system

Abstract

The system according to this embodiment aims to improve resource efficiency by having agents respond to inquiries and questions. [Solution] The system according to the embodiment comprises a collection unit, an analysis unit, a generation unit, a provision unit, and a correction unit. The collection unit collects question-and-answer patterns from past emails and communication tool history. The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and answer. The generation unit allows the agent to generate answers based on the information generated by the analysis unit. The provision unit provides the answers generated by the generation unit. The correction unit allows a human to provide corrected answers when corrections are needed to the answers provided by the provision unit, and accumulates these corrections so that the agent can refine future answer patterns.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the prior art, there was a problem that responses to inquiries and question - answer sessions consumed unplanned resources.

[0005] The system according to the embodiment aims to improve the efficiency of resources by having an agent respond to inquiries and question - answer sessions.

Means for Solving the Problems

[0006] The system according to this embodiment comprises a collection unit, an analysis unit, a generation unit, a provision unit, and a correction unit. The collection unit collects question-and-answer patterns from past emails and communication tool history. The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and answer. The generation unit allows the agent to generate answers based on the information generated by the analysis unit. The provision unit provides the answers generated by the generation unit. The correction unit allows a human to provide corrected answers if corrections are needed to the answers provided by the provision unit, and accumulates these corrections so that the agent can refine future answer patterns. [Effects of the Invention]

[0007] The system according to this embodiment can improve resource efficiency by having agents respond to inquiries and questions. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

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

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

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

[0014] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, etc. The communication I / F manages communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards such as 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". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

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

[0017] As shown in FIG. 1, the 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. Also, the database 24 and the communication I / F 26 are 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 comprises a computer 36, a receiving device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The receiving device 38, output device 40, and camera 42 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

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

[0025] Storage 32 stores the data generation model 58 and the 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 user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

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

[0027] Furthermore, other devices besides the data processing device 12 may also 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 processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example of form 1) The agent system according to an embodiment of the present invention is a system for solving the problem of unplanned resource consumption caused by answering inquiries and questions that arise daily, even in departments other than those whose "job is to receive inquiries," such as customer service or support desks. This agent system is added as an option to email and communication tools, and the agent can determine the type of inquiry from the history of past emails and communication tools and automatically provide an answer. The answer is also notified to the staff member in question, so corrections can be made as needed. This system utilizes generative AI to learn the patterns of questions and answers in past emails and communication tools, and the agent can intervene and provide an answer. If corrections are needed to the answer, a human will provide a corrected answer, and this information can be accumulated so that the agent can refine future answer patterns and improve accuracy. By introducing this system, employees will be freed from miscellaneous tasks and will be able to concentrate on their core work. Furthermore, in today's world, where digital communication is increasing due to the rise of remote work, the digital accumulation of knowledge communication is increasing, and the amount of learning for generative AI will be maximized, so it is considered that now is the time to introduce this system. The vision for this system is to free all workers from mundane tasks, increasing the time they can dedicate to creative tasks and essential missions. It is expected that implementing this system will enable companies to improve operational efficiency and increase employee satisfaction. In this way, the agent system can optimize employee resources and streamline operations.

[0029] The agent system according to this embodiment comprises a collection unit, an analysis unit, a generation unit, a provision unit, and a correction unit. The collection unit collects question-and-answer patterns from past email and communication tool history. The collection unit, for example, analyzes past email and communication tool history and extracts question-and-answer patterns. The collection unit can also learn and collect past question-and-answer patterns using a generation AI. For example, the collection unit inputs past email and communication tool history into the generation AI and extracts question-and-answer patterns. The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and answer. The analysis unit, for example, analyzes the collected patterns and generates information for the agent to answer. The analysis unit can also analyze the collected patterns and generate information using a generation AI. For example, the analysis unit inputs the collected patterns into the generation AI and generates information for the agent to answer. The generation unit allows the agent to generate answers based on the information generated by the analysis unit. The generation unit allows the agent to generate answers based on the analyzed information. The generation unit can also generate answers based on the analyzed information using a generation AI. For example, the generation unit inputs the analyzed information into the generation AI and generates an answer. The provision unit provides the answers generated by the generation unit. For example, the provision unit provides the generated answers via email or communication tools. The provision unit can also provide the generated answers using a generation AI. For example, the provision unit inputs the generated answers into the generation AI and provides them. The correction unit allows a human to provide corrected answers when corrections are needed to the answers provided by the provision unit, and accumulates these corrections so that the agent can refine future answer patterns. For example, the correction unit can learn from corrected answers using a generation AI and refine future answer patterns. As a result, the agent system according to this embodiment can automate question and answer and improve resource efficiency.

[0030] The data collection unit collects question-and-answer patterns from past email and communication tool histories. Specifically, the unit accesses the company's email servers and communication tool databases and analyzes past interactions. This includes a process that uses natural language processing technology to extract email and chat content as text data and identify question-and-answer patterns. For example, metadata such as email subject lines and body text, the relationship between sender and recipient, and the frequency and timing of interactions are also analyzed. The data collection unit centrally manages this data and builds a dataset for extracting question-and-answer patterns. Furthermore, the data collection unit can also use generative AI to learn and collect past question-and-answer patterns. Generative AI has the ability to process large amounts of text data and automatically extract question-and-answer patterns. For example, past email and communication tool histories are input into the generative AI to extract question-and-answer patterns. In this process, the generative AI analyzes the text data, identifies pairs of questions and answers, and learns them as patterns. This allows the data collection unit to efficiently collect question-and-answer patterns from past interactions and provide the data that forms the basis of the agent system.

[0031] The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and provide answers. Specifically, the analysis unit analyzes the collected question-and-answer patterns in detail and generates information for the agent to intervene at the appropriate time and provide accurate answers. The analysis unit uses natural language processing techniques and machine learning algorithms to analyze the collected patterns. For example, the analysis unit clusters the collected patterns and groups similar questions and answers. This provides the agent with information to select the most appropriate answer to a particular question. Furthermore, the analysis unit can also use generative AI to analyze the collected patterns and generate information. The generative AI takes the collected patterns as input and generates information for the agent to answer. For example, the generative AI analyzes the collected patterns, understands the intent of the question, and generates the optimal answer. In this process, the generative AI considers the context and background information of the question to improve the quality of the answer provided by the agent. This allows the analysis unit to efficiently analyze the collected data and provide information for the agent to answer accurately and quickly.

[0032] The generation unit allows agents to generate answers based on information generated by the analysis unit. Specifically, the generation unit is responsible for the process by which agents generate answers based on analyzed information. The generation unit uses natural language generation technology to express the analyzed information in natural language and generate answers to provide to the user. For example, the generation unit receives information provided by the analysis unit as input and generates an appropriate answer to a question. In this process, the generation AI plays a crucial role. The generation AI generates the optimal answer to a question based on the analyzed information. For example, the analyzed information is input to the generation AI, and an answer is generated. In this process, the generation AI considers the context and background information of the question and generates an answer expressed in natural language. The generation unit can also evaluate the quality of the generated answer and make corrections as needed. This allows the generation unit to provide users with high-quality answers. Furthermore, the generation unit integrates the generated answers with other parts of the agent system to provide a seamless user experience. For example, the generation unit sends the generated answers to the delivery unit to prepare them for provision to the user. This allows the generation unit to efficiently manage the answer generation process for the entire agent system and provide users with fast and accurate answers.

[0033] The delivery unit provides the answers generated by the generation unit. Specifically, the delivery unit is responsible for delivering the generated answers to users. The delivery unit provides the generated answers to users via email or communication tools. For example, the delivery unit inserts the generated answers into the body of an email and sends it to the user. It can also provide answers in a chat format using communication tools. The delivery unit can also automate the process of providing generated answers using generation AI. For example, the delivery unit inputs the generated answers into the generation AI and provides them to the user in an appropriate format. In this process, the generation AI optimizes the format and content of the answers and provides them in an easy-to-understand manner for the user. Furthermore, the delivery unit can collect feedback from users and evaluate the quality of the answers. For example, users evaluate the provided answers, and the evaluation results are collected. This allows the delivery unit to continuously improve the quality of the generated answers and increase user satisfaction. The delivery unit works in conjunction with other parts of the agent system to provide a seamless user experience. For example, the delivery unit quickly provides answers received from the generation unit to users and sends user feedback to the correction unit. This allows the delivery unit to improve the efficiency and effectiveness of the entire agent system.

[0034] The Correction Unit allows humans to correct answers when necessary in the responses provided by the Provider Unit, accumulating these corrections to refine future response patterns for the agent. Specifically, the Correction Unit receives user feedback on the provided answers, and humans correct the answers as needed. In this process, the Correction Unit evaluates the accuracy and appropriateness of the provided answers and makes necessary corrections. For example, if a provided answer is inaccurate or does not match the user's intent, the Correction Unit corrects the answer to provide accurate information. The Correction Unit can also use Generative AI to learn from corrected answers and refine future response patterns. For example, the Correction Unit inputs corrected answers into Generative AI to refine future response patterns. In this process, Generative AI learns from corrected answers, enabling the agent to provide more accurate answers to future questions. Furthermore, the Correction Unit accumulates corrected answers in a database, strengthening the knowledge base of the entire agent system. This allows the Correction Unit to continuously improve the quality of responses in the agent system and increase user satisfaction. The Correction Unit works in conjunction with other parts of the agent system to provide a seamless user experience. For example, the correction unit sends feedback received from the provisioning unit to the analysis unit, which then incorporates it into future response generation. This allows the correction unit to improve the overall efficiency and effectiveness of the agent system.

[0035] The system includes a notification unit. The notification unit notifies the staff of the generated response content. The notification unit notifies the staff of the generated response content, for example, via email or a communication tool. The notification unit can also notify the staff of the generated response content using a generation AI. For example, the notification unit inputs the generated response content into the generation AI and notifies it. This allows staff to be notified of the generated response content and correct it as needed. Some or all of the above processing in the notification unit may be performed using the generation AI or not. For example, the notification unit can input the generated response content into the generation AI and have the generation AI execute the notification.

[0036] The system includes a learning unit. The learning unit learns corrected responses, and the agent refines future response patterns. For example, the learning unit collects corrected responses and the agent learns to refine future response patterns. The learning unit can also use a generative AI to learn corrected responses and refine future response patterns. For example, the learning unit inputs corrected responses into the generative AI and refines future response patterns. This improves the agent's response accuracy by learning corrected responses. Some or all of the above processing in the learning unit may be performed using a generative AI or not. For example, the learning unit can input corrected responses into the generative AI and have the generative AI perform the learning.

[0037] The system includes a knowledge base unit. The knowledge base unit accumulates patterns of past question-and-answer interactions and provides a database for agents to refer to. For example, the knowledge base unit collects patterns of past question-and-answer interactions and stores them in the database. The knowledge base unit can also use a generative AI to accumulate patterns of past question-and-answer interactions and provide a database. For example, the knowledge base unit inputs patterns of past question-and-answer interactions into a generative AI and generates a database. This provides a database that agents can refer to by accumulating patterns of past question-and-answer interactions. Some or all of the above-described processes in the knowledge base unit may be performed using a generative AI or not. For example, the knowledge base unit can input patterns of past question-and-answer interactions into a generative AI and have the generative AI generate the database.

[0038] The data collection unit can analyze the user's past communication history and select the optimal collection method when collecting questions and answers. For example, the data collection unit can collect relevant question and answer patterns based on keywords frequently used by the user in the past. The data collection unit can also collect question and answer patterns that occur during specific time periods from the user's past communication history. The data collection unit can also analyze the user's past communication history and collect question and answer patterns related to specific projects. This allows the optimal collection method to be selected by analyzing the user's past communication history. Some or all of the above processing in the data collection unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the data collection unit can input the user's past communication history into a generative AI and have the generative AI select the optimal collection method.

[0039] The collection unit can filter the collected questions and answers based on the user's current projects and areas of interest. For example, the collection unit can prioritize collecting question and answer patterns related to the user's current projects. The collection unit can also collect relevant question and answer patterns based on the user's areas of interest. The collection unit can also collect question and answer patterns related to topics the user is currently interested in. This allows for the collection of highly relevant questions and answers by filtering based on the user's current projects and areas of interest. Some or all of the above processing in the collection unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the collection unit can input data on the user's current projects and areas of interest into a generative AI and have the generative AI perform the filtering.

[0040] The collection unit can prioritize the collection of highly relevant patterns when collecting questions and answers, taking into account the user's geographical location information. For example, the collection unit can prioritize the collection of question and answer patterns related to the user's current location. The collection unit can also collect region-specific question and answer patterns based on the user's geographical location information. If the user is on the move, the collection unit can also prioritize the collection of question and answer patterns related to their current location. This allows for the priority collection of highly relevant questions and answers by considering the user's geographical location information. Some or all of the above processing in the collection unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the collection unit can input the user's geographical location information into a generation AI and have the generation AI perform the collection of questions and answers.

[0041] The collection unit can analyze the user's social media activity and collect relevant patterns when collecting questions and answers. For example, the collection unit can collect question and answer patterns related to topics that the user frequently mentions on social media. The collection unit can also collect question and answer patterns related to topics of high interest from the user's social media activity. The collection unit can also collect question and answer patterns related to accounts that the user follows on social media. In this way, relevant question and answer patterns can be collected by analyzing the user's social media activity. Some or all of the above processing in the collection unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the collection unit can input the user's social media activity data into a generative AI and have the generative AI perform the collection of questions and answers.

[0042] The analysis unit can adjust the level of detail of the analysis based on the importance of the questions and answers during the analysis. For example, the analysis unit can perform a detailed analysis for questions and answers of high importance. The analysis unit can also perform a concise analysis for questions and answers of low importance. The analysis unit can also adjust the level of detail of the analysis in stages according to the importance of the questions and answers. By adjusting the level of detail of the analysis based on the importance of the questions and answers, more appropriate analysis results can be provided. Some or all of the above processing in the analysis unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the analysis unit can input the importance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the level of detail of the analysis.

[0043] The analysis unit can apply different analysis algorithms depending on the category of the question and answer during analysis. For example, the analysis unit can apply a specialized analysis algorithm to technical questions and answers. The analysis unit can also apply a simpler analysis algorithm to general questions and answers. The analysis unit can also apply an analysis algorithm that prioritizes customer satisfaction to questions and answers related to customer support. By applying different analysis algorithms depending on the category of the question and answer, more appropriate analysis results can be provided. Some or all of the above processing in the analysis unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the analysis unit can input the category data of the question and answer into a generative AI and have the generative AI execute the application of the analysis algorithm.

[0044] The analysis unit can determine the priority of analysis based on the submission date of the questions and answers during the analysis. For example, the analysis unit may prioritize the analysis of recently submitted questions and answers. The analysis unit may also postpone the analysis of older questions and answers. The analysis unit can also adjust the priority of analysis in stages according to the submission date. This allows for the provision of more appropriate analysis results by determining the priority of analysis based on the submission date of the questions and answers. Some or all of the above processing in the analysis unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the analysis unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the determination of the analysis priority.

[0045] The analysis unit can adjust the order of analysis based on the relevance of the questions and answers during the analysis. For example, the analysis unit may prioritize the analysis of highly relevant questions and answers. The analysis unit may also postpone the analysis of less relevant questions and answers. The analysis unit can also adjust the order of analysis in stages according to the relevance of the questions and answers. By adjusting the order of analysis based on the relevance of the questions and answers, more appropriate analysis results can be provided. Some or all of the above processing in the analysis unit may be performed using a generating AI, or it may be performed without a generating AI. For example, the analysis unit can input the relevance data of the questions and answers into a generating AI and have the generating AI perform the adjustment of the order of analysis.

[0046] The generation unit can adjust the level of detail in the answers based on the importance of the questions and answers when generating responses. For example, the generation unit can generate detailed answers for high-importance questions and answers. The generation unit can also generate concise answers for low-importance questions and answers. The generation unit can also adjust the level of detail in the answers in stages according to the importance of the questions and answers. This allows for the generation of more appropriate answers by adjusting the level of detail in the answers based on the importance of the questions and answers. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input question and answer importance data into a generation AI and have the generation AI perform the adjustment of the level of detail in the answers.

[0047] The generation unit can apply different generation algorithms depending on the category of the question and answer when generating answers. For example, the generation unit can apply a specialized generation algorithm to technical questions and answers. The generation unit can also apply a simple generation algorithm to general questions and answers. The generation unit can also apply a generation algorithm that prioritizes customer satisfaction to questions and answers related to customer support. By applying different generation algorithms depending on the category of the question and answer, more appropriate answers can be generated. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input the category data of the question and answer into a generation AI and have the generation AI execute the application of the generation algorithm.

[0048] The generation unit can determine the priority of answers based on the submission date of the questions and answers when generating responses. For example, the generation unit can prioritize generating answers for recently submitted questions and answers. The generation unit can also postpone generating answers for older questions and answers. The generation unit can also adjust the priority of answers in stages according to the submission date. This allows for the generation of more appropriate answers by determining the priority of answers based on the submission date of the questions and answers. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the determination of the priority of answers.

[0049] The generation unit can adjust the order of answers based on the relevance of the questions and answers when generating responses. For example, the generation unit can prioritize generating responses to highly relevant questions and answers. The generation unit can also postpone generating responses to less relevant questions and answers. The generation unit can also adjust the order of answers in stages according to the relevance of the questions and answers. This allows for the generation of more appropriate answers by adjusting the order of answers based on the relevance of the questions and answers. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input the relevance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of answers.

[0050] The answering unit can adjust the level of detail provided based on the importance of the question and answer. For example, the answering unit can provide detailed answers to high-importance questions and answers. It can also provide concise answers to low-importance questions and answers. The answering unit can also adjust the level of detail in stages according to the importance of the questions and answers. This allows for the provision of more appropriate answers by adjusting the level of detail based on the importance of the questions and answers. Some or all of the above processing in the answering unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the answering unit can input question and answer importance data into a generation AI and have the generation AI perform the adjustment of the level of detail in the answer.

[0051] The response unit can apply different response algorithms depending on the category of the question and answer when providing answers. For example, the response unit can apply a specialized response algorithm to technical questions. The response unit can also apply a simpler response algorithm to general questions. The response unit can also apply a response algorithm that prioritizes customer satisfaction to questions related to customer support. By applying different response algorithms depending on the category of the question and answer, more appropriate answers can be provided. Some or all of the above processing in the response unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the response unit can input question and answer category data into a generative AI and have the generative AI execute the application of the response algorithm.

[0052] The service provider can adjust the order in which answers are provided based on the submission date of the questions and answers. For example, the service provider can prioritize answers to recently submitted questions and answers. The service provider can also postpone the provision of answers to older questions and answers. The service provider can also adjust the order of provision in stages according to the submission date. This allows for the provision of more appropriate answers by adjusting the order of provision based on the submission date of the questions and answers. Some or all of the above processing in the service provider may be performed using a generation AI, or it may be performed without a generation AI. For example, the service provider can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of provision.

[0053] The answering unit can adjust the order in which answers are provided based on the relevance of the questions and answers. For example, the answering unit can prioritize answers to highly relevant questions and answers. The answering unit can also postpone answers to less relevant questions and answers. The answering unit can also adjust the order of answers in stages according to the relevance of the questions and answers. This allows for the provision of more appropriate answers by adjusting the order of answers based on the relevance of the questions and answers. Some or all of the above processing in the answering unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the answering unit can input the relevance data of the questions and answers into a generative AI and have the generative AI perform the adjustment of the order of answers.

[0054] The correction unit can adjust the level of detail of the correction based on the importance of the question and answer during the correction process. For example, the correction unit can perform detailed corrections for high-importance questions and answers. The correction unit can also perform concise corrections for low-importance questions and answers. The correction unit can also adjust the level of detail of the correction in stages according to the importance of the questions and answers. This allows for more appropriate corrections by adjusting the level of detail of the correction based on the importance of the questions and answers. Some or all of the above-described processes in the correction unit may be performed using a generation AI, or they may be performed without a generation AI. For example, the correction unit can input question and answer importance data into a generation AI and have the generation AI perform the adjustment of the level of detail of the correction.

[0055] The correction unit can apply different correction algorithms depending on the category of the question and answer during the correction process. For example, the correction unit can apply a specialized correction algorithm to technical questions and answers. The correction unit can also apply a simpler correction algorithm to general questions and answers. The correction unit can also apply a correction algorithm that prioritizes customer satisfaction to questions and answers related to customer support. By applying different correction algorithms depending on the category of the question and answer, more appropriate corrections can be made. Some or all of the above processing in the correction unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the correction unit can input the category data of the question and answer into a generation AI and have the generation AI execute the application of the correction algorithm.

[0056] The correction unit can adjust the order of corrections based on the submission date of the questions and answers. For example, the correction unit can prioritize corrections to recently submitted questions and answers. The correction unit can also postpone corrections to older questions and answers. The correction unit can also adjust the order of corrections in stages according to the submission date. This allows for more appropriate corrections by adjusting the order of corrections based on the submission date of the questions and answers. Some or all of the above processing in the correction unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the correction unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of corrections.

[0057] The correction unit can adjust the order of corrections based on the relevance of the questions and answers during the correction process. For example, the correction unit can prioritize correcting highly relevant questions and answers. The correction unit can also postpone correcting less relevant questions and answers. The correction unit can also adjust the order of corrections in stages according to the relevance of the questions and answers. This allows for more appropriate corrections by adjusting the order of corrections based on the relevance of the questions and answers. Some or all of the above-described processes in the correction unit may be performed using a generation AI, or they may be performed without a generation AI. For example, the correction unit can input the relevance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of corrections.

[0058] The notification unit can adjust the level of detail of the notification based on the importance of the question and answer. For example, the notification unit can provide detailed notifications for high-importance questions and answers. It can also provide concise notifications for low-importance questions and answers. The notification unit can also adjust the level of detail of the notification in stages according to the importance of the questions and answers. This allows for more appropriate notifications to be provided by adjusting the level of detail of the notification based on the importance of the questions and answers. Some or all of the above processing in the notification unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the notification unit can input the importance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the level of detail of the notification.

[0059] The notification unit can adjust the order of notifications based on when the questions and answers were submitted. For example, the notification unit can prioritize notifications for recently submitted questions and answers. It can also postpone notifications for older questions and answers. The notification unit can also adjust the order of notifications in stages according to the submission date. This allows for more appropriate notifications by adjusting the order of notifications based on the submission date of the questions and answers. Some or all of the above processing in the notification unit may be performed using a generation AI, or not. For example, the notification unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of notifications.

[0060] The learning unit can optimize the learning algorithm by referring to past learning data during the learning process. For example, the learning unit can select the optimal learning algorithm based on past learning data. The learning unit can also analyze past learning data and adjust the parameters of the learning algorithm. The learning unit can also improve the accuracy of the learning algorithm by referring to past learning data. As a result, the accuracy of learning is improved by optimizing the learning algorithm by referring to past learning data. Some or all of the above processes in the learning unit may be performed using a generative AI, or they may be performed without a generative AI. For example, the learning unit can input past learning data into a generative AI and have the generative AI perform the optimization of the learning algorithm.

[0061] The learning unit can weight the training data based on the submission date of the questions and answers during training. For example, the learning unit can give higher weight to recently submitted questions and answers during training. The learning unit can also give lower weight to older questions and answers during training. The learning unit can also adjust the weighting of the training data in stages according to the submission date. This allows for more appropriate training by weighting the training data based on the submission date of the questions and answers. Some or all of the above processing in the learning unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the learning unit can input the submission date data of the questions and answers into a generative AI and have the generative AI perform the weighting of the training data.

[0062] The knowledge base unit can select the optimal update method by referring to past Q&A data when updating the knowledge base. For example, the knowledge base unit selects the optimal update method based on past Q&A data. The knowledge base unit can also analyze past Q&A data and optimize the knowledge base update method. The knowledge base unit can also adjust the update frequency of the knowledge base by referring to past Q&A data. This improves the accuracy of the knowledge base by selecting the optimal update method by referring to past Q&A data. Some or all of the above processing in the knowledge base unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the knowledge base unit can input past Q&A data into a generation AI and have the generation AI select the update method.

[0063] The knowledge base unit can adjust the update order based on the submission date of questions and answers when updating the knowledge base. For example, the knowledge base unit can prioritize updating the knowledge base for recently submitted questions and answers. The knowledge base unit can also postpone updating the knowledge base for older questions and answers. The knowledge base unit can also adjust the update order of the knowledge base in stages according to the submission date. This allows the knowledge base to be updated in a more appropriate order by adjusting the update order based on the submission date of the questions and answers. Some or all of the above processing in the knowledge base unit may be performed using a generation AI, or not. For example, the knowledge base unit can input the submission date data of questions and answers into a generation AI and have the generation AI perform the adjustment of the update order.

[0064] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0065] The agent system can further analyze a user's past behavior history and provide responses optimized for each individual user. For example, if a user has previously shown interest in a particular topic, it can prioritize providing detailed information related to that topic. If a user has previously preferred a specific response format, it can also generate responses in that format. This allows for more personalized responses based on the user's past behavior history.

[0066] The agent system can further customize its responses based on the user's current situation. For example, if the user is in a meeting, the agent can provide a concise and to-the-point answer. If the user is on the go, the agent can also provide an answer in a format optimized for mobile devices. This ensures that responses are provided in an appropriate format according to the user's current situation.

[0067] The agent system can further analyze past user feedback to improve the accuracy of its responses. For example, based on feedback previously provided by the user, the agent can adjust the content and format of its responses. If a user gives a high rating to a particular response, the agent can use that format and content as a reference to generate new responses. This allows the system to provide more accurate responses based on the user's past feedback.

[0068] The agent system can further customize its responses by taking into account the user's geographical location. For example, if the user is in a specific region, it can prioritize providing information relevant to that region. If the user is traveling, it can also provide information relevant to their travel destination. This allows the system to provide more relevant responses based on the user's geographical location.

[0069] The agent system can further analyze the user's social media activity and provide relevant information. For example, it can prioritize providing information related to topics the user frequently mentions on social media. It can also provide information related to accounts the user follows. This allows for the provision of more relevant information based on the user's social media activity.

[0070] The following briefly describes the processing flow for example form 1.

[0071] Step 1: The collection unit collects question-and-answer patterns from past emails and communication tool history. For example, the collection unit can use generative AI to learn and collect past question-and-answer patterns. Step 2: The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and respond. For example, the analysis unit can use a generation AI to analyze the collected patterns and generate information. Step 3: The generation unit uses the information generated by the analysis unit to generate an answer from the agent. For example, the generation unit can use a generation AI to generate an answer based on the analyzed information. Step 4: The providing unit provides the answers generated by the generating unit. For example, the providing unit can use a generation AI to provide the generated answers via email or communication tools. Step 5: The correction unit has a human correct the answer provided by the supply unit if correction is needed, and this is accumulated so that the agent can refine future answer patterns. For example, the correction unit can use a generation AI to learn from the corrected answers and refine future answer patterns.

[0072] (Example of form 2) The agent system according to an embodiment of the present invention is a system for solving the problem of unplanned resource consumption caused by answering inquiries and questions that arise daily, even in departments other than those whose "job is to receive inquiries," such as customer service or support desks. This agent system is added as an option to email and communication tools, and the agent can determine the type of inquiry from the history of past emails and communication tools and automatically provide an answer. The answer is also notified to the staff member in question, so corrections can be made as needed. This system utilizes generative AI to learn the patterns of questions and answers in past emails and communication tools, and the agent can intervene and provide an answer. If corrections are needed to the answer, a human will provide a corrected answer, and this information can be accumulated so that the agent can refine future answer patterns and improve accuracy. By introducing this system, employees will be freed from miscellaneous tasks and will be able to concentrate on their core work. Furthermore, in today's world, where digital communication is increasing due to the rise of remote work, the digital accumulation of knowledge communication is increasing, and the amount of learning for generative AI will be maximized, so it is considered that now is the time to introduce this system. The vision for this system is to free all workers from mundane tasks, increasing the time they can dedicate to creative tasks and essential missions. It is expected that implementing this system will enable companies to improve operational efficiency and increase employee satisfaction. In this way, the agent system can optimize employee resources and streamline operations.

[0073] The agent system according to this embodiment comprises a collection unit, an analysis unit, a generation unit, a provision unit, and a correction unit. The collection unit collects question-and-answer patterns from past email and communication tool history. The collection unit, for example, analyzes past email and communication tool history and extracts question-and-answer patterns. The collection unit can also learn and collect past question-and-answer patterns using a generation AI. For example, the collection unit inputs past email and communication tool history into the generation AI and extracts question-and-answer patterns. The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and answer. The analysis unit, for example, analyzes the collected patterns and generates information for the agent to answer. The analysis unit can also analyze the collected patterns and generate information using a generation AI. For example, the analysis unit inputs the collected patterns into the generation AI and generates information for the agent to answer. The generation unit allows the agent to generate answers based on the information generated by the analysis unit. The generation unit allows the agent to generate answers based on the analyzed information. The generation unit can also generate answers based on the analyzed information using a generation AI. For example, the generation unit inputs the analyzed information into the generation AI and generates an answer. The provision unit provides the answers generated by the generation unit. For example, the provision unit provides the generated answers via email or communication tools. The provision unit can also provide the generated answers using a generation AI. For example, the provision unit inputs the generated answers into the generation AI and provides them. The correction unit allows a human to provide corrected answers when corrections are needed to the answers provided by the provision unit, and accumulates these corrections so that the agent can refine future answer patterns. For example, the correction unit can learn from corrected answers using a generation AI and refine future answer patterns. As a result, the agent system according to this embodiment can automate question and answer and improve resource efficiency.

[0074] The data collection unit collects question-and-answer patterns from past email and communication tool histories. Specifically, the unit accesses the company's email servers and communication tool databases and analyzes past interactions. This includes a process that uses natural language processing technology to extract email and chat content as text data and identify question-and-answer patterns. For example, metadata such as email subject lines and body text, the relationship between sender and recipient, and the frequency and timing of interactions are also analyzed. The data collection unit centrally manages this data and builds a dataset for extracting question-and-answer patterns. Furthermore, the data collection unit can also use generative AI to learn and collect past question-and-answer patterns. Generative AI has the ability to process large amounts of text data and automatically extract question-and-answer patterns. For example, past email and communication tool histories are input into the generative AI to extract question-and-answer patterns. In this process, the generative AI analyzes the text data, identifies pairs of questions and answers, and learns them as patterns. This allows the data collection unit to efficiently collect question-and-answer patterns from past interactions and provide the data that forms the basis of the agent system.

[0075] The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and provide answers. Specifically, the analysis unit analyzes the collected question-and-answer patterns in detail and generates information for the agent to intervene at the appropriate time and provide accurate answers. The analysis unit uses natural language processing techniques and machine learning algorithms to analyze the collected patterns. For example, the analysis unit clusters the collected patterns and groups similar questions and answers. This provides the agent with information to select the most appropriate answer to a particular question. Furthermore, the analysis unit can also use generative AI to analyze the collected patterns and generate information. The generative AI takes the collected patterns as input and generates information for the agent to answer. For example, the generative AI analyzes the collected patterns, understands the intent of the question, and generates the optimal answer. In this process, the generative AI considers the context and background information of the question to improve the quality of the answer provided by the agent. This allows the analysis unit to efficiently analyze the collected data and provide information for the agent to answer accurately and quickly.

[0076] The generation unit allows agents to generate answers based on information generated by the analysis unit. Specifically, the generation unit is responsible for the process by which agents generate answers based on analyzed information. The generation unit uses natural language generation technology to express the analyzed information in natural language and generate answers to provide to the user. For example, the generation unit receives information provided by the analysis unit as input and generates an appropriate answer to a question. In this process, the generation AI plays a crucial role. The generation AI generates the optimal answer to a question based on the analyzed information. For example, the analyzed information is input to the generation AI, and an answer is generated. In this process, the generation AI considers the context and background information of the question and generates an answer expressed in natural language. The generation unit can also evaluate the quality of the generated answer and make corrections as needed. This allows the generation unit to provide users with high-quality answers. Furthermore, the generation unit integrates the generated answers with other parts of the agent system to provide a seamless user experience. For example, the generation unit sends the generated answers to the delivery unit to prepare them for provision to the user. This allows the generation unit to efficiently manage the answer generation process for the entire agent system and provide users with fast and accurate answers.

[0077] The delivery unit provides the answers generated by the generation unit. Specifically, the delivery unit is responsible for delivering the generated answers to users. The delivery unit provides the generated answers to users via email or communication tools. For example, the delivery unit inserts the generated answers into the body of an email and sends it to the user. It can also provide answers in a chat format using communication tools. The delivery unit can also automate the process of providing generated answers using generation AI. For example, the delivery unit inputs the generated answers into the generation AI and provides them to the user in an appropriate format. In this process, the generation AI optimizes the format and content of the answers and provides them in an easy-to-understand manner for the user. Furthermore, the delivery unit can collect feedback from users and evaluate the quality of the answers. For example, users evaluate the provided answers, and the evaluation results are collected. This allows the delivery unit to continuously improve the quality of the generated answers and increase user satisfaction. The delivery unit works in conjunction with other parts of the agent system to provide a seamless user experience. For example, the delivery unit quickly provides answers received from the generation unit to users and sends user feedback to the correction unit. This allows the delivery unit to improve the efficiency and effectiveness of the entire agent system.

[0078] The Correction Unit allows humans to correct answers when necessary in the responses provided by the Provider Unit, accumulating these corrections to refine future response patterns for the agent. Specifically, the Correction Unit receives user feedback on the provided answers, and humans correct the answers as needed. In this process, the Correction Unit evaluates the accuracy and appropriateness of the provided answers and makes necessary corrections. For example, if a provided answer is inaccurate or does not match the user's intent, the Correction Unit corrects the answer to provide accurate information. The Correction Unit can also use Generative AI to learn from corrected answers and refine future response patterns. For example, the Correction Unit inputs corrected answers into Generative AI to refine future response patterns. In this process, Generative AI learns from corrected answers, enabling the agent to provide more accurate answers to future questions. Furthermore, the Correction Unit accumulates corrected answers in a database, strengthening the knowledge base of the entire agent system. This allows the Correction Unit to continuously improve the quality of responses in the agent system and increase user satisfaction. The Correction Unit works in conjunction with other parts of the agent system to provide a seamless user experience. For example, the correction unit sends feedback received from the provisioning unit to the analysis unit, which then incorporates it into future response generation. This allows the correction unit to improve the overall efficiency and effectiveness of the agent system.

[0079] The system includes a notification unit. The notification unit notifies the staff of the generated response content. The notification unit notifies the staff of the generated response content, for example, via email or a communication tool. The notification unit can also notify the staff of the generated response content using a generation AI. For example, the notification unit inputs the generated response content into the generation AI and notifies it. This allows staff to be notified of the generated response content and correct it as needed. Some or all of the above processing in the notification unit may be performed using the generation AI or not. For example, the notification unit can input the generated response content into the generation AI and have the generation AI execute the notification.

[0080] The system includes a learning unit. The learning unit learns corrected responses, and the agent refines future response patterns. For example, the learning unit collects corrected responses and the agent learns to refine future response patterns. The learning unit can also use a generative AI to learn corrected responses and refine future response patterns. For example, the learning unit inputs corrected responses into the generative AI and refines future response patterns. This improves the agent's response accuracy by learning corrected responses. Some or all of the above processing in the learning unit may be performed using a generative AI or not. For example, the learning unit can input corrected responses into the generative AI and have the generative AI perform the learning.

[0081] The system includes a knowledge base unit. The knowledge base unit accumulates patterns of past question-and-answer interactions and provides a database for agents to refer to. For example, the knowledge base unit collects patterns of past question-and-answer interactions and stores them in the database. The knowledge base unit can also use a generative AI to accumulate patterns of past question-and-answer interactions and provide a database. For example, the knowledge base unit inputs patterns of past question-and-answer interactions into a generative AI and generates a database. This provides a database that agents can refer to by accumulating patterns of past question-and-answer interactions. Some or all of the above-described processes in the knowledge base unit may be performed using a generative AI or not. For example, the knowledge base unit can input patterns of past question-and-answer interactions into a generative AI and have the generative AI generate the database.

[0082] The data collection unit can estimate the user's emotions and select question-and-answer patterns to collect based on the estimated emotions. For example, if the user is stressed, the data collection unit will prioritize collecting patterns from past question-and-answer sessions that can be resolved quickly. If the user is relaxed, the data collection unit may also collect question-and-answer patterns that include detailed explanations. If the user is in a hurry, the data collection unit may also collect concise and to-the-point question-and-answer patterns. This allows for the collection of more appropriate patterns by selecting question-and-answer patterns based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the data collection unit may be performed using or without a generative AI. For example, the data collection unit can input user emotion data into a generative AI and have the generative AI select question-and-answer patterns.

[0083] The data collection unit can analyze the user's past communication history and select the optimal collection method when collecting questions and answers. For example, the data collection unit can collect relevant question and answer patterns based on keywords frequently used by the user in the past. The data collection unit can also collect question and answer patterns that occur during specific time periods from the user's past communication history. The data collection unit can also analyze the user's past communication history and collect question and answer patterns related to specific projects. This allows the optimal collection method to be selected by analyzing the user's past communication history. Some or all of the above processing in the data collection unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the data collection unit can input the user's past communication history into a generative AI and have the generative AI select the optimal collection method.

[0084] The collection unit can filter the collected questions and answers based on the user's current projects and areas of interest. For example, the collection unit can prioritize collecting question and answer patterns related to the user's current projects. The collection unit can also collect relevant question and answer patterns based on the user's areas of interest. The collection unit can also collect question and answer patterns related to topics the user is currently interested in. This allows for the collection of highly relevant questions and answers by filtering based on the user's current projects and areas of interest. Some or all of the above processing in the collection unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the collection unit can input data on the user's current projects and areas of interest into a generative AI and have the generative AI perform the filtering.

[0085] The data collection unit can estimate the user's emotions and determine the priority of questions and answers to collect based on the estimated emotions. For example, if the user is stressed, the data collection unit will prioritize collecting question and answer patterns that can be resolved quickly. If the user is relaxed, the data collection unit may also prioritize collecting question and answer patterns that include detailed explanations. If the user is in a hurry, the data collection unit may also prioritize collecting concise and to-the-point question and answer patterns. This allows for the collection of more appropriate patterns by prioritizing questions and answers based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the processing described above in the data collection unit may be performed using or without a generative AI. For example, the data collection unit can input user emotion data into a generative AI and have the generative AI determine the priority of questions and answers.

[0086] The collection unit can prioritize the collection of highly relevant patterns when collecting questions and answers, taking into account the user's geographical location information. For example, the collection unit can prioritize the collection of question and answer patterns related to the user's current location. The collection unit can also collect region-specific question and answer patterns based on the user's geographical location information. If the user is on the move, the collection unit can also prioritize the collection of question and answer patterns related to their current location. This allows for the priority collection of highly relevant questions and answers by considering the user's geographical location information. Some or all of the above processing in the collection unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the collection unit can input the user's geographical location information into a generation AI and have the generation AI perform the collection of questions and answers.

[0087] The collection unit can analyze the user's social media activity and collect relevant patterns when collecting questions and answers. For example, the collection unit can collect question and answer patterns related to topics that the user frequently mentions on social media. The collection unit can also collect question and answer patterns related to topics of high interest from the user's social media activity. The collection unit can also collect question and answer patterns related to accounts that the user follows on social media. In this way, relevant question and answer patterns can be collected by analyzing the user's social media activity. Some or all of the above processing in the collection unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the collection unit can input the user's social media activity data into a generative AI and have the generative AI perform the collection of questions and answers.

[0088] The analysis unit can estimate the user's emotions and adjust the presentation of the analysis based on the estimated emotions. For example, if the user is stressed, the analysis unit can provide the analysis results using a concise and clear presentation. If the user is relaxed, the analysis unit can also provide the analysis results using a presentation that includes detailed explanations. If the user is in a hurry, the analysis unit can also provide the analysis results using a presentation that gets straight to the point. This allows for the provision of more appropriate analysis results by adjusting the presentation of the analysis based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the analysis unit may be performed using or without a generative AI. For example, the analysis unit can input user emotion data into a generative AI and have the generative AI adjust the presentation of the analysis.

[0089] The analysis unit can adjust the level of detail of the analysis based on the importance of the questions and answers during the analysis. For example, the analysis unit can perform a detailed analysis for questions and answers of high importance. The analysis unit can also perform a concise analysis for questions and answers of low importance. The analysis unit can also adjust the level of detail of the analysis in stages according to the importance of the questions and answers. By adjusting the level of detail of the analysis based on the importance of the questions and answers, more appropriate analysis results can be provided. Some or all of the above processing in the analysis unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the analysis unit can input the importance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the level of detail of the analysis.

[0090] The analysis unit can apply different analysis algorithms depending on the category of the question and answer during analysis. For example, the analysis unit can apply a specialized analysis algorithm to technical questions and answers. The analysis unit can also apply a simpler analysis algorithm to general questions and answers. The analysis unit can also apply an analysis algorithm that prioritizes customer satisfaction to questions and answers related to customer support. By applying different analysis algorithms depending on the category of the question and answer, more appropriate analysis results can be provided. Some or all of the above processing in the analysis unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the analysis unit can input the category data of the question and answer into a generative AI and have the generative AI execute the application of the analysis algorithm.

[0091] The analysis unit can estimate the user's emotions and adjust the length of the analysis based on the estimated emotions. For example, if the user is stressed, the analysis unit can provide a short, concise analysis result. If the user is relaxed, the analysis unit can also provide a detailed analysis result. If the user is in a hurry, the analysis unit can also provide a brief analysis result. By adjusting the length of the analysis based on the user's emotions, more appropriate analysis results can be provided. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the analysis unit may be performed using or without a generative AI. For example, the analysis unit can input user emotion data into a generative AI and have the generative AI adjust the length of the analysis.

[0092] The analysis unit can determine the priority of analysis based on the submission date of the questions and answers during the analysis. For example, the analysis unit may prioritize the analysis of recently submitted questions and answers. The analysis unit may also postpone the analysis of older questions and answers. The analysis unit can also adjust the priority of analysis in stages according to the submission date. This allows for the provision of more appropriate analysis results by determining the priority of analysis based on the submission date of the questions and answers. Some or all of the above processing in the analysis unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the analysis unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the determination of the analysis priority.

[0093] The analysis unit can adjust the order of analysis based on the relevance of the questions and answers during the analysis. For example, the analysis unit may prioritize the analysis of highly relevant questions and answers. The analysis unit may also postpone the analysis of less relevant questions and answers. The analysis unit can also adjust the order of analysis in stages according to the relevance of the questions and answers. By adjusting the order of analysis based on the relevance of the questions and answers, more appropriate analysis results can be provided. Some or all of the above processing in the analysis unit may be performed using a generating AI, or it may be performed without a generating AI. For example, the analysis unit can input the relevance data of the questions and answers into a generating AI and have the generating AI perform the adjustment of the order of analysis.

[0094] The generation unit can estimate the user's emotions and adjust the way it expresses the generated response based on the estimated emotions. For example, if the user is stressed, the generation unit can generate a response using a concise and clear expression. If the user is relaxed, the generation unit can also generate a response using an expression that includes detailed explanations. If the user is in a hurry, the generation unit can also generate a response using an expression that gets straight to the point. This allows for the generation of more appropriate responses by adjusting the expression of the response based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processes in the generation unit may be performed using or without a generative AI. For example, the generation unit can input user emotion data into a generative AI and have the generative AI adjust the expression of the response.

[0095] The generation unit can adjust the level of detail in the answers based on the importance of the questions and answers when generating responses. For example, the generation unit can generate detailed answers for high-importance questions and answers. The generation unit can also generate concise answers for low-importance questions and answers. The generation unit can also adjust the level of detail in the answers in stages according to the importance of the questions and answers. This allows for the generation of more appropriate answers by adjusting the level of detail in the answers based on the importance of the questions and answers. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input question and answer importance data into a generation AI and have the generation AI perform the adjustment of the level of detail in the answers.

[0096] The generation unit can apply different generation algorithms depending on the category of the question and answer when generating answers. For example, the generation unit can apply a specialized generation algorithm to technical questions and answers. The generation unit can also apply a simple generation algorithm to general questions and answers. The generation unit can also apply a generation algorithm that prioritizes customer satisfaction to questions and answers related to customer support. By applying different generation algorithms depending on the category of the question and answer, more appropriate answers can be generated. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input the category data of the question and answer into a generation AI and have the generation AI execute the application of the generation algorithm.

[0097] The generation unit can estimate the user's emotions and adjust the length of the response it generates based on the estimated emotions. For example, if the user is stressed, the generation unit can generate a short, concise response. If the user is relaxed, the generation unit can also generate a longer response with more detailed explanations. If the user is in a hurry, the generation unit can generate a brief response. By adjusting the length of the response based on the user's emotions, a more appropriate response can be generated. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above processing in the generation unit may be performed using or without a generative AI. For example, the generation unit can input user emotion data into a generative AI and have the generative AI adjust the length of the response.

[0098] The generation unit can determine the priority of answers based on the submission date of the questions and answers when generating responses. For example, the generation unit can prioritize generating answers for recently submitted questions and answers. The generation unit can also postpone generating answers for older questions and answers. The generation unit can also adjust the priority of answers in stages according to the submission date. This allows for the generation of more appropriate answers by determining the priority of answers based on the submission date of the questions and answers. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the determination of the priority of answers.

[0099] The generation unit can adjust the order of answers based on the relevance of the questions and answers when generating responses. For example, the generation unit can prioritize generating responses to highly relevant questions and answers. The generation unit can also postpone generating responses to less relevant questions and answers. The generation unit can also adjust the order of answers in stages according to the relevance of the questions and answers. This allows for the generation of more appropriate answers by adjusting the order of answers based on the relevance of the questions and answers. Some or all of the above processing in the generation unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the generation unit can input the relevance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of answers.

[0100] The service provider can estimate the user's emotions and adjust the way it provides answers based on those emotions. For example, if the user is stressed, the service provider may provide answers in a concise and clear manner. If the user is relaxed, the service provider may also provide answers that include detailed explanations. If the user is in a hurry, the service provider may also provide answers that get straight to the point. This allows for more appropriate answers to be provided by adjusting the way answers are provided based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the service provider may be performed using or without a generative AI. For example, the service provider can input user emotion data into a generative AI and have the generative AI adjust the way answers are provided.

[0101] The answering unit can adjust the level of detail provided based on the importance of the question and answer. For example, the answering unit can provide detailed answers to high-importance questions and answers. It can also provide concise answers to low-importance questions and answers. The answering unit can also adjust the level of detail in stages according to the importance of the questions and answers. This allows for the provision of more appropriate answers by adjusting the level of detail based on the importance of the questions and answers. Some or all of the above processing in the answering unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the answering unit can input question and answer importance data into a generation AI and have the generation AI perform the adjustment of the level of detail in the answer.

[0102] The response unit can apply different response algorithms depending on the category of the question and answer when providing answers. For example, the response unit can apply a specialized response algorithm to technical questions. The response unit can also apply a simpler response algorithm to general questions. The response unit can also apply a response algorithm that prioritizes customer satisfaction to questions related to customer support. By applying different response algorithms depending on the category of the question and answer, more appropriate answers can be provided. Some or all of the above processing in the response unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the response unit can input question and answer category data into a generative AI and have the generative AI execute the application of the response algorithm.

[0103] The service provider can estimate the user's emotions and determine the priority of the responses to provide based on the estimated emotions. For example, if the user is stressed, the service provider will provide a quick response. If the user is relaxed, the service provider may also provide a response that includes a detailed explanation. If the user is in a hurry, the service provider may also provide a concise and to-the-point response. This allows for the provision of more appropriate responses by prioritizing responses based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the service provider may be performed using or without a generative AI. For example, the service provider can input user emotion data into a generative AI and have the generative AI determine the priority of responses.

[0104] The service provider can adjust the order in which answers are provided based on the submission date of the questions and answers. For example, the service provider can prioritize answers to recently submitted questions and answers. The service provider can also postpone the provision of answers to older questions and answers. The service provider can also adjust the order of provision in stages according to the submission date. This allows for the provision of more appropriate answers by adjusting the order of provision based on the submission date of the questions and answers. Some or all of the above processing in the service provider may be performed using a generation AI, or it may be performed without a generation AI. For example, the service provider can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of provision.

[0105] The answering unit can adjust the order in which answers are provided based on the relevance of the questions and answers. For example, the answering unit can prioritize answers to highly relevant questions and answers. The answering unit can also postpone answers to less relevant questions and answers. The answering unit can also adjust the order of answers in stages according to the relevance of the questions and answers. This allows for the provision of more appropriate answers by adjusting the order of answers based on the relevance of the questions and answers. Some or all of the above processing in the answering unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the answering unit can input the relevance data of the questions and answers into a generative AI and have the generative AI perform the adjustment of the order of answers.

[0106] The correction unit can estimate the user's emotions and adjust the correction method based on the estimated emotions. For example, if the user is stressed, the correction unit will make corrections in a concise and clear manner. If the user is relaxed, the correction unit may also make corrections that include detailed explanations. If the user is in a hurry, the correction unit may also make corrections that get straight to the point. This allows for more appropriate corrections by adjusting the correction method based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the correction unit may be performed using or without a generative AI. For example, the correction unit can input user emotion data into a generative AI and have the generative AI adjust the correction method.

[0107] The correction unit can adjust the level of detail of the correction based on the importance of the question and answer during the correction process. For example, the correction unit can perform detailed corrections for high-importance questions and answers. The correction unit can also perform concise corrections for low-importance questions and answers. The correction unit can also adjust the level of detail of the correction in stages according to the importance of the questions and answers. This allows for more appropriate corrections by adjusting the level of detail of the correction based on the importance of the questions and answers. Some or all of the above-described processes in the correction unit may be performed using a generation AI, or they may be performed without a generation AI. For example, the correction unit can input question and answer importance data into a generation AI and have the generation AI perform the adjustment of the level of detail of the correction.

[0108] The correction unit can apply different correction algorithms depending on the category of the question and answer during the correction process. For example, the correction unit can apply a specialized correction algorithm to technical questions and answers. The correction unit can also apply a simpler correction algorithm to general questions and answers. The correction unit can also apply a correction algorithm that prioritizes customer satisfaction to questions and answers related to customer support. By applying different correction algorithms depending on the category of the question and answer, more appropriate corrections can be made. Some or all of the above processing in the correction unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the correction unit can input the category data of the question and answer into a generation AI and have the generation AI execute the application of the correction algorithm.

[0109] The correction unit can estimate the user's emotions and determine the priority of corrections based on the estimated emotions. For example, if the user is stressed, the correction unit can perform corrections quickly. If the user is relaxed, the correction unit can also perform corrections that include detailed explanations. If the user is in a hurry, the correction unit can also perform concise and to-the-point corrections. This allows for more appropriate corrections by determining the priority of corrections based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the correction unit may be performed using or without a generative AI. For example, the correction unit can input user emotion data into a generative AI and have the generative AI determine the priority of corrections.

[0110] The correction unit can adjust the order of corrections based on the submission date of the questions and answers. For example, the correction unit can prioritize corrections to recently submitted questions and answers. The correction unit can also postpone corrections to older questions and answers. The correction unit can also adjust the order of corrections in stages according to the submission date. This allows for more appropriate corrections by adjusting the order of corrections based on the submission date of the questions and answers. Some or all of the above processing in the correction unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the correction unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of corrections.

[0111] The correction unit can adjust the order of corrections based on the relevance of the questions and answers during the correction process. For example, the correction unit can prioritize correcting highly relevant questions and answers. The correction unit can also postpone correcting less relevant questions and answers. The correction unit can also adjust the order of corrections in stages according to the relevance of the questions and answers. This allows for more appropriate corrections by adjusting the order of corrections based on the relevance of the questions and answers. Some or all of the above-described processes in the correction unit may be performed using a generation AI, or they may be performed without a generation AI. For example, the correction unit can input the relevance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of corrections.

[0112] The notification unit can estimate the user's emotions and adjust the notification method based on the estimated emotions. For example, if the user is stressed, the notification unit will provide a concise and clear notification. If the user is relaxed, the notification unit may also provide a notification with a detailed explanation. If the user is in a hurry, the notification unit may also provide a notification that gets straight to the point. This allows for more appropriate notifications by adjusting the notification method based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the notification unit may be performed using or without a generative AI. For example, the notification unit can input user emotion data into a generative AI and have the generative AI adjust the notification method.

[0113] The notification unit can adjust the level of detail of the notification based on the importance of the question and answer. For example, the notification unit can provide detailed notifications for high-importance questions and answers. It can also provide concise notifications for low-importance questions and answers. The notification unit can also adjust the level of detail of the notification in stages according to the importance of the questions and answers. This allows for more appropriate notifications to be provided by adjusting the level of detail of the notification based on the importance of the questions and answers. Some or all of the above processing in the notification unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the notification unit can input the importance data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the level of detail of the notification.

[0114] The notification unit can estimate the user's emotions and determine notification priorities based on those emotions. For example, if the user is stressed, the notification unit can send a quick notification. If the user is relaxed, the notification unit can also send a notification with a detailed explanation. If the user is in a hurry, the notification unit can send a concise and to-the-point notification. This allows for more appropriate notifications by prioritizing them based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the notification unit may be performed using or without a generative AI. For example, the notification unit can input user emotion data into a generative AI and have the generative AI determine the notification priorities.

[0115] The notification unit can adjust the order of notifications based on when the questions and answers were submitted. For example, the notification unit can prioritize notifications for recently submitted questions and answers. It can also postpone notifications for older questions and answers. The notification unit can also adjust the order of notifications in stages according to the submission date. This allows for more appropriate notifications by adjusting the order of notifications based on the submission date of the questions and answers. Some or all of the above processing in the notification unit may be performed using a generation AI, or not. For example, the notification unit can input the submission date data of the questions and answers into a generation AI and have the generation AI perform the adjustment of the order of notifications.

[0116] The learning unit can estimate the user's emotions and select training data based on the estimated emotions. For example, if the user is stressed, the learning unit can select question-and-answer patterns that can be resolved quickly as training data. If the user is relaxed, the learning unit can also select question-and-answer patterns that include detailed explanations as training data. If the user is in a hurry, the learning unit can also select concise and to-the-point question-and-answer patterns as training data. This allows for the selection of more appropriate training data based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the learning unit may be performed using or without a generative AI. For example, the learning unit can input user emotion data into a generative AI and have the generative AI perform the selection of training data.

[0117] The learning unit can optimize the learning algorithm by referring to past learning data during the learning process. For example, the learning unit can select the optimal learning algorithm based on past learning data. The learning unit can also analyze past learning data and adjust the parameters of the learning algorithm. The learning unit can also improve the accuracy of the learning algorithm by referring to past learning data. As a result, the accuracy of learning is improved by optimizing the learning algorithm by referring to past learning data. Some or all of the above processes in the learning unit may be performed using a generative AI, or they may be performed without a generative AI. For example, the learning unit can input past learning data into a generative AI and have the generative AI perform the optimization of the learning algorithm.

[0118] The learning unit can estimate the user's emotions and adjust the frequency of learning based on the estimated emotions. For example, if the user is stressed, the learning unit can learn frequently to enable quick responses. If the user is relaxed, the learning unit can learn regularly to enable detailed responses. If the user is in a hurry, the learning unit can learn quickly to enable concise responses. By adjusting the frequency of learning based on the user's emotions, learning can be performed at a more appropriate frequency. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the learning unit may be performed using or without a generative AI. For example, the learning unit can input user emotion data into a generative AI and have the generative AI adjust the frequency of learning.

[0119] The learning unit can weight the training data based on the submission date of the questions and answers during training. For example, the learning unit can give higher weight to recently submitted questions and answers during training. The learning unit can also give lower weight to older questions and answers during training. The learning unit can also adjust the weighting of the training data in stages according to the submission date. This allows for more appropriate training by weighting the training data based on the submission date of the questions and answers. Some or all of the above processing in the learning unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the learning unit can input the submission date data of the questions and answers into a generative AI and have the generative AI perform the weighting of the training data.

[0120] The knowledge base unit can estimate the user's emotions and adjust the frequency of knowledge base updates based on the estimated emotions. For example, if the user is stressed, the knowledge base unit can update the knowledge base frequently to enable quick responses. If the user is relaxed, the knowledge base unit can also update the knowledge base regularly to enable detailed responses. If the user is in a hurry, the knowledge base unit can update the knowledge base quickly to enable concise responses. By adjusting the frequency of knowledge base updates based on the user's emotions, the knowledge base can be updated at a more appropriate frequency. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the knowledge base unit may be performed using generative AI or not. For example, the knowledge base unit can input user emotion data into a generating AI and have the AI ​​adjust the update frequency of the knowledge base.

[0121] The knowledge base unit can select the optimal update method by referring to past Q&A data when updating the knowledge base. For example, the knowledge base unit selects the optimal update method based on past Q&A data. The knowledge base unit can also analyze past Q&A data and optimize the knowledge base update method. The knowledge base unit can also adjust the update frequency of the knowledge base by referring to past Q&A data. This improves the accuracy of the knowledge base by selecting the optimal update method by referring to past Q&A data. Some or all of the above processing in the knowledge base unit may be performed using a generation AI, or it may be performed without a generation AI. For example, the knowledge base unit can input past Q&A data into a generation AI and have the generation AI select the update method.

[0122] The knowledge base unit can estimate the user's emotions and prioritize the knowledge base based on those emotions. For example, if the user is stressed, the knowledge base unit will prioritize updating knowledge base items that can be addressed quickly. If the user is relaxed, the knowledge base unit may also prioritize updating knowledge base items that contain detailed explanations. If the user is in a hurry, the knowledge base unit may also prioritize updating knowledge base items that are concise and to the point. This allows for prioritizing the knowledge base based on the user's emotions, thereby prioritizing the updating of more appropriate knowledge base items. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the knowledge base unit may be performed using generative AI or not. For example, the knowledge base unit can input user emotion data into a generating AI and have the generating AI determine the priorities of the knowledge base.

[0123] The knowledge base unit can adjust the update order based on the submission date of questions and answers when updating the knowledge base. For example, the knowledge base unit can prioritize updating the knowledge base for recently submitted questions and answers. The knowledge base unit can also postpone updating the knowledge base for older questions and answers. The knowledge base unit can also adjust the update order of the knowledge base in stages according to the submission date. This allows the knowledge base to be updated in a more appropriate order by adjusting the update order based on the submission date of the questions and answers. Some or all of the above processing in the knowledge base unit may be performed using a generation AI, or not. For example, the knowledge base unit can input the submission date data of questions and answers into a generation AI and have the generation AI perform the adjustment of the update order.

[0124] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0125] The agent system can further estimate the user's emotions and adjust the tone of its response based on those emotions. For example, if the user is stressed, the agent can generate a response in a more friendly and encouraging tone. If the user is relaxed, the agent can generate a response in a more detailed and polite tone. If the user is in a hurry, the agent can generate a response in a concise and to-the-point tone. This allows the system to provide responses in an appropriate tone that matches the user's emotions.

[0126] The agent system can further analyze a user's past behavior history and provide responses optimized for each individual user. For example, if a user has previously shown interest in a particular topic, it can prioritize providing detailed information related to that topic. If a user has previously preferred a specific response format, it can also generate responses in that format. This allows for more personalized responses based on the user's past behavior history.

[0127] The agent system can further estimate the user's emotions and adjust the timing of response delivery based on those estimates. For example, if the user is stressed, it can provide a quick response. If the user is relaxed, it can provide a response with detailed explanations. If the user is in a hurry, it can provide a concise and to-the-point response. This allows for responses to be delivered at the appropriate time according to the user's emotions.

[0128] The agent system can further customize its responses based on the user's current situation. For example, if the user is in a meeting, the agent can provide a concise and to-the-point answer. If the user is on the go, the agent can also provide an answer in a format optimized for mobile devices. This ensures that responses are provided in an appropriate format according to the user's current situation.

[0129] The agent system can further estimate the user's emotions and adjust its response based on those emotions. For example, if the user is stressed, the agent can provide a response that includes words of encouragement and comfort. If the user is relaxed, the agent can provide a response that includes detailed explanations and additional information. If the user is in a hurry, the agent can provide a concise and to-the-point response. This allows the system to provide responses that are appropriate to the user's emotions.

[0130] The agent system can further analyze past user feedback to improve the accuracy of its responses. For example, based on feedback previously provided by the user, the agent can adjust the content and format of its responses. If a user gives a high rating to a particular response, the agent can use that format and content as a reference to generate new responses. This allows the system to provide more accurate responses based on the user's past feedback.

[0131] The agent system can further estimate the user's emotions and prioritize responses based on those emotions. For example, if the user is stressed, it can prioritize providing answers that can be resolved quickly. If the user is relaxed, it can prioritize providing answers that include detailed explanations. If the user is in a hurry, it can prioritize providing concise and to-the-point answers. This allows the system to provide answers with appropriate priorities according to the user's emotions.

[0132] The agent system can further customize its responses by taking into account the user's geographical location. For example, if the user is in a specific region, it can prioritize providing information relevant to that region. If the user is traveling, it can also provide information relevant to their travel destination. This allows the system to provide more relevant responses based on the user's geographical location.

[0133] The agent system can further estimate the user's emotions and adjust the response format based on those estimates. For example, if the user is stressed, the agent can provide a response in a concise and visually easy-to-understand format. If the user is relaxed, it can provide a detailed text response. If the user is in a hurry, it can provide a response in a bulleted list format. This allows the system to provide responses in an appropriate format according to the user's emotions.

[0134] The agent system can further analyze the user's social media activity and provide relevant information. For example, it can prioritize providing information related to topics the user frequently mentions on social media. It can also provide information related to accounts the user follows. This allows for the provision of more relevant information based on the user's social media activity.

[0135] The following briefly describes the processing flow for example form 2.

[0136] Step 1: The collection unit collects question-and-answer patterns from past emails and communication tool history. For example, the collection unit can use generative AI to learn and collect past question-and-answer patterns. Step 2: The analysis unit analyzes the patterns collected by the collection unit and generates information for the agent to intervene and respond. For example, the analysis unit can use a generation AI to analyze the collected patterns and generate information. Step 3: The generation unit uses the information generated by the analysis unit to generate an answer from the agent. For example, the generation unit can use a generation AI to generate an answer based on the analyzed information. Step 4: The providing unit provides the answers generated by the generating unit. For example, the providing unit can use a generation AI to provide the generated answers via email or communication tools. Step 5: The correction unit has a human correct the answer provided by the supply unit if correction is needed, and this is accumulated so that the agent can refine future answer patterns. For example, the correction unit can use a generation AI to learn from the corrected answers and refine future answer patterns.

[0137] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0138] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0139] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0140] Each of the multiple elements described above, including the collection unit, analysis unit, generation unit, provision unit, correction unit, notification unit, learning unit, and knowledge base unit, is implemented in at least one of the smart device 14 and the data processing device 12. For example, the collection unit is implemented by the control unit 46A of the smart device 14 and collects question-and-answer patterns from past email and communication tool history. The analysis unit is implemented by the identification processing unit 290 of the data processing device 12 and analyzes the collected patterns to generate information for the agent to respond. The generation unit is implemented by the identification processing unit 290 of the data processing device 12 and the agent generates a response based on the analyzed information. The provision unit is implemented by the control unit 46A of the smart device 14 and provides the generated response via email or communication tool. The correction unit is implemented by the identification processing unit 290 of the data processing device 12 and a human makes a corrected response if the provided response needs correction and stores it. The notification unit is implemented, for example, by the control unit 46A of the smart device 14, and notifies the staff member of the generated response. The learning unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and learns the corrected response so that the agent can refine future response patterns. The knowledge base unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and stores past question-and-answer patterns, providing a database for the agent to refer to. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0141] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0142] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0143] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0144] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0145] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0146] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0147] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0148] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0149] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0150] Storage 32 stores the data generation model 58 and the 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 user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0151] In the smart glasses 214, 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 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0152] Furthermore, other devices besides the data processing device 12 may also 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 processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0153] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0154] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0155] 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 performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed 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 necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0156] Each of the multiple elements described above, including the collection unit, analysis unit, generation unit, provision unit, correction unit, notification unit, learning unit, and knowledge base unit, is implemented, for example, in at least one of the smart glasses 214 and the data processing device 12. For example, the collection unit is implemented by the control unit 46A of the smart glasses 214 and collects question-and-answer patterns from past email and communication tool history. The analysis unit is implemented, for example, by the identification processing unit 290 of the data processing device 12 and analyzes the collected patterns to generate information for the agent to respond. The generation unit is implemented, for example, by the identification processing unit 290 of the data processing device 12 and the agent generates a response based on the analyzed information. The provision unit is implemented, for example, by the control unit 46A of the smart glasses 214 and provides the generated response via email or communication tool. The correction unit is implemented, for example, by the identification processing unit 290 of the data processing device 12 and stores a corrected response made by a human when correction is needed for the provided response. The notification unit is implemented, for example, by the control unit 46A of the smart glasses 214, and notifies the staff member of the generated response. The learning unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and learns corrected responses so that the agent can refine future response patterns. The knowledge base unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and stores past question-and-answer patterns, providing a database for the agent to refer to. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0157] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0158] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0159] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0160] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0161] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0162] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0163] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0164] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0165] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0166] Storage 32 stores the data generation model 58 and the 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 user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0167] In the headset terminal 314, 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 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0168] Furthermore, other devices besides the data processing device 12 may also 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 processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0169] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0170] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0171] 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 performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0172] Each of the multiple elements described above, including the collection unit, analysis unit, generation unit, provision unit, correction unit, notification unit, learning unit, and knowledge base unit, is implemented in at least one of the headset terminal 314 and the data processing unit 12. For example, the collection unit is implemented by the control unit 46A of the headset terminal 314 and collects question-and-answer patterns from past email and communication tool history. The analysis unit is implemented by the specific processing unit 290 of the data processing unit 12 and analyzes the collected patterns to generate information for the agent to respond. The generation unit is implemented by the specific processing unit 290 of the data processing unit 12 and the agent generates a response based on the analyzed information. The provision unit is implemented by the control unit 46A of the headset terminal 314 and provides the generated response via email or communication tool. The correction unit is implemented by the specific processing unit 290 of the data processing unit 12 and stores a corrected response made by a human when correction is needed for the provided response. The notification unit is implemented, for example, by the control unit 46A of the headset terminal 314, and notifies the staff member of the generated response. The learning unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and learns the corrected response so that the agent can refine future response patterns. The knowledge base unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and stores past question-and-answer patterns, providing a database for the agent to refer to. The correspondence between each unit and the device or control unit is not limited to the example described above, and various changes are possible.

[0173] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0174] As shown in Figure 7, the 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.

[0175] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0176] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0177] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0178] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0179] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0180] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0181] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0182] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0183] Storage 32 stores the data generation model 58 and the 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 user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0184] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0185] Furthermore, other devices besides the data processing device 12 may also 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 processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0186] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0187] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0188] 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 performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed 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 necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0189] Each of the multiple elements described above, including the collection unit, analysis unit, generation unit, provision unit, correction unit, notification unit, learning unit, and knowledge base unit, is implemented, for example, in at least one of the robot 414 and the data processing unit 12. For example, the collection unit is implemented by the control unit 46A of the robot 414 and collects question-and-answer patterns from the history of past emails and communication tools. The analysis unit is implemented, for example, by the identification processing unit 290 of the data processing unit 12 and analyzes the collected patterns to generate information for the agent to answer. The generation unit is implemented, for example, by the identification processing unit 290 of the data processing unit 12 and the agent generates an answer based on the analyzed information. The provision unit is implemented, for example, by the control unit 46A of the robot 414 and provides the generated answer via email or communication tools. The correction unit is implemented, for example, by the identification processing unit 290 of the data processing unit 12 and stores a corrected answer made by a human when correction is needed for the provided answer. The notification unit is implemented, for example, by the control unit 46A of the robot 414, and notifies the staff member of the generated response. The learning unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and learns the corrected responses so that the agent can refine future response patterns. The knowledge base unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, and stores past question-and-answer patterns, providing a database for the agent to refer to. The correspondence between each unit and the devices and control units is not limited to the examples described above, and various changes are possible.

[0190] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0191] Figure 9 shows the emotion map 400, in 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 the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0192] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0193] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0194] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0195] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it 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 emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0196] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0197] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0198] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0199] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0200] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0201] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0202] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0203] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0204] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0205] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0206] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0207] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0208] (Note 1) A collection unit that collects question-and-answer patterns from past emails and communication tool history, An analysis unit analyzes the patterns collected by the aforementioned collection unit and generates information for the agent to intervene and respond, A generation unit in which an agent generates a response based on the information generated by the analysis unit, A providing unit that provides the answer generated by the generation unit, The system includes a correction unit in which a human provides a corrected response if the response provided by the aforementioned provision unit needs correction, and this correction unit is stored so that the agent can refine future response patterns. A system characterized by the following features. (Note 2) It includes a notification unit that notifies the staff member of the generated response content. The system described in Appendix 1, characterized by the features described herein. (Note 3) It features a learning unit that learns from corrected responses, allowing the agent to refine future response patterns. The system described in Appendix 1, characterized by the features described herein. (Note 4) It includes a knowledge base section that accumulates patterns of past questions and answers and provides a database for agents to refer to. The system described in Appendix 1, characterized by the features described herein. (Note 5) The aforementioned collection unit is The system estimates the user's emotions and selects question-and-answer patterns to collect based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 6) The aforementioned collection unit is When collecting Q&A data, the system analyzes the user's past communication history to select the most suitable collection method. The system described in Appendix 1, characterized by the features described herein. (Note 7) The aforementioned collection unit is When collecting questions and answers, filter them based on the user's current projects and areas of interest. The system described in Appendix 1, characterized by the features described herein. (Note 8) The aforementioned collection unit is It estimates the user's emotions and determines the priority of the questions and answers to be collected based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 9) The aforementioned collection unit is When collecting Q&A data, the system prioritizes collecting highly relevant patterns by considering the user's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 10) The aforementioned collection unit is When collecting Q&A data, analyze users' social media activity and collect relevant patterns. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned analysis unit, It estimates the user's emotions and adjusts the representation of the analysis based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned analysis unit, During the analysis, the level of detail of the analysis will be adjusted based on the importance of the questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned analysis unit, During analysis, different analysis algorithms are applied depending on the category of the question and answer. The system described in Appendix 1, characterized by the features described herein. (Note 14) The aforementioned analysis unit, It estimates the user's emotions and adjusts the length of the analysis based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 15) The aforementioned analysis unit, During the analysis, the priority of the analysis will be determined based on the timing of the submission of questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 16) The aforementioned analysis unit, During the analysis, the order of analysis will be adjusted based on the relevance of the questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 17) The generating unit is It estimates the user's emotions and adjusts how the generated responses are expressed based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 18) The generating unit is When generating answers, adjust the level of detail in the answers based on the importance of the question and answer. The system described in Appendix 1, characterized by the features described herein. (Note 19) The generating unit is When generating answers, different generation algorithms are applied depending on the category of the question and answer. The system described in Appendix 1, characterized by the features described herein. (Note 20) The generating unit is It estimates the user's emotions and adjusts the length of the response generated based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 21) The generating unit is When generating answers, the priority of the answers is determined based on when the questions and answers were submitted. The system described in Appendix 1, characterized by the features described herein. (Note 22) The generating unit is When generating answers, the order of the answers will be adjusted based on the relevance of the questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 23) The aforementioned supply unit is, We estimate the user's emotions and adjust how we provide responses based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 24) The aforementioned supply unit is, When providing answers, we adjust the level of detail based on the importance of the question and answer. The system described in Appendix 1, characterized by the features described herein. (Note 25) The aforementioned supply unit is, When providing answers, different answering algorithms are applied depending on the category of the question and answer. The system described in Appendix 1, characterized by the features described herein. (Note 26) The aforementioned supply unit is, It estimates the user's emotions and determines the priority of the responses to provide based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 27) The aforementioned supply unit is, When providing answers, we will adjust the order of provision based on when the questions and answers were submitted. The system described in Appendix 1, characterized by the features described herein. (Note 28) The aforementioned supply unit is, When providing answers, we adjust the order of presentation based on the relevance of the questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 29) The correction unit, The system estimates the user's emotions and adjusts the correction method based on the estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 30) The correction unit, During the correction process, the level of detail of the correction will be adjusted based on the importance of the questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 31) The correction unit, During correction, different correction algorithms are applied depending on the category of the question and answer session. The system described in Appendix 1, characterized by the features described herein. (Note 32) The correction unit, The system estimates the user's emotions and determines the priority of corrections based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 33) The correction unit, When making corrections, the order of corrections will be adjusted based on the timing of submission of questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 34) The correction unit, During the correction process, the order of corrections will be adjusted based on the relevance of the questions and answers. The system described in Appendix 1, characterized by the features described herein. (Note 35) The aforementioned notification unit, It estimates the user's emotions and adjusts the notification method based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 36) The aforementioned notification unit, When sending notifications, adjust the level of detail in the notification based on the importance of the Q&A session. The system described in Appendix 1, characterized by the features described herein. (Note 37) The aforementioned notification unit, It estimates the user's emotions and prioritizes notifications based on those emotions. The system described in Appendix 1, characterized by the features described herein. (Note 38) The aforementioned notification unit, When notifying, the order of notifications will be adjusted based on when questions and answers are submitted. The system described in Appendix 1, characterized by the features described herein. (Note 39) The aforementioned learning unit, The system estimates the user's emotions and selects training data based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 40) The aforementioned learning unit, During training, the learning algorithm is optimized by referring to past training data. The system described in Appendix 1, characterized by the features described herein. (Note 41) The aforementioned learning unit, It estimates the user's emotions and adjusts the learning frequency based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 42) The aforementioned learning unit, During training, the training data is weighted based on when questions and answers were submitted. The system described in Appendix 1, characterized by the features described herein. (Note 43) The aforementioned knowledge base unit is, We estimate user sentiment and adjust the frequency of knowledge base updates based on that estimated sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 44) The aforementioned knowledge base unit is, When updating the knowledge base, the optimal update method is selected by referring to past Q&A data. The system described in Appendix 1, characterized by the features described herein. (Note 45) The aforementioned knowledge base unit is, It estimates user sentiment and prioritizes the knowledge base based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 46) The aforementioned knowledge base unit is, When updating the knowledge base, the order of updates will be adjusted based on when questions and answers were submitted. The system described in Appendix 1, characterized by the features described herein. [Explanation of Symbols]

[0209] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. A collection unit that collects question-and-answer patterns from past emails and communication tool history, An analysis unit analyzes the patterns collected by the aforementioned collection unit and generates information for the agent to intervene and respond, A generation unit in which an agent generates a response based on the information generated by the analysis unit, A providing unit that provides the answer generated by the generation unit, The system includes a correction unit in which a human provides a corrected response if the response provided by the aforementioned provision unit needs correction, and this correction unit is stored so that the agent can refine future response patterns. A system characterized by the following features.

2. It includes a notification unit that notifies the staff member of the generated response content. The system according to feature 1.

3. It features a learning unit that learns from corrected responses, allowing the agent to refine future response patterns. The system according to feature 1.

4. It includes a knowledge base section that accumulates patterns of past questions and answers and provides a database for agents to refer to. The system according to feature 1.

5. The aforementioned collection unit is The system estimates the user's emotions and selects question-and-answer patterns to collect based on those estimated emotions. The system according to feature 1.

6. The aforementioned collection unit is When collecting Q&A data, the system analyzes the user's past communication history to select the most suitable collection method. The system according to feature 1.

7. The aforementioned collection unit is When collecting questions and answers, filter them based on the user's current projects and areas of interest. The system according to feature 1.

8. The aforementioned collection unit is It estimates the user's emotions and determines the priority of the questions and answers to be collected based on the estimated user emotions. The system according to feature 1.

9. The aforementioned collection unit is When collecting Q&A data, the system prioritizes collecting highly relevant patterns by considering the user's geographical location. The system according to feature 1.

10. The aforementioned collection unit is When collecting Q&A data, analyze users' social media activity and collect relevant patterns. The system according to feature 1.

Images