answers and preferentially stabilize answers from highly reliable sources. Furthermore, the stabilization unit can analyze the content of the collected answers and preferentially stabilize answers that include highly reliable information. In this way, more reliable information can be obtained by evaluating the reliability of the collected answers.

[0078] The processing flow of the first embodiment will be briefly explained below.

[0079] Step 1: The prompt generation unit accepts input from the user. For example, it receives text or speech input from the user and generates a prompt. It can also use generation AI to generate prompts to elicit knowledge that is the premise of the conversation. Step 2: The answer collection unit asks multiple generation AIs questions based on the prompts generated by the prompt generation unit and collects their answers. For example, the unit asks multiple generation AIs a question such as "Please tell me some application examples of this technology," and each generation AI generates an answer, which is then collected. Step 3: The question generator generates advanced questions based on the collected answers. For example, based on the collected answers, it generates advanced questions such as "What is the most effective application of this technology?" Step 4: The stabilization unit repeats the process of asking multiple AI generators questions based on the questions generated by the question generator, until the answers reach a certain standard. For example, the stabilization unit repeats the questions until multiple AI generators give the same answer to the same question.

[0080] (Example 2) A system according to an embodiment of the present invention combines multi-turn and multi-LLM to maximize the capabilities of a generative AI. This system accepts user input, extracts prerequisite knowledge from the generative AI, asks the same question to multiple generative AIs, collects responses, generates more advanced questions, and asks them again, repeating this process until the answers stabilize. For example, when a user inputs a prompt such as "Please tell me some basic information about a specific technology," the generative AI generates an answer. The system then asks multiple generative AIs a question such as "Please tell me some application examples of this technology," each of which generates an answer and collects the answers. Based on the collected answers, the system generates a more advanced question such as "What is the most effective application example of this technology?" and asks the question again to multiple generative AIs, collecting their answers. This process is repeated until multiple generative AIs repeatedly give the same answer to the same question, at which point the system determines that the answer has stabilized. This allows the system to maximize the capabilities of the generative AI and obtain reliable information. For example, generative AI can be used to efficiently collect and analyze information in the research and development of new technologies.

[0081] An information processing system according to an embodiment includes a prompt generation unit, an answer collection unit, a question generation unit, and a stabilization unit. The prompt generation unit receives input from a user. For example, the prompt generation unit can receive text or voice input from a user and generate it as a prompt. The prompt generation unit can also use a generation AI to generate prompts to elicit knowledge that is the basis of a conversation. For example, the generation AI generates a prompt such as, "Please tell me some basic information about a specific technology." The answer collection unit asks multiple generation AIs questions based on the prompt generated by the prompt generation unit and collects their answers. For example, the answer collection unit asks multiple generation AIs a question such as, "Please tell me some application examples of this technology," and each generation AI generates an answer and collects the answers. The question generation unit generates an advanced question based on the collected answers. For example, the question generation unit generates an advanced question such as, "Among the application examples of this technology, what is the most effective one?" based on the collected answers. The stabilization unit again asks multiple generation AIs questions based on the question generated by the question generation unit until the answers are stabilized. For example, the stabilization unit repeats a question until multiple generation AIs repeatedly give the same answer to the same question. This allows the information processing system according to the embodiment to maximize the capabilities of the generation AI and obtain highly reliable information. For example, the stabilization unit determines that the generation AI's capabilities have been fully utilized when the answers become stable.

[0082] The prompt generation unit can generate prompts to elicit knowledge that is a prerequisite for the conversation. The prompt generation unit generates prompts to elicit knowledge that is a prerequisite for the conversation, for example, by using a generation AI. For example, the prompt generation unit generates a prompt such as, "Please tell me some basic information about a specific technology." The prompt generation unit can also generate an appropriate prompt based on user input. For example, if the user inputs, "I would like to know about application examples of a new technology," the prompt generation unit generates a prompt such as, "Please tell me some application examples of this technology." In this way, by generating a prompt that elicits knowledge that is a prerequisite for the conversation, preparations for the dialogue are completed.

[0083] The answer collection unit can ask the same question to multiple generation AIs and collect their answers. For example, the answer collection unit asks the same question to multiple generation AIs and collects their answers. For example, the answer collection unit asks multiple generation AIs a question such as "Please tell me some application examples of this technology," and each generation AI generates an answer, which is then collected. The answer collection unit can also organize the collected answers and use them to generate the next question. For example, the answer collection unit organizes the collected answers by category and provides them to the question generation unit. In this way, by collecting answers from multiple generation AIs, information from various perspectives can be obtained.

[0084] The question generation unit can generate advanced questions based on the collected answers. For example, the question generation unit generates advanced questions based on the collected answers. For example, the question generation unit generates advanced questions such as "What is the most effective application example of this technology?" based on the collected answers. The question generation unit can also analyze the collected answers using a generation AI to generate appropriate questions. For example, the question generation unit generates advanced questions by analyzing the collected answers using a generation AI and extracting important information. In this way, by generating advanced questions based on the collected answers, deeper information can be elicited.

[0085] The stabilization unit can repeatedly ask questions to multiple generation AIs until the answers become stable. For example, the stabilization unit repeats questions to multiple generation AIs until the answers become stable. For example, the stabilization unit repeats questions until multiple generation AIs repeatedly give the same answer to the same question. The stabilization unit can also use the generation AI to evaluate the stability of the answers. For example, the stabilization unit evaluates the consistency and reliability of the answers and repeats the questions until a stable answer is obtained by the generation AI. In this way, by repeating the questions until the answers become stable, highly reliable information can be obtained.

[0086] The stabilization unit can determine that the generation AI's capabilities have been fully utilized when the answer becomes stable. For example, the stabilization unit determines that the generation AI's capabilities have been fully utilized when the answer becomes stable. For example, the stabilization unit determines that the answer has stabilized if multiple generation AIs begin to repeatedly give the same answer to the same question. The stabilization unit can also use the generation AI to evaluate the stability of the answer. For example, the stabilization unit evaluates the consistency and reliability of the answer and determines that the generation AI's capabilities have been fully utilized when a stable answer is obtained. In this way, optimal information can be obtained by determining that the generation AI's capabilities have been fully utilized when the answer becomes stable.

[0087] The prompt generation unit can estimate the user's emotions and adjust the content and expression of the prompt based on the estimated user emotions. For example, the prompt generation unit estimates the user's emotions and adjusts the content and expression of the prompt based on the estimated user emotions. For example, if the user is feeling stressed, the prompt generation unit generates a simple and intuitive prompt to reduce the burden of input. If the user is relaxed, the prompt generation unit can generate a prompt containing detailed information to promote in-depth dialogue. If the user is in a hurry, the prompt generation unit can generate a short and to-the-point prompt to obtain a quick response. This enables more appropriate dialogue by adjusting the content and expression of the prompt according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0088] The prompt generation unit can analyze the user's past input history and generate an appropriate prompt. The prompt generation unit, for example, analyzes the user's past input history and generates an appropriate prompt. For example, the prompt generation unit generates a relevant prompt based on questions or themes that the user has frequently input in the past. The prompt generation unit can also extract specific patterns from the user's past input history and customize prompts based on those patterns. For example, the prompt generation unit generates a familiar prompt by referring to the language and expressions used by the user in the past. In this way, by analyzing the user's past input history, it is possible to generate more relevant prompts.

[0089] The prompt generation unit can customize prompts based on the user's current areas of interest or projects. The prompt generation unit customizes prompts based on, for example, the user's current areas of interest or projects. For example, the prompt generation unit generates specific prompts based on information related to a project the user is currently working on. The prompt generation unit can also analyze the user's recent search history or browsing history to generate prompts that are aligned with the user's themes of interest. For example, the prompt generation unit generates highly relevant prompts by referring to topics in communities or forums in which the user participates. This enables more specific interactions by customizing prompts based on the user's current areas of interest or projects.

[0090] The prompt generation unit can select an appropriate prompt format depending on the user's input method. For example, the prompt generation unit selects an appropriate prompt format depending on the user's input method. For example, if the user is using voice input, the prompt generation unit generates a prompt suitable for voice recognition. Furthermore, if the user is using text input, the prompt generation unit can also generate a concise and easy-to-read prompt. Furthermore, if the user is using image input, the prompt generation unit can also generate a prompt based on image analysis. In this way, by selecting the optimal prompt format depending on the user's input method, more user-friendly interactions are possible.

[0091] The prompt generation unit can estimate the user's emotions and determine the priority of prompts based on the estimated user emotions. The prompt generation unit, for example, estimates the user's emotions and determines the priority of prompts based on the estimated user emotions. For example, if the user is nervous, the prompt generation unit can preferentially generate prompts to relax the user. Also, if the user is excited, the prompt generation unit can preferentially generate prompts to calm the user down. Also, if the user is tired, the prompt generation unit can preferentially generate simple and less burdensome prompts. This enables more appropriate dialogue by determining the priority of prompts according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0092] The prompt generation unit can generate a highly relevant prompt based on the user's geographical location information. The prompt generation unit generates a highly relevant prompt based on, for example, the user's geographical location information. For example, if the user is in a specific area, the prompt generation unit generates a prompt including information related to the area. Furthermore, if the user is traveling, the prompt generation unit can generate a prompt related to the travel destination. Furthermore, if the user is participating in a specific event, the prompt generation unit can generate a prompt related to the event. In this way, by taking the user's geographical location information into consideration, more relevant prompts can be generated.

[0093] The prompt generation unit can analyze the user's social media activity and generate relevant prompts. The prompt generation unit, for example, analyzes the user's social media activity and generates relevant prompts. For example, the prompt generation unit generates relevant prompts based on content shared by the user on social media. The prompt generation unit can also generate interesting prompts by referring to the activities of the user's friends on social media. The prompt generation unit can also analyze the content posted by accounts the user follows and generate relevant prompts. In this way, more relevant prompts can be generated by analyzing the user's social media activity.

[0094] The prompt generation unit can customize the prompt content by reflecting the user's past feedback. The prompt generation unit customizes the prompt content by reflecting the user's past feedback, for example. For example, the prompt generation unit generates an improved prompt based on feedback provided by the user in the past. Furthermore, if the user gives a high rating to a specific prompt, the prompt generation unit can generate a similar prompt. Furthermore, the prompt generation unit can avoid prompts that the user has previously expressed dissatisfaction with and generate more appropriate prompts. In this way, more appropriate prompts can be generated by reflecting the user's past feedback.

[0095] The answer collection unit can estimate the user's emotions and adjust the timing of answer collection based on the estimated user emotions. The answer collection unit, for example, estimates the user's emotions and adjusts the timing of answer collection based on the estimated user emotions. For example, if the user is relaxed, the answer collection unit collects answers immediately. Also, if the user is feeling stressed, the answer collection unit can wait a short time before collecting answers. Also, if the user is in a hurry, the answer collection unit can collect answers quickly. In this way, by adjusting the timing of answer collection according to the user's emotions, more appropriate answers can be obtained. Emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0096] The answer collection unit can analyze the generation AI's past answer history and select an appropriate collection method. The answer collection unit, for example, analyzes the generation AI's past answer history and selects an appropriate collection method. For example, the answer collection unit prioritizes collecting highly reliable answers from the generation AI's past answer history. The answer collection unit can also analyze the generation AI's past answer patterns and select an efficient collection method. The answer collection unit can also prioritize collecting answers related to specific topics based on the generation AI's past answer history. In this way, more reliable answers can be collected by analyzing the generation AI's past answer history.

[0097] The answer collection unit can customize the collection means based on the characteristics and performance of the generation AI. The answer collection unit customizes the collection means based on, for example, the characteristics and performance of the generation AI. For example, the answer collection unit preferentially collects answers from high-performance generation AI. The answer collection unit can also preferentially collect answers from generation AIs that are strong in a particular field. The answer collection unit can also adjust the collection means according to the characteristics of the generation AI to obtain the optimal answer. In this way, by customizing the collection means based on the characteristics and performance of the generation AI, it is possible to obtain a more optimal answer.

[0098] The answer collection unit can evaluate the reliability of the collected answers and preferentially collect highly reliable answers. The answer collection unit, for example, evaluates the reliability of the collected answers and preferentially collects highly reliable answers. For example, the answer collection unit evaluates the reliability of the collected answers and preferentially collects highly rated answers. The answer collection unit can also check the source of the collected answers and preferentially collect answers from highly reliable sources. The answer collection unit can also analyze the content of the collected answers and preferentially collect answers that include highly reliable information. In this way, by evaluating the reliability of the collected answers, more reliable information can be obtained.

[0099] The answer collection unit can estimate the user's emotions and determine the priority of answers to be collected based on the estimated user emotions. The answer collection unit, for example, estimates the user's emotions and determines the priority of answers to be collected based on the estimated user emotions. For example, if the user is nervous, the answer collection unit can preferentially collect answers to relax the user. Also, if the user is excited, the answer collection unit can preferentially collect answers to help the user regain their composure. Also, if the user is tired, the answer collection unit can preferentially collect answers that are easy and less burdensome. In this way, by determining the priority of answers to be collected according to the user's emotions, more appropriate answers can be obtained. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0100] The answer collection unit can prioritize collecting highly relevant answers by taking into account the geographical distribution of the generation AI. The answer collection unit, for example, prioritizes collecting highly relevant answers by taking into account the geographical distribution of the generation AI. For example, if the generation AI has information related to a specific region, the answer collection unit prioritizes collecting answers related to that region. The answer collection unit can also analyze the geographical distribution of the generation AI and prioritize collecting answers from generation AIs that are strong in a specific region. The answer collection unit can also prioritize collecting answers that are in line with the characteristics of each region by taking into account the geographical distribution of the generation AI. In this way, by taking into account the geographical distribution of the generation AI, more relevant answers can be collected.

[0101] The answer collection unit can improve the accuracy of collection by referring to literature related to the generation AI. The answer collection unit can improve the accuracy of collection by, for example, referring to literature related to the generation AI. For example, the answer collection unit collects highly accurate answers based on the related literature referred to by the generation AI. The answer collection unit can also analyze literature related to the generation AI and preferentially collect answers that include highly reliable information. The answer collection unit can also collect detailed answers on a specific topic by referring to literature related to the generation AI. In this way, by referring to literature related to the generation AI, more accurate answers can be collected.

[0102] The response collection unit can perform collection based on the market value of the generating AI. The response collection unit performs collection based on, for example, the market value of the generating AI. For example, the response collection unit preferentially collects responses from generating AIs with high market value. The response collection unit can also analyze the market value of the generating AI and preferentially collect responses from generating AIs that are strong in a particular field. The response collection unit can also adjust the collection means, taking into account the market value of the generating AI, to obtain optimal responses. In this way, by taking into account the market value of the generating AI, it is possible to collect more valuable responses.

[0103] The question generation unit can estimate the user's emotions and adjust the content and expression of the question based on the estimated user's emotions. For example, the question generation unit estimates the user's emotions and adjusts the content and expression of the question based on the estimated user's emotions. For example, if the user is feeling stressed, the question generation unit generates a simple and intuitive question. Also, if the user is relaxed, the question generation unit can generate a question that includes detailed information. Also, if the user is in a hurry, the question generation unit can generate a short and to-the-point question. This enables more appropriate dialogue by adjusting the content and expression of the question according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0104] The question generation unit can adjust the level of detail of the question based on the importance of the collected answers. The question generation unit adjusts the level of detail of the question based on, for example, the importance of the collected answers. For example, the question generation unit generates a detailed question based on an answer with a high level of importance. The question generation unit can also generate a concise question based on an answer with a low level of importance. The question generation unit can also evaluate the importance of the collected answers and generate a question with an appropriate level of detail. In this way, by adjusting the level of detail of the question based on the importance of the collected answers, it is possible to generate a more appropriate question.

[0105] The question generation unit can apply different question algorithms depending on the category of the collected answers. For example, the question generation unit applies different question algorithms depending on the category of the collected answers. For example, the question generation unit applies a specialized question algorithm to technical answers. The question generation unit can also apply a simple question algorithm to general answers. The question generation unit can also analyze the category of the collected answers and apply the most appropriate question algorithm. In this way, by applying different question algorithms depending on the category of the collected answers, more appropriate questions can be generated.

[0106] The question generation unit can improve the accuracy of questions by referring to the user's past question history. The question generation unit improves the accuracy of questions by referring to the user's past question history, for example. For example, the question generation unit generates related questions based on questions the user has previously asked. The question generation unit can also extract specific patterns from the user's past question history and customize questions based on those patterns. The question generation unit can also generate familiar questions by referring to the language and expressions the user has used in the past. In this way, more accurate questions can be generated by referring to the user's past question history.

[0107] The question generation unit can estimate the user's emotion and adjust the length of the question based on the estimated user's emotion. The question generation unit, for example, estimates the user's emotion and adjusts the length of the question based on the estimated user's emotion. For example, the question generation unit generates a short and concise question when the user is nervous. The question generation unit can also generate a longer question including detailed information when the user is relaxed. The question generation unit can also generate a short and to-the-point question when the user is in a hurry. This allows for more appropriate dialogue by adjusting the length of the question according to the user's emotion. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0108] The question generation unit can determine the priority of questions based on the submission time of collected answers. The question generation unit determines the priority of questions based on, for example, the submission time of collected answers. For example, the question generation unit generates questions preferentially based on the most recently submitted answers. The question generation unit can also generate questions to be postponed based on older answers. The question generation unit can also analyze the submission time of collected answers and generate questions with appropriate priority. In this way, by determining the priority of questions based on the submission time of collected answers, more appropriate questions can be generated.

[0109] The question generation unit can adjust the order of questions based on the relevance of the collected answers. The question generation unit adjusts the order of questions based on, for example, the relevance of the collected answers. For example, the question generation unit generates questions preferentially based on answers with high relevance. The question generation unit can also generate questions to be postponed based on answers with low relevance. The question generation unit can also evaluate the relevance of the collected answers and generate questions in an appropriate order. In this way, by adjusting the order of questions based on the relevance of the collected answers, more appropriate questions can be generated.

[0110] The question generation unit can appropriately use technical terms in questions depending on the user's level of expertise. The question generation unit appropriately uses technical terms in questions depending on, for example, the user's level of expertise. For example, if the user is an expert, the question generation unit generates questions that use a lot of technical terms. Furthermore, if the user is a beginner, the question generation unit can also generate questions that explain things in simple terms. Furthermore, the question generation unit can evaluate the user's level of expertise and generate questions that use appropriate technical terms. This allows for more appropriate dialogue by adjusting the use of technical terms in questions depending on the user's level of expertise.

[0111] The stabilization unit can estimate the user's emotion and adjust the stabilization method based on the estimated user's emotion. For example, the stabilization unit estimates the user's emotion and adjusts the stabilization method based on the estimated user's emotion. For example, if the user is tense, the stabilization unit applies a stabilization method to relax the user. Furthermore, if the user is excited, the stabilization unit can apply a stabilization method to regain composure. Furthermore, if the user is tired, the stabilization unit can apply a simple and less burdensome stabilization method. This allows for more appropriate dialogue by adjusting the stabilization method according to the user's emotion. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0112] The stabilization unit can select the optimal stabilization method by analyzing past fluctuations in the collected answers. For example, the stabilization unit analyzes past fluctuations in the collected answers to select the optimal stabilization method. For example, the stabilization unit analyzes past fluctuations in the collected answers and preferentially selects stable answers. The stabilization unit can also evaluate fluctuation patterns in the collected answers to select the optimal stabilization method. The stabilization unit can also select a highly reliable stabilization method based on past fluctuations in the collected answers. In this way, a more reliable stabilization method can be selected by analyzing past fluctuations in the collected answers.

[0113] The stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly reliable answers. The stabilization unit, for example, evaluates the reliability of the collected answers and preferentially stabilizes highly reliable answers. For example, the stabilization unit evaluates the reliability of the collected answers and preferentially stabilizes highly rated answers. The stabilization unit can also check the source of the collected answers and preferentially stabilize answers from highly reliable sources. The stabilization unit can also analyze the content of the collected answers and preferentially stabilize answers that include highly reliable information. In this way, by evaluating the reliability of the collected answers, more reliable information can be obtained.

[0114] The stabilization unit can improve the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method based on user feedback. The stabilization unit can also apply a more effective stabilization method by reflecting feedback provided by the user. The stabilization unit can also analyze user feedback and select an optimal stabilization method. In this way, a more effective stabilization method can be applied by reflecting user feedback.

[0115] The stabilization unit can estimate the user's emotions and determine the priority of stabilization based on the estimated user emotions. For example, the stabilization unit estimates the user's emotions and determines the priority of stabilization based on the estimated user emotions. For example, if the user is nervous, the stabilization unit can prioritize stabilization to relax the user. Also, if the user is excited, the stabilization unit can prioritize stabilization to restore composure. Also, if the user is tired, the stabilization unit can prioritize simple and less burdensome stabilization. This enables more appropriate dialogue by determining the priority of stabilization according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0116] The stabilization unit can select the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit selects the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit analyzes the geographic distribution of the collected responses and selects a stabilization method according to the characteristics of each region. The stabilization unit can also prioritize stabilization of responses related to a specific region. The stabilization unit can also select the optimal stabilization method by taking into account the geographic distribution of the collected responses. In this way, a more appropriate stabilization method can be selected by taking into account the geographic distribution of the collected responses.

[0117] The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers. The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers, for example. For example, the stabilization unit can perform highly accurate stabilization by referring to literature related to the collected answers. The stabilization unit can also analyze literature related to the collected answers and preferentially stabilize answers containing highly reliable information. The stabilization unit can also perform detailed stabilization on a specific topic by referring to literature related to the collected answers. In this way, by referring to literature related to the collected answers, more accurate stabilization can be performed.

[0118] The stabilization unit can perform stabilization taking into account the market value of the collected answers. The stabilization unit, for example, performs stabilization taking into account the market value of the collected answers. For example, the stabilization unit prioritizes stabilizing answers with high market value. The stabilization unit can also analyze the market value of the collected answers and prioritize stabilizing answers that are strong in a particular field. The stabilization unit can also select an optimal stabilization method taking into account the market value of the collected answers. In this way, by taking into account the market value of the collected answers, more valuable information can be stabilized.

[0119] The question generation unit can estimate the user's emotions and adjust the content and expression of the question based on the estimated user's emotions. For example, the question generation unit estimates the user's emotions and adjusts the content and expression of the question based on the estimated user's emotions. For example, if the user is feeling stressed, the question generation unit generates a simple and intuitive question. Also, if the user is relaxed, the question generation unit can generate a question that includes detailed information. Also, if the user is in a hurry, the question generation unit can generate a short and to-the-point question. This enables more appropriate dialogue by adjusting the content and expression of the question according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0120] The question generation unit can adjust the level of detail of the question based on the importance of the collected answers. The question generation unit adjusts the level of detail of the question based on, for example, the importance of the collected answers. For example, the question generation unit generates a detailed question based on an answer with a high level of importance. The question generation unit can also generate a concise question based on an answer with a low level of importance. The question generation unit can also evaluate the importance of the collected answers and generate a question with an appropriate level of detail. In this way, by adjusting the level of detail of the question based on the importance of the collected answers, it is possible to generate a more appropriate question.

[0121] The question generation unit can apply different question algorithms depending on the category of the collected answers. For example, the question generation unit applies different question algorithms depending on the category of the collected answers. For example, the question generation unit applies a specialized question algorithm to technical answers. The question generation unit can also apply a simple question algorithm to general answers. The question generation unit can also analyze the category of the collected answers and apply the most appropriate question algorithm. In this way, by applying different question algorithms depending on the category of the collected answers, more appropriate questions can be generated.

[0122] The question generation unit can improve the accuracy of questions by referring to the user's past question history. The question generation unit improves the accuracy of questions by referring to the user's past question history, for example. For example, the question generation unit generates related questions based on questions the user has previously asked. The question generation unit can also extract specific patterns from the user's past question history and customize questions based on those patterns. The question generation unit can also generate familiar questions by referring to the language and expressions the user has used in the past. In this way, more accurate questions can be generated by referring to the user's past question history.

[0123] The question generation unit can estimate the user's emotion and adjust the length of the question based on the estimated user's emotion. The question generation unit, for example, estimates the user's emotion and adjusts the length of the question based on the estimated user's emotion. For example, the question generation unit generates a short and concise question when the user is nervous. The question generation unit can also generate a longer question including detailed information when the user is relaxed. The question generation unit can also generate a short and to-the-point question when the user is in a hurry. This allows for more appropriate dialogue by adjusting the length of the question according to the user's emotion. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0124] The question generation unit can determine the priority of questions based on the submission time of collected answers. The question generation unit determines the priority of questions based on, for example, the submission time of collected answers. For example, the question generation unit generates questions preferentially based on the most recently submitted answers. The question generation unit can also generate questions to be postponed based on older answers. The question generation unit can also analyze the submission time of collected answers and generate questions with appropriate priority. In this way, by determining the priority of questions based on the submission time of collected answers, more appropriate questions can be generated.

[0125] The question generation unit can adjust the order of questions based on the relevance of the collected answers. The question generation unit adjusts the order of questions based on, for example, the relevance of the collected answers. For example, the question generation unit generates questions preferentially based on answers with high relevance. The question generation unit can also generate questions to be postponed based on answers with low relevance. The question generation unit can also evaluate the relevance of the collected answers and generate questions in an appropriate order. In this way, by adjusting the order of questions based on the relevance of the collected answers, more appropriate questions can be generated.

[0126] The question generation unit can adjust the use of technical terms in questions according to the user's level of expertise. The question generation unit adjusts the use of technical terms in questions according to the user's level of expertise, for example. For example, if the user is an expert, the question generation unit generates questions that use a lot of technical terms. Furthermore, if the user is a beginner, the question generation unit can also generate questions that explain things in simple terms. Furthermore, the question generation unit can evaluate the user's level of expertise and generate questions that use appropriate technical terms. In this way, adjusting the use of technical terms in questions according to the user's level of expertise enables more appropriate dialogue.

[0127] The stabilization unit can estimate the user's emotion and adjust the stabilization method based on the estimated user's emotion. For example, the stabilization unit estimates the user's emotion and adjusts the stabilization method based on the estimated user's emotion. For example, if the user is tense, the stabilization unit applies a stabilization method to relax the user. Furthermore, if the user is excited, the stabilization unit can apply a stabilization method to regain composure. Furthermore, if the user is tired, the stabilization unit can apply a simple and less burdensome stabilization method. This allows for more appropriate dialogue by adjusting the stabilization method according to the user's emotion. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0128] The stabilization unit can select the optimal stabilization method by analyzing past fluctuations in the collected answers. For example, the stabilization unit analyzes past fluctuations in the collected answers to select the optimal stabilization method. For example, the stabilization unit analyzes past fluctuations in the collected answers and preferentially selects stable answers. The stabilization unit can also evaluate fluctuation patterns in the collected answers to select the optimal stabilization method. The stabilization unit can also select a highly reliable stabilization method based on past fluctuations in the collected answers. In this way, a more reliable stabilization method can be selected by analyzing past fluctuations in the collected answers.

[0129] The stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly reliable answers. The stabilization unit, for example, evaluates the reliability of the collected answers and preferentially stabilizes highly reliable answers. For example, the stabilization unit evaluates the reliability of the collected answers and preferentially stabilizes highly rated answers. The stabilization unit can also check the source of the collected answers and preferentially stabilize answers from highly reliable sources. The stabilization unit can also analyze the content of the collected answers and preferentially stabilize answers that include highly reliable information. In this way, by evaluating the reliability of the collected answers, more reliable information can be obtained.

[0130] The stabilization unit can improve the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method by reflecting user feedback. For example, the stabilization unit improves the stabilization method based on user feedback. The stabilization unit can also apply a more effective stabilization method by reflecting feedback provided by the user. The stabilization unit can also analyze user feedback and select an optimal stabilization method. In this way, a more effective stabilization method can be applied by reflecting user feedback.

[0131] The stabilization unit can estimate the user's emotions and determine the priority of stabilization based on the estimated user emotions. For example, the stabilization unit estimates the user's emotions and determines the priority of stabilization based on the estimated user emotions. For example, if the user is nervous, the stabilization unit can prioritize stabilization to relax the user. Also, if the user is excited, the stabilization unit can prioritize stabilization to restore composure. Also, if the user is tired, the stabilization unit can prioritize simple and less burdensome stabilization. This enables more appropriate dialogue by determining the priority of stabilization according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0132] The stabilization unit can select the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit selects the optimal stabilization method by taking into account the geographic distribution of the collected responses. For example, the stabilization unit analyzes the geographic distribution of the collected responses and selects a stabilization method according to the characteristics of each region. The stabilization unit can also prioritize stabilization of responses related to a specific region. The stabilization unit can also select the optimal stabilization method by taking into account the geographic distribution of the collected responses. In this way, a more appropriate stabilization method can be selected by taking into account the geographic distribution of the collected responses.

[0133] The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers. The stabilization unit can improve the accuracy of stabilization by referring to literature related to the collected answers, for example. For example, the stabilization unit can perform highly accurate stabilization by referring to literature related to the collected answers. The stabilization unit can also analyze literature related to the collected answers and preferentially stabilize answers containing highly reliable information. The stabilization unit can also perform detailed stabilization on a specific topic by referring to literature related to the collected answers. In this way, by referring to literature related to the collected answers, more accurate stabilization can be performed.

[0134] The stabilization unit can perform stabilization taking into account the market value of the collected answers. The stabilization unit, for example, performs stabilization taking into account the market value of the collected answers. For example, the stabilization unit prioritizes stabilizing answers with high market value. The stabilization unit can also analyze the market value of the collected answers and prioritize stabilizing answers that are strong in a particular field. The stabilization unit can also select an optimal stabilization method taking into account the market value of the collected answers. In this way, by taking into account the market value of the collected answers, more valuable information can be stabilized. === Hard Collateral 1-1 === Each of the multiple elements, including the prompt generation unit, answer collection unit, question generation unit, and stabilization unit, described above, is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the prompt generation unit is realized by the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12. For example, the prompt generation unit receives input from a user using the touch panel 38A or the microphone 38B of the smart device 14. The answer collection unit, realized, for example, by the specific processing unit 290 of the data processing device 12, asks questions to multiple generation AIs and collects answers. The question generation unit, realized, for example, by the specific processing unit 290 of the data processing device 12, generates an advanced question based on the collected answers. The stabilization unit, realized, for example, by the specific processing unit 290 of the data processing device 12, repeats the question until the answer stabilizes. === Hard Collateral 1-2 === Each of the multiple elements including the above-described prompt generation unit, answer collection unit, question generation unit, and stabilization unit is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the prompt generation unit is realized by the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. For example, the prompt generation unit receives input from a user using the microphone 238 of the smart glasses 214. The answer collection unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and asks questions to multiple generation AIs and collects answers. The question generation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates an advanced question based on the collected answers. The stabilization unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and repeats the question until the answer stabilizes. === Hard Collateral 1-3 === Each of the multiple elements including the above-mentioned prompt generation unit, answer collection unit, question generation unit, and stabilization unit is realized, for example, by at least one of the headset-type terminal 314 and the data processing device 12. For example, the prompt generation unit is realized by the control unit 46A of the headset-type terminal 314 or the specific processing unit 290 of the data processing device 12. For example, the prompt generation unit receives input from a user using the microphone 238 of the headset-type terminal 314. The answer collection unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and asks questions to multiple generation AIs and collects answers. The question generation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates advanced questions based on the collected answers. The stabilization unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and repeats questions until the answers become stable. === Hard Collateral 1-4 === Each of the multiple elements including the above-mentioned prompt generation unit, answer collection unit, question generation unit, and stabilization unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the prompt generation unit is realized by the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12. For example, the prompt generation unit receives input from a user using the microphone 238 of the robot 414. The answer collection unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and asks questions to multiple generation AIs and collects answers. The question generation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates an advanced question based on the collected answers. The stabilization unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and repeats asking questions until the answers become stable.

[0135] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0136] The information processing system may further include a history analysis unit that analyzes the user's past behavioral history. The history analysis unit analyzes patterns of questions and answers asked by the user in the past to identify the user's interests. For example, the history analysis unit may identify topics about which the user has frequently asked questions in the past and generate prompts related to those topics. The history analysis unit may also generate similar questions based on answers that the user has previously rated highly. Furthermore, the history analysis unit may analyze the user's behavioral patterns and provide prompts at optimal times. This enables more personalized interactions by utilizing the user's past behavioral history.

[0137] The prompt generation unit can estimate the user's current emotional state and adjust the content and expression of the prompt based on the estimated emotion. For example, if the user is feeling stressed, the prompt generation unit can generate a simple and intuitive prompt to reduce the burden of input. If the user is relaxed, the prompt generation unit can generate a prompt containing detailed information to promote in-depth dialogue. Furthermore, if the user is in a hurry, the prompt generation unit can generate a short and to-the-point prompt to obtain a quick response. This allows for more appropriate dialogue by adjusting the content and expression of the prompt according to the user's emotion.

[0138] The answer collection unit can analyze the generation AI's past answer history and select an appropriate collection method. For example, the answer collection unit prioritizes collecting highly reliable answers from the generation AI's past answer history. The answer collection unit can also analyze the generation AI's past answer patterns and select an efficient collection method. Furthermore, the answer collection unit can prioritize collecting answers related to specific topics based on the generation AI's past answer history. In this way, by analyzing the generation AI's past answer history, more reliable answers can be collected.

[0139] The question generation unit can adjust the level of detail of the question based on the importance of the collected answers. For example, the question generation unit generates a detailed question based on answers with high importance. The question generation unit can also generate a concise question based on answers with low importance. Furthermore, the question generation unit can evaluate the importance of the collected answers and generate a question with an appropriate level of detail. In this way, by adjusting the level of detail of the question based on the importance of the collected answers, more appropriate questions can be generated.

[0140] The stabilization unit can estimate the user's emotions and adjust the stabilization method based on the estimated user's emotions. For example, if the user is nervous, the stabilization unit can apply a stabilization method to relax the user. If the user is excited, the stabilization unit can also apply a stabilization method to calm the user down. Furthermore, if the user is tired, the stabilization unit can apply a simple and less burdensome stabilization method. In this way, adjusting the stabilization method according to the user's emotions enables more appropriate dialogue.

[0141] The prompt generation unit can generate highly relevant prompts based on the user's geographical location information. For example, if the user is in a specific area, the prompt generation unit can generate a prompt including information related to that area. If the user is traveling, the prompt generation unit can also generate a prompt related to the user's travel destination. Furthermore, if the user is participating in a specific event, the prompt generation unit can also generate a prompt related to the event. In this way, more relevant prompts can be generated by taking the user's geographical location information into consideration.

[0142] The answer collection unit can estimate the user's emotions and determine the priority of answers to be collected based on the estimated user's emotions. For example, if the user is nervous, the answer collection unit can preferentially collect answers to relax the user. Also, if the user is excited, the answer collection unit can preferentially collect answers to help the user regain their composure. Furthermore, if the user is tired, the answer collection unit can preferentially collect answers that are easy and less burdensome. In this way, by determining the priority of answers to be collected according to the user's emotions, more appropriate answers can be obtained.

[0143] The question generator can adjust the use of technical terms in questions according to the user's level of expertise. For example, if the user is an expert, the question generator can generate questions that use a lot of technical terms. If the user is a beginner, the question generator can also generate questions that explain things in simple terms. Furthermore, the question generator can evaluate the user's level of expertise and generate questions that use appropriate technical terms. This allows for more appropriate dialogue by adjusting the use of technical terms in questions according to the user's level of expertise.

[0144] The stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly reliable answers. For example, the stabilization unit can evaluate the reliability of the collected answers and preferentially stabilize highly rated answers. The stabilization unit can also check the source of the collected answers and preferentially stabilize answers from highly reliable sources. Furthermore, the stabilization unit can analyze the content of the collected answers and preferentially stabilize answers that include highly reliable information. In this way, more reliable information can be obtained by evaluating the reliability of the collected answers.

[0145] The question generation unit can estimate the user's emotions and adjust the length of the questions based on the estimated user's emotions. For example, if the user is nervous, the question generation unit can generate short and concise questions. If the user is relaxed, the question generation unit can also generate longer questions that include detailed information. Furthermore, if the user is in a hurry, the question generation unit can also generate short and to the point questions. This allows for more appropriate dialogue by adjusting the length of the questions according to the user's emotions.

[0146] The processing flow of the second embodiment will be briefly explained below.

[0147] Step 1: The prompt generation unit accepts input from the user. For example, it receives text or speech input from the user and generates a prompt. It can also use generation AI to generate prompts to elicit knowledge that is the premise of the conversation. Step 2: The answer collection unit asks multiple generation AIs questions based on the prompts generated by the prompt generation unit and collects their answers. For example, the unit asks multiple generation AIs a question such as "Please tell me some application examples of this technology," and each generation AI generates an answer, which is then collected. Step 3: The question generator generates advanced questions based on the collected answers. For example, based on the collected answers, it generates advanced questions such as "What is the most effective application of this technology?" Step 4: The stabilization unit repeats the process of asking multiple AI generators questions based on the questions generated by the question generator, until the answers reach a certain standard. For example, the stabilization unit repeats the questions until multiple AI generators give the same answer to the same question.

[0148] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0149] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0150] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0151] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0152] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0153] 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0154] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0155] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0156] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0157] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0158] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0159] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0160] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0161] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0162] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0163] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0164] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0165] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0166] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0167] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0168] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0169] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0170] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0171] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0172] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0173] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0174] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0175] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0176] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0177] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0178] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0179] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0180] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0181] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0182] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0183] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0184] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0185] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0186] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0187] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0188] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0189] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0190] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0191] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0192] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0193] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0194] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0195] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0196] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0197] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0198] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0199] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0200] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0201] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0202] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0203] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0204] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0205] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0206] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[0207] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[0208] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0209] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0210] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0211] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0212] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. A processor also includes a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0213] The hardware resource that executes the specific process may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0214] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0215] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0216] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0217] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0218] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0219] [Explanation of symbols]

[0220] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. a prompt generator that accepts input from a user; an answer collection unit that asks questions to a plurality of generation AIs based on the prompts generated by the prompt generation unit and collects answers; a question generation unit that generates an advanced question based on the answers collected by the answer collection unit; a stabilization unit that asks a plurality of generation AIs questions again based on the questions generated by the question generation unit, and repeats this process until the answers reach a certain standard; Equipped with A system characterized by:

2. The prompt generation unit Generate prompts to elicit prior knowledge for the conversation 2. The system of claim 1.

3. The response collection unit Ask multiple AI generators the same questions and collect their responses.

2. The system of claim 1.

4. The question generation unit Generate advanced questions based on collected answers 2. The system of claim 1.

5. The stabilizing section Repeatedly ask questions to multiple AI generators until the answers stabilize 2. The system of claim 1.

6. The stabilizing section Once the answer becomes stable, it is determined that the capabilities of the generative AI have been fully utilized.

2. The system of claim 1.

7. The prompt generation unit Inferring user emotions and adjusting prompt content and wording based on the inferred user emotions 2. The system of claim 1.

8. The prompt generation unit Analyzes the user's past input history and generates appropriate prompts 2. The system of claim 1.

Images