System

Abstract

To provide a system that automatically generates appropriate answers to user questions.SOLUTION: A system according to an embodiment comprises: an acquisition unit; a generation unit; a reception unit; a provision unit; and an update unit. The acquisition unit acquires information on a pair of a question and an answer. The generation unit generates a learning model that inputs a question and outputs an answer on the basis of the information acquired by the acquisition unit. The reception unit receives questions from users. The provision unit provides answers to the questions received by the reception unit to the users by using the learning model generated by the generation unit. The update unit evaluates the learning model and updates the learning model on the basis of the evaluation results.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 a description and related instruction sentence regarding 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] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional technology, there is room for improvement in the performance of systems that automatically generate appropriate answers to user questions.

[0005] The system according to the embodiment aims to automatically generate appropriate answers to user questions. [Means for solving the problem]

[0006] The system according to the embodiment includes an acquisition unit, a generation unit, a reception unit, a provision unit, and an update unit. The acquisition unit acquires question and answer pair information. The generation unit generates a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The reception unit accepts questions from a user. The provision unit uses the learning model generated by the generation unit to provide the user with an answer to the question accepted by the reception unit. The update unit evaluates the learning model and updates the learning model based on the evaluation results. Effect of the Invention

[0007] The system according to the embodiment can automatically generate appropriate answers to user questions. [Brief description 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. [Diagram 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. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 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. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 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 PREFERRED EMBODIMENTS

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

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

[0011] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one 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 signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs and 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 code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or 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. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "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, the RAM 30, and the 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 (e.g., 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., voice and / or text). Display 40A displays visible information such as text and images according to instructions from processor 46. Speaker 40B outputs voice according to instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

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

[0024] As shown in Fig. 2, 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. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process 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 process 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[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 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. Also, the data processing device 12 may be a server device, or may be a terminal device (e.g., a mobile phone, a robot, a home appliance, etc.) owned by a user. Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A question and answer system according to an embodiment of the present invention is a system that acquires information on question and answer pairs, generates a learning model, provides answers to questions from users, and updates the learning model. This question and answer system has an acquisition unit that acquires information on question and answer pairs, thereby collecting training data. Next, there is a generation unit that generates a learning model that inputs questions and outputs answers based on the acquired information. This generation unit builds a natural language generation model and generates an optimal answer to a question from a user. A question from a user is accepted by a reception unit, and an answer is provided to the user from a provision unit using the generated learning model. Furthermore, there is an update unit that evaluates the learning model and updates the learning model based on the evaluation result. This update unit periodically trains and evaluates the model to improve performance. In addition, there is a management unit that manages the flow of dialogue with the user, understands the context of the dialogue, and generates appropriate follow-up questions. The provision unit provides the user with follow-up questions according to the flow of dialogue managed by the management unit. It is also possible to generate follow-up questions using a generation AI. The acquisition unit has a web scraping function that automatically collects question and answer pairs from the Internet. The update unit includes a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. It also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. The provision unit has a function of adjusting the output speed of the answer according to the user's input speed. The management unit learns the user's dialogue style and generates follow-up questions according to the dialogue style. It also has a function of analyzing the frequency and intervals of speech during the user's dialogue and measuring the user's interest level. The provision unit highlights keywords during the user's dialogue and generates answers based on the keywords. This allows the question and answer system to provide quick and accurate answers to questions from the user and generate follow-up questions according to the user's dialogue style and interest level. In addition, the performance of the system can be improved by periodically updating the model.

[0029] The question and answer system according to the embodiment includes an acquisition unit, a generation unit, a reception unit, a provision unit, and an update unit. The acquisition unit acquires information on pairs of questions and answers. The question and answer pairs include, for example, text format, audio format, and those related to a specific topic, but are not limited to such examples. The acquisition unit has, for example, a web scraping function that automatically collects question and answer pairs from the Internet. The web scraping function is realized by using, for example, a library such as BeautifulSoup or Scrapy. The generation unit uses a generation AI to generate a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The generation unit generates, for example, a learning model such as a neural network, a decision tree, or a support vector machine. The reception unit accepts questions from a user. The reception unit can accept questions by, for example, a method such as text input or voice input. The provision unit uses the learning model generated by the generation unit to provide the user with an answer to the question accepted by the reception unit. The provision unit has a function of adjusting the output speed of the answer according to, for example, the input speed of the user. The update unit evaluates the learning model and updates the learning model based on the evaluation result. The update unit includes, for example, a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The update unit also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. This allows the question and answer system according to the embodiment to acquire information on question and answer pairs, generate a learning model, provide answers to questions from users, and update the learning model.

[0030] The acquisition unit acquires information on question and answer pairs. The question and answer pairs include, but are not limited to, for example, text format, audio format, and those related to a specific topic. The acquisition unit has, for example, a web scraping function that automatically collects question and answer pairs from the Internet. The web scraping function is realized using, for example, libraries such as BeautifulSoup and Scrapy. Specifically, the web scraping function analyzes HTML structures and extracts necessary information in order to extract question and answer pairs from a specific website, forum, or Q&A platform. This allows the acquisition unit to collect a variety of question and answer pairs from a wide range of information sources. Furthermore, the acquisition unit has a function of centrally managing the collected data and storing it in a database. In the database, question and answer pairs are organized by topic and indexed for efficient access. This allows the acquisition unit to quickly search for necessary information and provide it to the generation unit and the provision unit. In addition, the acquisition unit has a filtering function for evaluating the quality of the collected data and removing noise and duplicate data. For example, natural language processing technology can be used to select grammatically correct question and answer pairs and retain only the most reliable data, allowing the acquisition unit to collect high-quality data and improve the performance of the entire system.

[0031] The generation unit uses the generation AI to generate a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The generation unit generates a learning model such as a neural network, a decision tree, or a support vector machine. Specifically, the generation unit trains the generation AI using question and answer pairs provided by the acquisition unit. The training data is preprocessed and converted into a format suitable for the model. For example, text data is tokenized and converted into an appropriate vector representation. The generation AI uses these data to learn patterns for generating optimal answers to questions. During the training process, the generation unit evaluates the performance of the model and adjusts hyperparameters and improves the model as necessary. For example, the generation unit evaluates the accuracy and reproducibility of the model using cross-validation and a test dataset, and selects the optimal model. In addition, the generation unit can generate multiple models to respond to questions of different topics and formats, and optimize each model for a specific purpose. As a result, the generation unit can generate learning models to provide highly accurate answers to a variety of questions, thereby improving the performance of the entire system.

[0032] The reception unit receives questions from the user. The reception unit can receive questions by, for example, text input or voice input. Specifically, the reception unit has a function of receiving text or voice data input by the user and converting it into an appropriate format. In the case of text input, the question input by the user is received as is, and in the case of voice input, the voice data is converted into text using a voice recognition technology. For the voice recognition technology, for example, a voice recognition model using deep learning is used. The reception unit has an interface for quickly processing the user's input and providing it to the generation unit. Furthermore, the reception unit uses a natural language processing technology for analyzing the contents of the user's input and identifying the intention and topic of the question. For example, the reception unit extracts keywords of the question and classifies them into related topics, making it easier for the generation unit to generate an appropriate answer. This allows the reception unit to efficiently receive a variety of questions from users and improve the response speed and accuracy of the entire system.

[0033] The providing unit uses the learning model generated by the generating unit to provide the user with an answer to the question accepted by the accepting unit. The providing unit has a function of adjusting the output speed of the answer according to, for example, the input speed of the user. Specifically, the providing unit has an interface for displaying the answer provided by the generating unit to the user. The answer in text format is displayed on the user's screen, and the answer in voice format is provided to the user using voice synthesis technology. For example, text-to-speech (TTS) technology is used as the voice synthesis technology. The providing unit can adjust the output speed and format of the answer according to the input speed and situation of the user. For example, when the user is in a hurry, a concise answer is quickly provided, and when detailed information is required, a more detailed answer is provided. In addition, the providing unit has a function of collecting user feedback and evaluating the quality of the answer and the user satisfaction. As a result, the providing unit can quickly provide an appropriate answer to the user and improve the user experience of the entire system.

[0034] The update unit evaluates the learning model and updates the learning model based on the evaluation result. The update unit includes, for example, a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The update unit also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. Specifically, the update unit analyzes the feedback from the user and calculates an index for evaluating the performance of the learning model. For example, the learning model is evaluated using indexes such as the accuracy of the answer, relevance, and user satisfaction. Based on the evaluation result, the parameters of the learning model are adjusted to improve the accuracy of the model. The update unit also periodically collects new data and updates the learning model based on the latest information. As a result, the update unit can always maintain a highly accurate learning model that reflects the latest information and improve the performance of the entire system. Furthermore, the update unit can use an anomaly detection algorithm to detect abnormal operation or performance degradation of the learning model early and take measures quickly. As a result, the update unit can ensure the reliability and stability of the learning model and improve the reliability of the entire system.

[0035] The question and answer system includes a management unit that manages the flow of dialogue with the user, and the provision unit provides the user with follow-up questions according to the flow of dialogue managed by the management unit. The management unit manages the flow of dialogue with the user. The flow of dialogue includes, for example, steps of dialogue and timing of follow-up, but is not limited to such examples. The management unit, for example, learns the dialogue style of the user and generates follow-up questions according to the dialogue style. The dialogue style includes, for example, formal, informal, question-centered, etc. The management unit has a function of analyzing the frequency and interval of speech during the user's dialogue and measuring the user's interest level. The frequency and interval of speech include, for example, the number of speeches per minute and the average time between speeches. The provision unit provides the user with follow-up questions according to the flow of dialogue managed by the management unit. The provision unit can also generate follow-up questions using a generation AI. This allows the question and answer system to manage the flow of dialogue with the user and provide appropriate follow-up questions.

[0036] The providing unit can generate follow-up questions using the generation AI. The generation AI understands the context of the user's dialogue and generates appropriate follow-up questions. For example, the generation AI analyzes the content of the user's remarks and generates the next appropriate question based on the content. The generation AI can also generate follow-up questions according to the user's dialogue style and interest level. This allows the providing unit to generate follow-up questions using the generation AI.

[0037] The acquisition unit may have a web scraping function that automatically collects question and answer pairs from the Internet. The web scraping function is realized, for example, by using a library such as BeautifulSoup or Scrapy. The web scraping function extracts question and answer pairs from web pages on the Internet and collects them as training data. For example, the web scraping function periodically collects question and answer pairs from a specific website and stores them in a database. The web scraping function can also automatically filter the collected data and select highly relevant question and answer pairs. This allows the acquisition unit to automatically collect question and answer pairs from the Internet.

[0038] The update unit may include a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The feedback collection unit collects feedback such as user reviews and survey results. The feedback collection unit analyzes the collected feedback and identifies improvements to the learning model. For example, the feedback collection unit analyzes user ratings and comments and extracts improvements for improving the accuracy and recall of the learning model. The feedback collection unit may also generate data for updating the learning model based on the collected feedback. This allows the update unit to update the learning model based on the user feedback.

[0039] The update unit may have a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. The external database may include, but is not limited to, a specific API or a type of database. The update unit periodically acquires the latest question and answer pairs from the external database and uses them as training data. For example, the update unit acquires data from the external database based on a weekly or monthly schedule and updates the learning model. The update unit may also automatically filter the acquired data and select highly relevant question and answer pairs. This allows the update unit to periodically acquire the latest question and answer pairs from the external database and update the learning model.

[0040] The providing unit may have a function of adjusting the answer output speed according to the user's input speed. The input speed includes, but is not limited to, for example, the number of characters per second and the input delay time. The providing unit monitors the user's input speed in real time and sets an appropriate output speed. For example, the providing unit may speed up the answer output speed when the user is inputting at a high speed. Also, the providing unit may slow down the answer output speed when the user is inputting at a low speed. This allows the providing unit to adjust the answer output speed according to the user's input speed.

[0041] The management unit can learn the user's dialogue style and generate follow-up questions according to the dialogue style. The dialogue style includes, for example, formal, informal, and question-centered. The management unit has a function of analyzing the user's speech frequency and intervals during dialogue and measuring the user's interest level. The speech frequency and intervals include, for example, the number of speeches per minute and the average time between speeches. The management unit learns the user's dialogue style and generates follow-up questions according to the dialogue style. For example, the management unit analyzes the content of the user's speech and generates the next appropriate question based on the content. The management unit can also generate follow-up questions according to the user's dialogue style and interest level. This allows the management unit to generate follow-up questions according to the user's dialogue style.

[0042] The management unit can have a function of analyzing the speech frequency and intervals during a user's dialogue and measuring the user's level of interest. The speech frequency and intervals include, for example, the number of speeches per minute and the average time between speeches. The management unit monitors the user's speech frequency and intervals in real time and measures the user's level of interest. For example, the management unit may determine that the user's level of interest is high when the user speaks frequently. Also, the management unit may determine that the user's level of interest is low when the user speaks at long intervals. This allows the management unit to analyze the speech frequency and intervals during a user's dialogue and measure the user's level of interest.

[0043] The providing unit can highlight keywords in the user's dialogue and generate an answer based on the keywords. The keywords include important words and phrases extracted by methods such as TF-IDF and co-occurrence network analysis. The providing unit analyzes the content of the user's speech and highlights important keywords. For example, the providing unit highlights a specific keyword in the user's speech and generates an answer based on the keyword. The providing unit can also adjust the detail level of the answer depending on the importance of the keyword. This allows the providing unit to highlight keywords in the user's dialogue and generate an answer based on the keyword.

[0044] When acquiring question and answer pairs, the acquisition unit can analyze the user's past question history and select the optimal acquisition method. The optimal acquisition method includes, for example, filtering based on the past history and a recommendation system. The acquisition unit analyzes the user's past question history and finds a specific pattern. For example, the acquisition unit preferentially acquires topics that the user has frequently asked about in the past. In addition, the acquisition unit can select an acquisition method based on a specific pattern from the user's past question history. This allows the acquisition unit to analyze the user's past question history and select the optimal acquisition method.

[0045] The acquisition unit can perform filtering based on the user's current field of interest when acquiring question and answer pairs. The field of interest includes, for example, the user's search history and survey results. The acquisition unit analyzes the user's current field of interest and preferentially acquires highly relevant question and answer pairs. For example, the acquisition unit preferentially acquires question and answer pairs related to a topic in which the user is currently interested. In addition, the acquisition unit can filter highly relevant question and answer pairs based on the user's current field of interest. This allows the acquisition unit to filter question and answer pairs based on the user's current field of interest.

[0046] When acquiring question and answer pairs, the acquisition unit can preferentially acquire highly relevant pairs in consideration of the geographical location information of the user. The geographical location information includes, for example, GPS data and an IP address. The acquisition unit analyzes the geographical location information of the user and preferentially acquires question and answer pairs related to the region. For example, when the user is in a specific region, the acquisition unit preferentially acquires question and answer pairs related to the region. In addition, the acquisition unit can filter question and answer pairs specific to the region based on the geographical location information of the user. This allows the acquisition unit to preferentially acquire question and answer pairs related to the user in consideration of the geographical location information of the user.

[0047] When acquiring the question and answer pairs, the acquisition unit can analyze the user's social media activity and acquire related pairs. The social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The acquisition unit analyzes the user's social media activity and identifies topics of interest. For example, the acquisition unit preferentially acquires question and answer pairs related to topics that the user frequently mentions on social media. In addition, the acquisition unit can identify topics of interest from the user's social media activity and acquire question and answer pairs related to the topics. This allows the acquisition unit to analyze the user's social media activity and acquire related question and answer pairs.

[0048] When generating an answer, the generating unit can adjust the level of detail of the answer based on the importance of the question. The importance of a question includes, for example, the frequency of the question and the user's interest in the question. The generating unit analyzes the importance of the question and sets an appropriate level of detail. For example, the generating unit can generate a detailed and comprehensive answer for an important question. Also, the generating unit can generate a concise answer that gets to the point for a general question. This allows the generating unit to adjust the level of detail of the answer based on the importance of the question.

[0049] When generating an answer, the generation unit can apply different generation algorithms depending on the category of the question. Examples of generation algorithms include RNN and Transformer. The generation unit analyzes the category of the question and selects an appropriate generation algorithm. For example, the generation unit can apply a generation algorithm with technical expertise to a technical question. Also, the generation unit can apply a generation algorithm with general knowledge to a general question. This allows the generation unit to apply the optimal generation algorithm depending on the category of the question.

[0050] When generating answers, the generation unit can determine the priority of the answers based on the time when the question was submitted. The submission time includes, for example, the submission date and time and the time elapsed since submission. The generation unit analyzes the submission time of the question and sets an appropriate priority. For example, the generation unit can generate an answer quickly for an urgent question. Also, for a general question, the generation unit can generate an answer with normal priority. This allows the generation unit to determine the priority of the answers based on the time when the question was submitted.

[0051] The generation unit can adjust the order of answers based on the relevance of the questions when generating answers. The relevance of the questions includes, for example, the similarity of the question contents and the degree of matching of the topics. The generation unit analyzes the relevance of the questions and sets an appropriate order. For example, when the relevance of the questions is high, the generation unit generates answers with priority. Also, when the relevance of the questions is medium, the generation unit can generate answers in a normal order. This allows the generation unit to adjust the order of answers based on the relevance of the questions.

[0052] When accepting a question, the reception unit can select the optimal reception method by referring to the user's past question history. The optimal reception method includes, for example, filtering based on the past history and selection based on the user's preference. The reception unit analyzes the user's past question history and finds a specific pattern. For example, the reception unit can automatically display questions that the user frequently input in the past as candidates. In addition, the reception unit can preferentially suggest input methods (voice, text, etc.) that the user used in the past. This allows the reception unit to select the optimal reception method by referring to the user's past question history.

[0053] The reception unit can filter questions based on the user's current field of interest when receiving a question. The field of interest includes, for example, the user's search history and survey results. The reception unit analyzes the user's current field of interest and preferentially receives questions with high relevance. For example, the reception unit preferentially receives questions related to topics in which the user is currently interested. Also, the reception unit can filter questions with high relevance based on the user's current field of interest. This allows the reception unit to filter questions based on the user's current field of interest.

[0054] When accepting a question, the reception unit can preferentially accept highly relevant questions in consideration of the geographical location information of the user. Highly relevant questions include, for example, filtering based on geographical location information and selection based on the user's field of interest. The reception unit analyzes the user's geographical location information and preferentially accepts questions related to the area. For example, when the user is in a specific area, the reception unit preferentially accepts questions related to the area. In addition, questions specific to the area can be filtered based on the user's geographical location information. This allows the reception unit to preferentially accept highly relevant questions in consideration of the user's geographical location information.

[0055] When receiving a question, the reception unit can analyze the user's social media activity and receive related questions. Social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The reception unit analyzes the user's social media activity and identifies topics of interest. For example, the reception unit preferentially receives questions related to topics that the user frequently mentions on social media. In addition, the reception unit can identify topics of interest from the user's social media activity and receive questions related to those topics. This allows the reception unit to analyze the user's social media activity and receive related questions.

[0056] When providing an answer, the providing unit can select an optimal providing method by referring to the user's past question history. The optimal providing method includes, for example, filtering based on the past history and selection based on the user's preferences. The providing unit analyzes the user's past question history and finds a specific pattern. For example, the providing unit preferentially provides answers related to topics that the user has frequently asked questions about in the past. In addition, the providing unit can select a providing method based on a specific pattern from the user's past question history. This allows the providing unit to select an optimal providing method by referring to the user's past question history.

[0057] When providing an answer, the providing unit can customize the answer based on the user's current field of interest. The field of interest includes, for example, the user's search history and survey results. The providing unit analyzes the user's current field of interest and provides highly relevant answers preferentially. For example, the providing unit provides answers related to topics in which the user is currently interested preferentially. In addition, the providing unit can customize highly relevant answers based on the user's current field of interest. This allows the providing unit to customize answers based on the user's current field of interest.

[0058] When providing an answer, the providing unit can provide an optimal answer by taking into consideration the geographical location information of the user. The optimal answer includes, for example, filtering based on the geographical location information and selection based on the user's field of interest. The providing unit analyzes the geographical location information of the user and preferentially provides answers related to the region. For example, when the user is in a specific region, the providing unit preferentially provides answers related to the region. In addition, the providing unit can customize region-specific answers based on the geographical location information of the user. This allows the providing unit to provide an optimal answer by taking into consideration the geographical location information of the user.

[0059] When providing an answer, the providing unit can analyze the user's social media activity and provide a related answer. The social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The providing unit analyzes the user's social media activity and identifies topics of interest. For example, the providing unit preferentially provides answers related to topics that the user frequently mentions on social media. In addition, the providing unit can identify topics of interest from the user's social media activity and provide answers related to the topics. This allows the providing unit to analyze the user's social media activity and provide related answers.

[0060] When updating the learning model, the update unit can select an optimal update method by referring to past learning data. The optimal update method includes, for example, selection based on past learning data and selection based on user feedback. The update unit analyzes past learning data and finds a specific pattern. For example, the update unit analyzes past learning data and selects the most effective update method. In addition, the update unit can select an update method based on a specific pattern from the past learning data. This allows the update unit to select an optimal update method by referring to past learning data.

[0061] The update unit can filter the update data based on the user's current field of interest when updating the learning model. Filtering of the update data includes, for example, filtering based on the user's field of interest and filtering based on past history. The update unit analyzes the user's current field of interest and preferentially updates highly relevant data. For example, the update unit preferentially updates data related to topics in which the user is currently interested. In addition, highly relevant data can be filtered based on the user's current field of interest. This allows the update unit to filter the update data based on the user's current field of interest.

[0062] The update unit can select optimal update data in consideration of the geographical location information of the user when updating the learning model. The optimal update data includes, for example, selection based on geographical location information and selection based on the user's field of interest. The update unit analyzes the user's geographical location information and preferentially updates data related to the region. For example, when the user is in a specific region, the update unit preferentially updates data related to the region. In addition, region-specific data can be filtered based on the user's geographical location information. This allows the update unit to select optimal update data in consideration of the user's geographical location information.

[0063] The update unit can analyze the user's social media activity and select relevant update data when updating the learning model. The social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The update unit analyzes the user's social media activity and identifies topics of interest. For example, the update unit preferentially updates data related to topics that the user frequently mentions on social media. It is also possible to identify topics of interest from the user's social media activity and update data related to the topics. This allows the update unit to analyze the user's social media activity and select relevant update data.

[0064] When managing the dialogue flow, the management unit can select an optimal flow by referring to the user's past dialogue history. The optimal flow includes, for example, selection based on the past dialogue history and selection based on the user's feedback. The management unit analyzes the user's past dialogue history and finds a specific pattern. For example, the management unit preferentially provides dialogue flows that the user has frequently performed in the past. In addition, the management unit can select a dialogue flow based on a specific pattern from the user's past dialogue history. This allows the management unit to select an optimal flow by referring to the user's past dialogue history.

[0065] The management unit can customize the dialogue based on the user's current field of interest when managing the dialogue flow. The customization of the dialogue includes, for example, customization based on the user's field of interest and customization based on past history. The management unit analyzes the user's current field of interest and provides highly relevant dialogues preferentially. For example, the management unit provides a dialogue flow related to a topic in which the user is currently interested preferentially. The management unit can also customize highly relevant dialogues based on the user's current field of interest. This allows the management unit to customize the dialogue based on the user's current field of interest.

[0066] The management unit can provide an optimal dialogue taking into consideration the geographic location information of the user when managing the dialogue flow. The optimal dialogue includes, for example, selection based on the geographic location information or selection based on the user's field of interest. The management unit analyzes the user's geographic location information and preferentially provides dialogue related to the region. For example, when the user is in a specific region, the management unit preferentially provides dialogue related to the region. Region-specific dialogue can also be customized based on the user's geographic location information. This allows the management unit to provide an optimal dialogue taking into consideration the user's geographic location information.

[0067] During dialogue flow management, the management unit can analyze the user's social media activity and provide related dialogues. Social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The management unit analyzes the user's social media activity and identifies topics of interest. For example, the management unit preferentially provides dialogues related to topics that the user frequently mentions on social media. It is also possible to identify topics of interest from the user's social media activity and provide dialogues related to those topics. This allows the management unit to analyze the user's social media activity and provide related dialogues.

[0068] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0069] The acquisition unit can analyze the user's past question history and acquire optimal question and answer pairs. For example, the acquisition unit preferentially acquires topics about which the user has frequently asked questions in the past. In addition, the acquisition unit can select an acquisition method based on a specific pattern from the user's past question history. This allows the acquisition unit to analyze the user's past question history and acquire optimal question and answer pairs.

[0070] The providing unit can provide highly relevant question and answer pairs preferentially in consideration of the geographical location information of the user. For example, when the user is in a specific region, the providing unit can provide question and answer pairs related to the region preferentially. In addition, region-specific question and answer pairs can be filtered based on the geographical location information of the user. This allows the providing unit to provide highly relevant question and answer pairs preferentially in consideration of the geographical location information of the user.

[0071] When generating an answer, the generating unit can apply different generating algorithms depending on the category of the question. For example, the generating unit can apply a generating algorithm having technical expertise to a technical question. Also, the generating unit can apply a generating algorithm having general knowledge to a general question. This allows the generating unit to apply the optimal generating algorithm depending on the category of the question.

[0072] The management unit can analyze the user's social media activities and provide related interactions when managing the flow of interactions. For example, the management unit can provide interactions related to topics that the user frequently mentions on social media with a priority. It can also identify topics of interest from the user's social media activities and provide interactions related to those topics. This allows the management unit to analyze the user's social media activities and provide related interactions.

[0073] The acquisition unit can filter question and answer pairs based on the user's current field of interest when acquiring the question and answer pairs. For example, the acquisition unit preferentially acquires question and answer pairs related to topics in which the user is currently interested. Also, the acquisition unit can filter question and answer pairs with high relevance based on the user's current field of interest. This allows the acquisition unit to filter question and answer pairs based on the user's current field of interest.

[0074] When updating the learning model, the update unit can select optimal update data in consideration of the geographical location information of the user. For example, when the user is in a specific area, the update unit preferentially updates data related to the area. In addition, the update unit can filter area-specific data based on the geographical location information of the user. This allows the update unit to select optimal update data in consideration of the geographical location information of the user.

[0075] The process flow of the first embodiment will be briefly described below.

[0076] Step 1: The acquisition unit acquires information on question and answer pairs. The question and answer pairs may be in text format, audio format, or on a specific topic. The acquisition unit has a web scraping function that automatically collects question and answer pairs from the Internet, and is realized by using a library such as BeautifulSoup or Scrapy. Step 2: The generation unit uses the generation AI to generate a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The generation AI generates a learning model such as a neural network, a decision tree, or a support vector machine. Step 3: The reception unit receives a question from the user. The reception unit can receive the question by a method such as text input or voice input. Step 4: The providing unit uses the learning model generated by the generating unit to provide the user with an answer to the question accepted by the accepting unit. The providing unit has a function of adjusting the output speed of the answer according to the input speed of the user. Step 5: The update unit evaluates the learning model and updates the learning model based on the evaluation result. The update unit includes a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The update unit also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model.

[0077] (Example 2) A question and answer system according to an embodiment of the present invention is a system that acquires information on question and answer pairs, generates a learning model, provides answers to questions from users, and updates the learning model. This question and answer system has an acquisition unit that acquires information on question and answer pairs, thereby collecting training data. Next, there is a generation unit that generates a learning model that inputs questions and outputs answers based on the acquired information. This generation unit builds a natural language generation model and generates an optimal answer to a question from a user. A question from a user is accepted by a reception unit, and an answer is provided to the user from a provision unit using the generated learning model. Furthermore, there is an update unit that evaluates the learning model and updates the learning model based on the evaluation result. This update unit periodically trains and evaluates the model to improve performance. In addition, there is a management unit that manages the flow of dialogue with the user, understands the context of the dialogue, and generates appropriate follow-up questions. The provision unit provides the user with follow-up questions according to the flow of dialogue managed by the management unit. It is also possible to generate follow-up questions using a generation AI. The acquisition unit has a web scraping function that automatically collects question and answer pairs from the Internet. The update unit includes a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. It also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. The provision unit has a function of adjusting the output speed of the answer according to the user's input speed. The management unit learns the user's dialogue style and generates follow-up questions according to the dialogue style. It also has a function of analyzing the frequency and intervals of speech during the user's dialogue and measuring the user's interest level. The provision unit highlights keywords during the user's dialogue and generates answers based on the keywords. This allows the question and answer system to provide quick and accurate answers to questions from the user and generate follow-up questions according to the user's dialogue style and interest level. In addition, the performance of the system can be improved by periodically updating the model.

[0078] The question and answer system according to the embodiment includes an acquisition unit, a generation unit, a reception unit, a provision unit, and an update unit. The acquisition unit acquires information on pairs of questions and answers. The question and answer pairs include, for example, text format, audio format, and those related to a specific topic, but are not limited to such examples. The acquisition unit has, for example, a web scraping function that automatically collects question and answer pairs from the Internet. The web scraping function is realized by using, for example, a library such as BeautifulSoup or Scrapy. The generation unit uses a generation AI to generate a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The generation unit generates, for example, a learning model such as a neural network, a decision tree, or a support vector machine. The reception unit accepts questions from a user. The reception unit can accept questions by, for example, a method such as text input or voice input. The provision unit uses the learning model generated by the generation unit to provide the user with an answer to the question accepted by the reception unit. The provision unit has a function of adjusting the output speed of the answer according to, for example, the input speed of the user. The update unit evaluates the learning model and updates the learning model based on the evaluation result. The update unit includes, for example, a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The update unit also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. This allows the question and answer system according to the embodiment to acquire information on question and answer pairs, generate a learning model, provide answers to questions from users, and update the learning model.

[0079] The acquisition unit acquires information on question and answer pairs. The question and answer pairs include, but are not limited to, for example, text format, audio format, and those related to a specific topic. The acquisition unit has, for example, a web scraping function that automatically collects question and answer pairs from the Internet. The web scraping function is realized using, for example, libraries such as BeautifulSoup and Scrapy. Specifically, the web scraping function analyzes HTML structures and extracts necessary information in order to extract question and answer pairs from a specific website, forum, or Q&A platform. This allows the acquisition unit to collect a variety of question and answer pairs from a wide range of information sources. Furthermore, the acquisition unit has a function of centrally managing the collected data and storing it in a database. In the database, question and answer pairs are organized by topic and indexed for efficient access. This allows the acquisition unit to quickly search for necessary information and provide it to the generation unit and the provision unit. In addition, the acquisition unit has a filtering function for evaluating the quality of the collected data and removing noise and duplicate data. For example, natural language processing technology can be used to select grammatically correct question and answer pairs and retain only the most reliable data, allowing the acquisition unit to collect high-quality data and improve the performance of the entire system.

[0080] The generation unit uses the generation AI to generate a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The generation unit generates a learning model such as a neural network, a decision tree, or a support vector machine. Specifically, the generation unit trains the generation AI using question and answer pairs provided by the acquisition unit. The training data is preprocessed and converted into a format suitable for the model. For example, text data is tokenized and converted into an appropriate vector representation. The generation AI uses these data to learn patterns for generating optimal answers to questions. During the training process, the generation unit evaluates the performance of the model and adjusts hyperparameters and improves the model as necessary. For example, the generation unit evaluates the accuracy and reproducibility of the model using cross-validation and a test dataset, and selects the optimal model. In addition, the generation unit can generate multiple models to respond to questions of different topics and formats, and optimize each model for a specific purpose. As a result, the generation unit can generate learning models to provide highly accurate answers to a variety of questions, thereby improving the performance of the entire system.

[0081] The reception unit receives questions from the user. The reception unit can receive questions by, for example, text input or voice input. Specifically, the reception unit has a function of receiving text or voice data input by the user and converting it into an appropriate format. In the case of text input, the question input by the user is received as is, and in the case of voice input, the voice data is converted into text using a voice recognition technology. For the voice recognition technology, for example, a voice recognition model using deep learning is used. The reception unit has an interface for quickly processing the user's input and providing it to the generation unit. Furthermore, the reception unit uses a natural language processing technology for analyzing the contents of the user's input and identifying the intention and topic of the question. For example, the reception unit extracts keywords of the question and classifies them into related topics, making it easier for the generation unit to generate an appropriate answer. This allows the reception unit to efficiently receive a variety of questions from users and improve the response speed and accuracy of the entire system.

[0082] The providing unit uses the learning model generated by the generating unit to provide the user with an answer to the question accepted by the accepting unit. The providing unit has a function of adjusting the output speed of the answer according to, for example, the input speed of the user. Specifically, the providing unit has an interface for displaying the answer provided by the generating unit to the user. The answer in text format is displayed on the user's screen, and the answer in voice format is provided to the user using voice synthesis technology. For example, text-to-speech (TTS) technology is used as the voice synthesis technology. The providing unit can adjust the output speed and format of the answer according to the input speed and situation of the user. For example, when the user is in a hurry, a concise answer is quickly provided, and when detailed information is required, a more detailed answer is provided. In addition, the providing unit has a function of collecting user feedback and evaluating the quality of the answer and the user satisfaction. As a result, the providing unit can quickly provide an appropriate answer to the user and improve the user experience of the entire system.

[0083] The update unit evaluates the learning model and updates the learning model based on the evaluation result. The update unit includes, for example, a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The update unit also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. Specifically, the update unit analyzes the feedback from the user and calculates an index for evaluating the performance of the learning model. For example, the learning model is evaluated using indexes such as the accuracy of the answer, relevance, and user satisfaction. Based on the evaluation result, the parameters of the learning model are adjusted to improve the accuracy of the model. The update unit also periodically collects new data and updates the learning model based on the latest information. As a result, the update unit can always maintain a highly accurate learning model that reflects the latest information and improve the performance of the entire system. Furthermore, the update unit can use an anomaly detection algorithm to detect abnormal operation or performance degradation of the learning model early and take measures quickly. As a result, the update unit can ensure the reliability and stability of the learning model and improve the reliability of the entire system.

[0084] The question and answer system includes a management unit that manages the flow of dialogue with the user, and the provision unit provides the user with follow-up questions according to the flow of dialogue managed by the management unit. The management unit manages the flow of dialogue with the user. The flow of dialogue includes, for example, steps of dialogue and timing of follow-up, but is not limited to such examples. The management unit, for example, learns the dialogue style of the user and generates follow-up questions according to the dialogue style. The dialogue style includes, for example, formal, informal, question-centered, etc. The management unit has a function of analyzing the frequency and interval of speech during the user's dialogue and measuring the user's interest level. The frequency and interval of speech include, for example, the number of speeches per minute and the average time between speeches. The provision unit provides the user with follow-up questions according to the flow of dialogue managed by the management unit. The provision unit can also generate follow-up questions using a generation AI. This allows the question and answer system to manage the flow of dialogue with the user and provide appropriate follow-up questions.

[0085] The providing unit can generate follow-up questions using the generation AI. The generation AI understands the context of the user's dialogue and generates appropriate follow-up questions. For example, the generation AI analyzes the content of the user's remarks and generates the next appropriate question based on the content. The generation AI can also generate follow-up questions according to the user's dialogue style and interest level. This allows the providing unit to generate follow-up questions using the generation AI.

[0086] The acquisition unit may have a web scraping function that automatically collects question and answer pairs from the Internet. The web scraping function is realized, for example, by using a library such as BeautifulSoup or Scrapy. The web scraping function extracts question and answer pairs from web pages on the Internet and collects them as training data. For example, the web scraping function periodically collects question and answer pairs from a specific website and stores them in a database. The web scraping function can also automatically filter the collected data and select highly relevant question and answer pairs. This allows the acquisition unit to automatically collect question and answer pairs from the Internet.

[0087] The update unit may include a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The feedback collection unit collects feedback such as user reviews and survey results. The feedback collection unit analyzes the collected feedback and identifies improvements to the learning model. For example, the feedback collection unit analyzes user ratings and comments and extracts improvements for improving the accuracy and recall of the learning model. The feedback collection unit may also generate data for updating the learning model based on the collected feedback. This allows the update unit to update the learning model based on the user feedback.

[0088] The update unit may have a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model. The external database may include, but is not limited to, a specific API or a type of database. The update unit periodically acquires the latest question and answer pairs from the external database and uses them as training data. For example, the update unit acquires data from the external database based on a weekly or monthly schedule and updates the learning model. The update unit may also automatically filter the acquired data and select highly relevant question and answer pairs. This allows the update unit to periodically acquire the latest question and answer pairs from the external database and update the learning model.

[0089] The providing unit may have a function of adjusting the answer output speed according to the user's input speed. The input speed includes, but is not limited to, for example, the number of characters per second and the input delay time. The providing unit monitors the user's input speed in real time and sets an appropriate output speed. For example, the providing unit may speed up the answer output speed when the user is inputting at a high speed. Also, the providing unit may slow down the answer output speed when the user is inputting at a low speed. This allows the providing unit to adjust the answer output speed according to the user's input speed.

[0090] The management unit can learn the user's dialogue style and generate follow-up questions according to the dialogue style. The dialogue style includes, for example, formal, informal, and question-centered. The management unit has a function of analyzing the user's speech frequency and intervals during dialogue and measuring the user's interest level. The speech frequency and intervals include, for example, the number of speeches per minute and the average time between speeches. The management unit learns the user's dialogue style and generates follow-up questions according to the dialogue style. For example, the management unit analyzes the content of the user's speech and generates the next appropriate question based on the content. The management unit can also generate follow-up questions according to the user's dialogue style and interest level. This allows the management unit to generate follow-up questions according to the user's dialogue style.

[0091] The management unit can have a function of analyzing the speech frequency and intervals during a user's dialogue and measuring the user's level of interest. The speech frequency and intervals include, for example, the number of speeches per minute and the average time between speeches. The management unit monitors the user's speech frequency and intervals in real time and measures the user's level of interest. For example, the management unit may determine that the user's level of interest is high when the user speaks frequently. Also, the management unit may determine that the user's level of interest is low when the user speaks at long intervals. This allows the management unit to analyze the speech frequency and intervals during a user's dialogue and measure the user's level of interest.

[0092] The providing unit can highlight keywords in the user's dialogue and generate an answer based on the keywords. The keywords include important words and phrases extracted by methods such as TF-IDF and co-occurrence network analysis. The providing unit analyzes the content of the user's speech and highlights important keywords. For example, the providing unit highlights a specific keyword in the user's speech and generates an answer based on the keyword. The providing unit can also adjust the detail level of the answer depending on the importance of the keyword. This allows the providing unit to highlight keywords in the user's dialogue and generate an answer based on the keyword.

[0093] The acquisition unit can estimate the user's emotion and adjust the timing of acquiring question and answer pairs based on the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The acquisition unit monitors the user's emotion in real time and sets an appropriate acquisition timing. For example, when the user is feeling stressed, the acquisition unit delays the acquisition timing and waits until the user relaxes. In addition, when the user is concentrating, the acquisition unit can quickly acquire question and answer pairs and efficiently collect training data. This allows the acquisition unit to adjust the timing of acquiring question and answer pairs based on the user's emotion.

[0094] When acquiring question and answer pairs, the acquisition unit can analyze the user's past question history and select the optimal acquisition method. The optimal acquisition method includes, for example, filtering based on the past history and a recommendation system. The acquisition unit analyzes the user's past question history and finds a specific pattern. For example, the acquisition unit preferentially acquires topics that the user has frequently asked about in the past. In addition, the acquisition unit can select an acquisition method based on a specific pattern from the user's past question history. This allows the acquisition unit to analyze the user's past question history and select the optimal acquisition method.

[0095] The acquisition unit can perform filtering based on the user's current field of interest when acquiring question and answer pairs. The field of interest includes, for example, the user's search history and survey results. The acquisition unit analyzes the user's current field of interest and preferentially acquires highly relevant question and answer pairs. For example, the acquisition unit preferentially acquires question and answer pairs related to a topic in which the user is currently interested. In addition, the acquisition unit can filter highly relevant question and answer pairs based on the user's current field of interest. This allows the acquisition unit to filter question and answer pairs based on the user's current field of interest.

[0096] The acquisition unit can estimate the user's emotion and determine the priority of question and answer pairs to be acquired based on the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The acquisition unit monitors the user's emotion in real time and sets an appropriate priority. For example, when the user is excited, the acquisition unit preferentially acquires question and answer pairs that are interesting. Also, when the user is relaxed, it is possible to preferentially acquire question and answer pairs that are suitable for learning. This allows the acquisition unit to determine the priority of question and answer pairs based on the user's emotion.

[0097] When acquiring question and answer pairs, the acquisition unit can preferentially acquire highly relevant pairs in consideration of the geographical location information of the user. The geographical location information includes, for example, GPS data and an IP address. The acquisition unit analyzes the geographical location information of the user and preferentially acquires question and answer pairs related to the region. For example, when the user is in a specific region, the acquisition unit preferentially acquires question and answer pairs related to the region. In addition, the acquisition unit can filter question and answer pairs specific to the region based on the geographical location information of the user. This allows the acquisition unit to preferentially acquire question and answer pairs related to the user in consideration of the geographical location information of the user.

[0098] When acquiring the question and answer pairs, the acquisition unit can analyze the user's social media activity and acquire related pairs. The social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The acquisition unit analyzes the user's social media activity and identifies topics of interest. For example, the acquisition unit preferentially acquires question and answer pairs related to topics that the user frequently mentions on social media. In addition, the acquisition unit can identify topics of interest from the user's social media activity and acquire question and answer pairs related to the topics. This allows the acquisition unit to analyze the user's social media activity and acquire related question and answer pairs.

[0099] The generation unit can estimate the user's emotion and adjust the way of expressing the answer based on the estimated user's emotion. The estimation of emotion is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The generation unit monitors the user's emotion in real time and sets an appropriate expression method. For example, the generation unit generates a polite and detailed answer when the user is relaxed. Also, when the user is in a hurry, it can generate a concise answer that hits the point. This allows the generation unit to adjust the way of expressing the answer based on the user's emotion.

[0100] When generating an answer, the generating unit can adjust the level of detail of the answer based on the importance of the question. The importance of a question includes, for example, the frequency of the question and the user's interest in the question. The generating unit analyzes the importance of the question and sets an appropriate level of detail. For example, the generating unit can generate a detailed and comprehensive answer for an important question. Also, the generating unit can generate a concise answer that gets to the point for a general question. This allows the generating unit to adjust the level of detail of the answer based on the importance of the question.

[0101] When generating an answer, the generation unit can apply different generation algorithms depending on the category of the question. Examples of generation algorithms include RNN and Transformer. The generation unit analyzes the category of the question and selects an appropriate generation algorithm. For example, the generation unit can apply a generation algorithm with technical expertise to a technical question. Also, the generation unit can apply a generation algorithm with general knowledge to a general question. This allows the generation unit to apply the optimal generation algorithm depending on the category of the question.

[0102] The generation unit can estimate the user's emotion and adjust the length of the answer based on the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The generation unit monitors the user's emotion in real time and sets an appropriate answer length. For example, when the user is in a hurry, the generation unit generates a short and to the point answer. Also, when the user is relaxed, it can generate a longer answer including a detailed explanation. This allows the generation unit to adjust the length of the answer based on the user's emotion.

[0103] When generating answers, the generation unit can determine the priority of the answers based on the time when the question was submitted. The submission time includes, for example, the submission date and time and the time elapsed since submission. The generation unit analyzes the submission time of the question and sets an appropriate priority. For example, the generation unit can generate an answer quickly for an urgent question. Also, for a general question, the generation unit can generate an answer with normal priority. This allows the generation unit to determine the priority of the answers based on the time when the question was submitted.

[0104] The generation unit can adjust the order of answers based on the relevance of the questions when generating answers. The relevance of the questions includes, for example, the similarity of the question contents and the degree of matching of the topics. The generation unit analyzes the relevance of the questions and sets an appropriate order. For example, when the relevance of the questions is high, the generation unit generates answers with priority. Also, when the relevance of the questions is medium, the generation unit can generate answers in a normal order. This allows the generation unit to adjust the order of answers based on the relevance of the questions.

[0105] The reception unit can estimate the user's emotions and adjust the method of receiving the question based on the estimated user's emotions. The estimation of emotions is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The reception unit monitors the user's emotions in real time and sets an appropriate reception method. For example, when the user is feeling stressed, the reception unit provides a simple interface and minimizes input steps. Also, when the user is relaxed, the reception unit can provide detailed input options and suggest a customizable input method. This allows the reception unit to adjust the method of receiving the question based on the user's emotions.

[0106] When accepting a question, the reception unit can select the optimal reception method by referring to the user's past question history. The optimal reception method includes, for example, filtering based on the past history and selection based on the user's preference. The reception unit analyzes the user's past question history and finds a specific pattern. For example, the reception unit can automatically display questions that the user frequently input in the past as candidates. In addition, the reception unit can preferentially suggest input methods (voice, text, etc.) that the user used in the past. This allows the reception unit to select the optimal reception method by referring to the user's past question history.

[0107] The reception unit can filter questions based on the user's current field of interest when receiving a question. The field of interest includes, for example, the user's search history and survey results. The reception unit analyzes the user's current field of interest and preferentially receives questions with high relevance. For example, the reception unit preferentially receives questions related to topics in which the user is currently interested. Also, the reception unit can filter questions with high relevance based on the user's current field of interest. This allows the reception unit to filter questions based on the user's current field of interest.

[0108] The reception unit can estimate the user's emotion and determine the priority of questions based on the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The reception unit monitors the user's emotion in real time and sets appropriate priorities. For example, when the user is excited, the reception unit preferentially receives questions that are interesting. Also, when the user is relaxed, the reception unit can preferentially receive questions that are suitable for learning. This allows the reception unit to determine the priority of questions based on the user's emotion.

[0109] When accepting a question, the reception unit can preferentially accept highly relevant questions in consideration of the geographical location information of the user. Highly relevant questions include, for example, filtering based on geographical location information and selection based on the user's field of interest. The reception unit analyzes the user's geographical location information and preferentially accepts questions related to the area. For example, when the user is in a specific area, the reception unit preferentially accepts questions related to the area. In addition, questions specific to the area can be filtered based on the user's geographical location information. This allows the reception unit to preferentially accept highly relevant questions in consideration of the user's geographical location information.

[0110] When receiving a question, the reception unit can analyze the user's social media activity and receive related questions. Social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The reception unit analyzes the user's social media activity and identifies topics of interest. For example, the reception unit preferentially receives questions related to topics that the user frequently mentions on social media. In addition, the reception unit can identify topics of interest from the user's social media activity and receive questions related to those topics. This allows the reception unit to analyze the user's social media activity and receive related questions.

[0111] The providing unit can estimate the user's emotion and adjust the answer providing method based on the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The providing unit monitors the user's emotion in real time and sets an appropriate providing method. For example, the providing unit provides an answer including a detailed explanation when the user is relaxed. Also, when the user is in a hurry, it can provide a concise answer that is to the point. This allows the providing unit to adjust the answer providing method based on the user's emotion.

[0112] When providing an answer, the providing unit can select an optimal providing method by referring to the user's past question history. The optimal providing method includes, for example, filtering based on the past history and selection based on the user's preferences. The providing unit analyzes the user's past question history and finds a specific pattern. For example, the providing unit preferentially provides answers related to topics that the user has frequently asked questions about in the past. In addition, the providing unit can select a providing method based on a specific pattern from the user's past question history. This allows the providing unit to select an optimal providing method by referring to the user's past question history.

[0113] When providing an answer, the providing unit can customize the answer based on the user's current field of interest. The field of interest includes, for example, the user's search history and survey results. The providing unit analyzes the user's current field of interest and provides highly relevant answers preferentially. For example, the providing unit provides answers related to topics in which the user is currently interested preferentially. In addition, the providing unit can customize highly relevant answers based on the user's current field of interest. This allows the providing unit to customize answers based on the user's current field of interest.

[0114] The providing unit can estimate the user's emotions and determine the priority of answers based on the estimated user's emotions. The estimation of emotions is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The providing unit monitors the user's emotions in real time and sets appropriate priorities. For example, when the user is excited, the providing unit provides answers that are interesting with priority. Also, when the user is relaxed, it can provide answers that are suitable for learning with priority. This allows the providing unit to determine the priority of answers based on the user's emotions.

[0115] When providing an answer, the providing unit can provide an optimal answer by taking into consideration the geographical location information of the user. The optimal answer includes, for example, filtering based on the geographical location information and selection based on the user's field of interest. The providing unit analyzes the geographical location information of the user and preferentially provides answers related to the region. For example, when the user is in a specific region, the providing unit preferentially provides answers related to the region. In addition, the providing unit can customize region-specific answers based on the geographical location information of the user. This allows the providing unit to provide an optimal answer by taking into consideration the geographical location information of the user.

[0116] When providing an answer, the providing unit can analyze the user's social media activity and provide a related answer. The social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The providing unit analyzes the user's social media activity and identifies topics of interest. For example, the providing unit preferentially provides answers related to topics that the user frequently mentions on social media. In addition, the providing unit can identify topics of interest from the user's social media activity and provide answers related to the topics. This allows the providing unit to analyze the user's social media activity and provide related answers.

[0117] The update unit can estimate the user's emotion and adjust the update method of the learning model based on the estimated user's emotion. The emotion estimation is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The update unit monitors the user's emotion in real time and sets an appropriate update method. For example, when the user is relaxed, the update unit collects detailed feedback and updates the learning model. Also, when the user is in a hurry, the update unit can collect brief feedback and update the learning model. This allows the update unit to adjust the update method of the learning model based on the user's emotion.

[0118] When updating the learning model, the update unit can select an optimal update method by referring to past learning data. The optimal update method includes, for example, selection based on past learning data and selection based on user feedback. The update unit analyzes past learning data and finds a specific pattern. For example, the update unit analyzes past learning data and selects the most effective update method. In addition, the update unit can select an update method based on a specific pattern from the past learning data. This allows the update unit to select an optimal update method by referring to past learning data.

[0119] The update unit can filter the update data based on the user's current field of interest when updating the learning model. Filtering of the update data includes, for example, filtering based on the user's field of interest and filtering based on past history. The update unit analyzes the user's current field of interest and preferentially updates highly relevant data. For example, the update unit preferentially updates data related to topics in which the user is currently interested. In addition, highly relevant data can be filtered based on the user's current field of interest. This allows the update unit to filter the update data based on the user's current field of interest.

[0120] The update unit can estimate the user's emotion and adjust the update frequency of the learning model based on the estimated user's emotion. The emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The update unit monitors the user's emotion in real time and sets an appropriate update frequency. For example, the update unit updates the learning model frequently when the user is relaxed. Also, when the user is in a hurry, the update unit can update the learning model with a reduced update frequency. This allows the update unit to adjust the update frequency of the learning model based on the user's emotion.

[0121] The update unit can select optimal update data in consideration of the geographical location information of the user when updating the learning model. The optimal update data includes, for example, selection based on geographical location information and selection based on the user's field of interest. The update unit analyzes the user's geographical location information and preferentially updates data related to the region. For example, when the user is in a specific region, the update unit preferentially updates data related to the region. In addition, region-specific data can be filtered based on the user's geographical location information. This allows the update unit to select optimal update data in consideration of the user's geographical location information.

[0122] The update unit can analyze the user's social media activity and select relevant update data when updating the learning model. The social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The update unit analyzes the user's social media activity and identifies topics of interest. For example, the update unit preferentially updates data related to topics that the user frequently mentions on social media. It is also possible to identify topics of interest from the user's social media activity and update data related to the topics. This allows the update unit to analyze the user's social media activity and select relevant update data.

[0123] The management unit can estimate the user's emotions and adjust the dialogue flow based on the estimated user's emotions. The emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. The management unit monitors the user's emotions in real time and sets an appropriate dialogue flow. For example, the management unit provides a dialogue flow including detailed explanations when the user is relaxed. Also, when the user is in a hurry, it can provide a dialogue flow that is concise and to the point. This allows the management unit to adjust the dialogue flow based on the user's emotions.

[0124] When managing the dialogue flow, the management unit can select an optimal flow by referring to the user's past dialogue history. The optimal flow includes, for example, selection based on the past dialogue history and selection based on the user's feedback. The management unit analyzes the user's past dialogue history and finds a specific pattern. For example, the management unit preferentially provides dialogue flows that the user has frequently performed in the past. In addition, the management unit can select a dialogue flow based on a specific pattern from the user's past dialogue history. This allows the management unit to select an optimal flow by referring to the user's past dialogue history.

[0125] The management unit can customize the dialogue based on the user's current field of interest when managing the dialogue flow. The customization of the dialogue includes, for example, customization based on the user's field of interest and customization based on past history. The management unit analyzes the user's current field of interest and provides highly relevant dialogues preferentially. For example, the management unit provides a dialogue flow related to a topic in which the user is currently interested preferentially. The management unit can also customize highly relevant dialogues based on the user's current field of interest. This allows the management unit to customize the dialogue based on the user's current field of interest.

[0126] The management unit can estimate the user's emotions and determine the priority of dialogues based on the estimated user's emotions. The estimation of emotions is realized by using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. The management unit monitors the user's emotions in real time and sets appropriate priorities. For example, when the user is excited, the management unit can provide an interesting dialogue with priority. Also, when the user is relaxed, the management unit can provide a dialogue suitable for learning with priority. This allows the management unit to determine the priority of dialogues based on the user's emotions.

[0127] The management unit can provide an optimal dialogue taking into consideration the geographic location information of the user when managing the dialogue flow. The optimal dialogue includes, for example, selection based on the geographic location information or selection based on the user's field of interest. The management unit analyzes the user's geographic location information and preferentially provides dialogue related to the region. For example, when the user is in a specific region, the management unit preferentially provides dialogue related to the region. Region-specific dialogue can also be customized based on the user's geographic location information. This allows the management unit to provide an optimal dialogue taking into consideration the user's geographic location information.

[0128] During dialogue flow management, the management unit can analyze the user's social media activity and provide related dialogues. Social media activity includes, for example, the content of posts, the number of likes, the number of followers, and the like. The management unit analyzes the user's social media activity and identifies topics of interest. For example, the management unit preferentially provides dialogues related to topics that the user frequently mentions on social media. It is also possible to identify topics of interest from the user's social media activity and provide dialogues related to those topics. This allows the management unit to analyze the user's social media activity and provide related dialogues.

[0129] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0130] The acquisition unit can analyze the user's past question history and acquire optimal question and answer pairs. For example, the acquisition unit preferentially acquires topics about which the user has frequently asked questions in the past. In addition, the acquisition unit can select an acquisition method based on a specific pattern from the user's past question history. This allows the acquisition unit to analyze the user's past question history and acquire optimal question and answer pairs.

[0131] The management unit can estimate the user's emotion and adjust the dialogue flow based on the estimated user's emotion. For example, the management unit can provide a dialogue flow including detailed explanations when the user is relaxed. Also, the management unit can provide a dialogue flow that is concise and to the point when the user is in a hurry. This allows the management unit to adjust the dialogue flow based on the user's emotion.

[0132] The providing unit can provide highly relevant question and answer pairs preferentially in consideration of the geographical location information of the user. For example, when the user is in a specific region, the providing unit can provide question and answer pairs related to the region preferentially. In addition, region-specific question and answer pairs can be filtered based on the geographical location information of the user. This allows the providing unit to provide highly relevant question and answer pairs preferentially in consideration of the geographical location information of the user.

[0133] The update unit can estimate the user's emotion and adjust the update frequency of the learning model based on the estimated user's emotion. For example, the update unit can update the learning model frequently when the user is relaxed. Also, when the user is in a hurry, the update unit can update the learning model with a reduced update frequency. This allows the update unit to adjust the update frequency of the learning model based on the user's emotion.

[0134] When generating an answer, the generating unit can apply different generating algorithms depending on the category of the question. For example, the generating unit can apply a generating algorithm having technical expertise to a technical question. Also, the generating unit can apply a generating algorithm having general knowledge to a general question. This allows the generating unit to apply the optimal generating algorithm depending on the category of the question.

[0135] The providing unit can estimate the user's emotion and adjust the method of providing an answer based on the estimated user's emotion. For example, the providing unit can provide an answer including a detailed explanation when the user is relaxed. Also, the providing unit can provide a concise answer that is to the point when the user is in a hurry. This allows the providing unit to adjust the method of providing an answer based on the user's emotion.

[0136] The management unit can analyze the user's social media activities and provide related interactions when managing the flow of interactions. For example, the management unit can provide interactions related to topics that the user frequently mentions on social media with a priority. It can also identify topics of interest from the user's social media activities and provide interactions related to those topics. This allows the management unit to analyze the user's social media activities and provide related interactions.

[0137] The acquisition unit can filter question and answer pairs based on the user's current field of interest when acquiring the question and answer pairs. For example, the acquisition unit preferentially acquires question and answer pairs related to topics in which the user is currently interested. Also, the acquisition unit can filter question and answer pairs with high relevance based on the user's current field of interest. This allows the acquisition unit to filter question and answer pairs based on the user's current field of interest.

[0138] The generation unit can estimate the user's emotions and adjust the way the answer is expressed based on the estimated user's emotions. For example, the generation unit can generate a polite and detailed answer when the user is relaxed. Also, the generation unit can generate a concise answer that gets to the point when the user is in a hurry. This allows the generation unit to adjust the way the answer is expressed based on the user's emotions.

[0139] When updating the learning model, the update unit can select optimal update data in consideration of the geographical location information of the user. For example, when the user is in a specific area, the update unit preferentially updates data related to the area. In addition, the update unit can filter area-specific data based on the geographical location information of the user. This allows the update unit to select optimal update data in consideration of the geographical location information of the user.

[0140] The process flow of the second embodiment will be briefly described below.

[0141] Step 1: The acquisition unit acquires information on question and answer pairs. The question and answer pairs may be in text format, audio format, or on a specific topic. The acquisition unit has a web scraping function that automatically collects question and answer pairs from the Internet, and is realized by using a library such as BeautifulSoup or Scrapy. Step 2: The generation unit uses the generation AI to generate a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit. The generation AI generates a learning model such as a neural network, a decision tree, or a support vector machine. Step 3: The reception unit receives a question from the user. The reception unit can receive the question by a method such as text input or voice input. Step 4: The providing unit uses the learning model generated by the generating unit to provide the user with an answer to the question accepted by the accepting unit. The providing unit has a function of adjusting the output speed of the answer according to the input speed of the user. Step 5: The update unit evaluates the learning model and updates the learning model based on the evaluation result. The update unit includes a feedback collection unit for collecting user feedback and updating the learning model based on the feedback. The update unit also has a function of periodically acquiring the latest question and answer pairs from an external database and updating the learning model.

[0142] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice 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 voice data.

[0143] 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<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. 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 a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) are input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may also be an AI agent. In addition, when the processes of each of the above-mentioned parts are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples.In addition, processing performed by AI including the generation AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI including the generation AI.

[0144] Furthermore, the above-described processing by the data processing system 10 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 required for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or an external device.

[0145] For example, the acquisition unit is realized by the specific processing unit 290 of the data processing device 12. For example, the generation unit is realized by the specific processing unit 290 of the data processing device 12. For example, the reception unit is realized by the control unit 46A of the smart device 14. For example, the provision unit is realized by the specific processing unit 290 of the data processing device 12. For example, the update unit is realized by the specific processing unit 290 of the data processing device 12. For example, the management unit is realized by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0146] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0147] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0148] 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, the RAM 30, and the 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.

[0149] 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.

[0150] 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 the voice according to instructions from the processor 46.

[0151] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0152] 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 performed in a secure state.

[0153] Fig. 4 shows an example of 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.

[0154] 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.

[0155] 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0156] In the smart glasses 214, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. 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 glasses 214 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 also perform processing similar to that of the specific processing unit 290 using these models.

[0157] 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 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0158] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the 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.

[0159] 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0160] 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 be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, 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, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device.

[0161] For example, the acquisition unit is realized by the specific processing unit 290 of the data processing device 12. For example, the generation unit is realized by the specific processing unit 290 of the data processing device 12. For example, the reception unit is realized by the control unit 46A of the smart glasses 214. For example, the provision unit is realized by the specific processing unit 290 of the data processing device 12. For example, the update unit is realized by the specific processing unit 290 of the data processing device 12. For example, the management unit is realized by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0162] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0163] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0164] 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, the RAM 30, and the 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.

[0165] 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.

[0166] 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 the voice according to instructions from the processor 46.

[0167] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0168] 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 performed in a secure state.

[0169] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type 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.

[0170] 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.

[0171] 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0172] In the headset type terminal 314, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50, and executes the read specific 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 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.

[0173] 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 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0174] 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 voice indicating a user input for 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.

[0175] 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0176] 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 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. In addition, 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, and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device.

[0177] For example, the acquisition unit is realized by the specific processing unit 290 of the data processing device 12. For example, the generation unit is realized by the specific processing unit 290 of the data processing device 12. For example, the reception unit is realized by the control unit 46A of the headset type terminal 314. For example, the provision unit is realized by the specific processing unit 290 of the data processing device 12. For example, the update unit is realized by the specific processing unit 290 of the data processing device 12. For example, the management unit is realized by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0178] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0179] 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.

[0180] 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, the RAM 30, and the 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.

[0181] 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.

[0182] 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 the voice according to instructions from the processor 46.

[0183] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of a typical healthy person).

[0184] 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 performed in a secure state.

[0185] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and legs. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, and legs. 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.

[0186] 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.

[0187] 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.

[0188] 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0189] In the robot 414, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads out the specific program 60 from the storage 50, and executes the specific 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 program 60 executed on the RAM 48. The robot 414 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.

[0190] 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 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0191] 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 a voice indicating a user input for 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.

[0192] 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0193] 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 be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, 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.

[0194] For example, the acquisition unit is realized by the specific processing unit 290 of the data processing device 12. For example, the generation unit is realized by the specific processing unit 290 of the data processing device 12. For example, the reception unit is realized by the control unit 46A of the robot 414. For example, the provision unit is realized by the specific processing unit 290 of the data processing device 12. For example, the update unit is realized by the specific processing unit 290 of the data processing device 12. For example, the management unit is realized by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0195] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0196] FIG. 9 is a diagram showing 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. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0197] These emotions are distributed in the three o'clock direction of emotion map 400 and usually 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.

[0198] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out you go on emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0199] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0200] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this is 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 the positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0201] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the 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, "relief," "calm," and "encouraging," have similar emotion values.

[0202] 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.

[0203] 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 Universal Serial Bus (USB) 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.

[0204] In addition, 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 upon request from the data processing device 12.

[0205] 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.

[0206] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing a specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[0207] The hardware resource that executes the specific process may be one of these various processors, or may be 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.

[0208] As an example of a configuration using one 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 configuration using a processor that realizes the functions of the 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.

[0209] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[0210] In the above example, the first to fourth embodiments are described separately, but some or all of these embodiments may be combined. The smart device 14, the smart glasses 214, the headset terminal 314, and the robot 414 are only examples, and they may be combined with each other, or may be other devices. In the above example, the first and second embodiments are described separately, but they may be combined.

[0211] The above description and illustrations are detailed descriptions 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, function, action, and effect is an example of the configuration, function, action, and effect 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 replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require particular explanation in enabling the implementation of the technology of the present disclosure.

[0212] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0213] (Appendix 1) an acquisition unit for acquiring the question and answer pair information; a generation unit that generates a learning model that receives a question and outputs an answer based on the information acquired by the acquisition unit; A reception unit that receives questions from users; a providing unit that provides a user with an answer to the question accepted by the accepting unit, using the learning model generated by the generating unit; An update unit that evaluates the learning model and updates the learning model based on the evaluation result. (Appendix 2) The system described in Appendix 1, further comprising a management unit that manages the flow of dialogue with a user, and a providing unit that provides the user with follow-up questions in accordance with the flow of dialogue managed by the management unit. (Appendix 3) The providing unit is Generate follow-up questions using generative AI 3. The system of claim 2, (Appendix 4) The system described in Appendix 1, wherein the acquisition unit has a web scraping function that automatically collects question and answer pairs from the Internet. (Appendix 5) The update unit is a feedback collection unit for collecting feedback from the users and updating the learning model based on the feedback; 2. The system of claim 1, (Appendix 6) The update unit is The function of periodically obtaining the latest question and answer pairs from an external database and updating the learning model. 2. The system of claim 1, (Appendix 7) The providing unit is A function for adjusting the speed at which the answer is output in response to the user's input speed 2. The system of claim 1, (Appendix 8) The management unit Learning the user's dialogue style and generating follow-up questions according to the dialogue style 3. The system of claim 2, (Appendix 9) The management unit The function of analyzing the frequency and intervals of speech of the user during the conversation and measuring the level of interest of the user is provided. 3. The system of claim 2, (Appendix 10) The providing unit is Highlighting keywords in the user's dialogue and generating answers based on the keywords 2. The system of claim 1, (Appendix 11) The acquisition unit is A method for estimating a user's emotion and adjusting the timing of acquiring question and answer pairs based on the estimated user's emotion 2. The system of claim 1, (Appendix 12) The acquisition unit is When acquiring question and answer pairs, the system analyzes the user's past question history and selects the optimal acquisition method. 2. The system of claim 1, (Appendix 13) The acquisition unit is When retrieving question and answer pairs, filter them based on the user's current areas of interest 2. The system of claim 1, (Appendix 14) The acquisition unit is Estimate the user's sentiment and prioritize question-answer pairs to be retrieved based on the estimated user's sentiment. 2. The system of claim 1, (Appendix 15) The acquisition unit is When fetching question and answer pairs, the system takes into account the user's geographic location to prioritize the most relevant pairs. 2. The system of claim 1, (Appendix 16) The acquisition unit is Upon obtaining question and answer pairs, the system analyzes the user's social media activity to obtain relevant pairs. 2. The system of claim 1, (Appendix 17) The generation unit is Infer user sentiment and adjust answer expression based on the inferred sentiment 2. The system of claim 1, (Appendix 18) The generation unit is When generating answers, adjust the detail of the answer based on the importance of the question 2. The system of claim 1, (Appendix 19) The generation unit is When generating answers, different generation algorithms are applied depending on the question category. 2. The system of claim 1, (Appendix 20) The generation unit is Infer user sentiment and adjust the length of answers based on the estimated user sentiment 2. The system of claim 1, (Appendix 21) The generation unit is When generating answers, prioritize answers based on when the question was submitted 2. The system of claim 1, (Appendix 22) The generation unit is During answer generation, adjust the order of answers based on the relevance of the question 2. The system of claim 1, (Appendix 23) The reception unit is Estimate the user's emotions and adjust the way questions are accepted based on the estimated user's emotions 2. The system of claim 1, (Appendix 24) The reception unit is When accepting a question, the system selects the most appropriate method of acceptance by referring to the user's past question history. 2. The system of claim 1, (Appendix 25) The reception unit is When questions are asked, filter them based on the user's current areas of interest 2. The system of claim 1, (Appendix 26) The reception unit is Estimate user sentiment and prioritize questions based on the estimated user sentiment 2. The system of claim 1, (Appendix 27) The reception unit is When accepting questions, the system takes into account the user's geographic location information and prioritizes questions that are highly relevant. 2. The system of claim 1, (Appendix 28) The reception unit is When a question is received, the user's social media activity is analyzed and related questions are received. 2. The system of claim 1, (Appendix 29) The providing unit is Inferring user sentiment and adjusting answer delivery based on the inferred user sentiment 2. The system of claim 1, (Appendix 30) The providing unit is When providing an answer, the system selects the most appropriate method by referring to the user's past question history. 2. The system of claim 1, (Appendix 31) The providing unit is When providing answers, customize the answers based on the user's current areas of interest 2. The system of claim 1, (Appendix 32) The providing unit is Infer user sentiment and prioritize answers based on the inferred sentiment 2. The system of claim 1, (Appendix 33) The providing unit is When providing answers, the system takes into account the user's geographic location to provide the most appropriate answer. 2. The system of claim 1, (Appendix 34) The providing unit is When providing an answer, analyze the user's social media activity to provide a relevant answer 2. The system of claim 1, (Appendix 35) The update unit is Estimate user emotions and adjust the method of updating the learning model based on the estimated user emotions 2. The system of claim 1, (Appendix 36) The update unit is When updating a learning model, the optimal update method is selected by referring to past learning data. 2. The system of claim 1, (Appendix 37) The update unit is When updating learning models, filter the update data based on the user's current areas of interest 2. The system of claim 1, (Appendix 38) The update unit is Estimate user emotions and adjust the update frequency of the learning model based on the estimated user emotions. 2. The system of claim 1, (Appendix 39) The update unit is When updating a learning model, the system selects the most appropriate update data taking into account the user's geographic location information. 2. The system of claim 1, (Appendix 40) The update unit is When updating the learning model, analyze users' social media activities and select relevant update data. 2. The system of claim 1, (Appendix 41) The management unit Estimate the user's emotions and adjust the flow of the dialogue based on the estimated user's emotions. 3. The system of claim 2, (Appendix 42) The management unit When managing the dialogue flow, select the optimal flow by referring to the user's past dialogue history. 3. The system of claim 2, (Appendix 43) The management unit During dialogue flow management, customizing dialogue based on the user's current areas of interest 3. The system of claim 2, (Appendix 44) The management unit Estimate user emotions and prioritize interactions based on the estimated user emotions 3. The system of claim 2, (Appendix 45) The management unit When managing dialogue flow, provide optimal dialogue by considering the user's geographic location information 3. The system of claim 2, (Appendix 46) The management unit Analyzing users' social media activity and providing relevant conversations during conversation flow management 3. The system of claim 2, [Explanation of symbols]

[0214] 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. an acquisition unit for acquiring question and answer pair information; a generation unit that generates a learning model that inputs a question and outputs an answer based on the information acquired by the acquisition unit; A reception unit that receives questions from users; a providing unit that provides the user with an answer to the question accepted by the accepting unit, using the learning model generated by the generating unit; An update unit that evaluates the learning model and updates the learning model based on a result of the evaluation. A system characterized by:

2. A management unit that manages a flow of dialogue with the user, The providing unit is Providing the user with follow-up questions according to the flow of the dialogue managed by the management unit 2. The system of claim 1 .

3. The providing unit is Generate follow-up questions using generative AI 3. The system of claim 2.

4. The acquisition unit is The system according to claim 1, further comprising a web scraping function for automatically collecting question and answer pairs from the Internet.

5. The update unit is a feedback collection unit for collecting feedback from the users and updating the learning model based on the feedback; 2. The system of claim 1 .

6. The update unit is The function of periodically obtaining the latest question and answer pairs from an external database and updating the learning model.

2. The system of claim 1 .

7. The providing unit is A function for adjusting the speed at which the answer is output in response to the user's input speed 2. The system of claim 1 .

8. The management unit Learning the user's dialogue style and generating follow-up questions according to the dialogue style 3. The system of claim 2.

9. The management unit The function of analyzing the frequency and intervals of speech of the user during the conversation and measuring the level of interest of the user is provided.

3. The system of claim 2.

10. The providing unit is Highlighting keywords in the user's dialogue and generating answers based on the keywords 2. The system of claim 1 .

Images