System

Abstract

To provide a system for preventing the deterioration of productivity by reducing time and labor for dealing with electronic mail in a company.SOLUTION: The specific processing unit 290 of the data processing device 12 in the system executes a process of periodically detecting a new e-mail and capturing the content thereof, a process of analyzing the content of the captured e-mail and extracting an important keyword, a process of searching for related information from an in-house information source and an external information source based on the extracted keyword, a process of generating an appropriate answer based on the search result, and a process of displaying the generated answer on the user terminal so that the user can confirm and correct the answer.SELECTED DRAWING: Figure 2

Description

[Technical Field]

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

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

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

[0004] Responding to emails within a company often requires a great deal of time and effort, resulting in a decline in productivity. In particular, the process of checking new emails, analyzing their content, searching for information, and creating responses takes time, which can be a burden in a busy work environment. Furthermore, responding manually carries the risk of human error. Therefore, there is a need to develop a system that automates the email response process, improves time efficiency, and ensures reliable responses. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides a system having the following features:

[0006] The system includes a means for periodically detecting new e-mails and capturing their contents, a means for analyzing the contents of the captured e-mails and extracting important keywords, a means for searching for related information from internal and external information sources based on the extracted keywords, a means for generating appropriate answers based on the search results, and a means for displaying the generated answers on a user terminal so that the user can check and correct them.

[0007] This automates the email response process, reducing the user's workload while enabling quick and accurate responses. Furthermore, by using natural language processing technology to automatically generate responses, it is possible to further improve efficiency. Furthermore, by utilizing past emails, the company's intranet, and the company's knowledge base as internal information sources, more appropriate and consistent responses can be obtained.

[0008] "New email" refers to unread email that has just arrived in the user's mailbox.

[0009] "Periodic detection" means that the system automatically checks the email box at regular intervals to determine whether there are any new emails.

[0010] "Capture" means that the system stores the contents of the newly arrived email internally and makes them available for subsequent processing.

[0011] "Content analysis" means analyzing the body and subject of an email using natural language processing technology to extract important information and keywords.

[0012] "Extracting important keywords" refers to identifying and isolating words and phrases that should be of particular interest from the content of the analyzed emails.

[0013] "Internal and external information sources" refers to a collection of information that can be referenced to generate an answer, such as past email history, company intranet, company knowledge base, and external internet information.

[0014] "Searching for related information" means locating the necessary information and data within internal and external sources based on the extracted keywords.

[0015] "Generating an answer" means automatically creating an appropriate response sentence based on the search results.

[0016] "Display on user terminal" means that the generated answer can be viewed on the screen of a computer or device operated by the user.

[0017] "Confirm and correct" means that the user can check the generated answer and change the content if necessary.

[0018] "Natural language processing technology" is a technology that enables computers to understand and generate human language, and is primarily used for text analysis and generation.

[0019] "Past emails" are historical data of emails that the user has previously received or sent.

[0020] An "internal intranet" is an intranet environment operated within a company, and is intended for the purpose of sharing information and communicating within the company.

[0021] An "internal knowledge base" is a database in which knowledge and information accumulated within a company are organized and stored. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2]1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0023] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0024] First, the terms used in the following description will be explained.

[0025] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0026] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0027] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0028] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0029] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0030] [First embodiment]

[0031] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0032] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0033] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0034] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0035] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0036] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0037] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0038] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0039] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0040] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0041] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0042] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0043] An embodiment of the present invention relates to a system that automates the process of responding to emails. This system coordinates automatic email extraction, content analysis, information search, response generation, and confirmation / sending via a user interface. The system of the present invention operates primarily through the coordination of a server and a user terminal.

[0044] Server Operation

[0045] Automatic email extraction

[0046] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[0047] Content analysis

[0048] The server uses natural language processing technology to analyze the content of the email and extracts important keywords to understand the intent and context of the email. For example, the server can extract the keywords "project" and "progress" from an email with a subject such as "Regarding the progress of a new project."

[0049] Information Search

[0050] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[0051] Answer generation

[0052] The server uses a generative AI model to generate appropriate answers based on the search results. For example, based on the keywords above, it might generate an answer such as, "The current project progress is on schedule and within budget."

[0053] User terminal operation

[0054] User Interface

[0055] The user terminal presents the answer generated by the server to the user. The user can check this answer and modify it as necessary. For example, the user may modify the displayed answer by adding more specific progress information. The user then submits the final answer.

[0056] Specific examples

[0057] For example, consider the following email sent to a user:

[0058] Please tell me about the progress of your new project.

[0059] Server behavior:

[0060] The server periodically checks the mailbox to detect new mail and retrieves it.

[0061] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[0062] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[0063] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[0064] User terminal behavior:

[0065] The user terminal displays the generated answer to the user.

[0066] The user checks the answers and makes corrections, such as adding "more specific progress information."

[0067] After making corrections, the user creates a final response and sends it to the other party.

[0068] In this way, the present invention can significantly automate the process of responding to e-mails, reducing the workload of users and improving productivity.

[0069] The processing flow will be explained below.

[0070] Step 1:

[0071] The server periodically accesses the user's mailbox to check for new emails, and if there are any new emails, it imports them into the system.

[0072] Step 2:

[0073] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords.

[0074] Step 3:

[0075] The server searches for related information from internal and external sources based on the extracted keywords, utilizing past email history, the company intranet, the company knowledge base, the Internet, and other sources.

[0076] Step 4:

[0077] The server uses a generative AI model to generate appropriate answers based on the search results, and outputs the answers as syntactically and semantically consistent sentences using natural language processing technology.

[0078] Step 5:

[0079] The user terminal displays the generated answer sent from the server to the user, who can then check the displayed answer and modify it as necessary.

[0080] Step 6:

[0081] After the user has confirmed and corrected the answer, he or she clicks the send button to send the final answer. At this time, the user's operation details are sent to the server and then to the other party as an e-mail.

[0082] In this way, the process of responding to emails is automated, significantly reducing the amount of work required by the user.

[0083] Example 1

[0084] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0085] The traditional email response process requires a lot of manual work, which is time-consuming and labor-intensive, reducing the user's work efficiency. Furthermore, in order to understand the content of an email and respond appropriately, the user must analyze each piece of content and generate a response, which places a heavy burden on the user. Therefore, there is a need for a system that automates the email response process, reduces the user's workload, and improves productivity.

[0086] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0087] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords using natural language processing technology, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating appropriate answers based on the search results using a generative AI model, means for displaying the generated answers on a user terminal so that the user can confirm and modify them, and means for sending the final answers. This allows the email response process to be largely automated, reducing the user's workload and improving productivity.

[0088] "New email" is a new electronic message that has arrived in a user's mailbox.

[0089] "Periodic detection" means that new e-mails are automatically checked according to a preset time interval.

[0090] The "means for capturing" is a function for storing the contents of detected new e-mails in the system.

[0091] "Analyzing the content" refers to analyzing the text information of the acquired email using natural language processing technology.

[0092] "Natural language processing technology" is a general term for technologies that enable computers to understand, interpret, and generate human language.

[0093] "Extracting important keywords" means identifying key words and phrases from the content of an email that are necessary to understand its intent and context.

[0094] "Internal sources" refer to information repositories within an organization, including archived emails, internal intranets, and internal knowledge bases.

[0095] "External information sources" refer to repositories of information that can be obtained from outside the organization, primarily including the Internet.

[0096] "Searching for related information" refers to using the extracted keywords to research and obtain related information from internal and external sources.

[0097] A "generative AI model" refers to an algorithm or system that uses artificial intelligence technology to generate sentences in natural language.

[0098] "Means for generating answers" refers to a function that uses a generative AI model to create appropriate answers based on search results.

[0099] "User terminal" refers to equipment such as a computer or mobile device used by a user.

[0100] The "means for checking and correcting" is a function that displays the generated answer to the user, and allows the user to check the content and edit it as necessary.

[0101] The "means for sending the final answer" is a function for the user to send the answer that has been confirmed and corrected by the user to the other party as an email.

[0102] An embodiment of the present invention relates to a system for automating the process of responding to emails. This system operates mainly through cooperation between a server and a user terminal.

[0103] Server Operation

[0104] Automatic email extraction

[0105] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. Specifically, it periodically connects to the mail server using a mail protocol such as IMAP or POP3 to retrieve new emails. For example, the server connects to the mail server once every hour to check for new emails.

[0106] Content analysis

[0107] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy or NLTK). During this process, the email text is tokenized, tagged with parts of speech, and important keywords are extracted. For example, the keywords "project" and "progress" are extracted from an email that asks, "Please tell us about the progress of your new project."

[0108] Information Search

[0109] The server uses the extracted keywords to search for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet searches). Specifically, it executes SQL queries using the keywords to retrieve information from the internal database. It also uses external sources such as the Google Search API.

[0110] Answer generation

[0111] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT series) based on the search results. The search results and prompt are input into the generative AI model, which generates an appropriate answer in natural language. For example, the prompt "Project progress is progressing smoothly and within budget" is input.

[0112] User terminal operation

[0113] User Interface

[0114] The user terminal displays the generated answer to the user. The user can review the displayed answer and edit it if necessary. The user interface provides text boxes and editing tools to allow the user to easily modify the answer. For example, the user can enter additional information such as "The next milestone is the start of the testing phase."

[0115] Last sent

[0116] Once the user has finished making corrections, they can finalize the answer and send it to the other party as an email. The email is sent via the user's mail server using the SMTP protocol, making it easy for the user to send the final answer.

[0117] Specific examples

[0118] Consider the following example:

[0119] Please tell me about the progress of your new project.

[0120] Server behavior:

[0121] 1. The server connects to the user's mail server once every hour and checks for new email.

[0122] 2. Detect new emails and save their contents to a local database.

[0123] 3. Use Spacy to tokenize the text and extract the keywords "project" and "progress."

[0124] 4. Use SQL queries to retrieve relevant information from an internal knowledge base and use the Google Search API to retrieve information from the internet.

[0125] 5. The prompt sentence "Project progress is progressing smoothly and within budget" is input into the generative AI model, which generates the answer "Current project progress is progressing as planned and within budget."

[0126] User terminal behavior:

[0127] 1. The user's device displays the generated answer, and the user confirms the content.

[0128] 2. The user corrects their answer by adding additional information: "The next milestone is the start of the testing phase."

[0129] 3. The user confirms the revised answer and clicks the submit button.

[0130] 4. The final response is sent via the mail server using the SMTP protocol.

[0131] Prompt Sentence Examples

[0132] You receive an email asking, "How's your new project going?" Please generate an answer based on the following information:

[0133] Project name: AI system development project

[0134] Progress: 50% complete

[0135] Next Milestone: Testing Phase Begins

[0136] Budget status: On track and within budget

[0137] Using this prompt, the appropriate generative AI model will respond, generating the answer, "The AI ​​system development project is now 50% complete and is progressing towards the next milestone, the start of the testing phase. We are on track and within budget."

[0138] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0139] Step 1:

[0140] Automatic email extraction

[0141] The server connects to the user's mail server (e.g., using IMAP or POP3 protocol) at set intervals to detect and retrieve new emails. The input is the connection information to the user's mail server, and the output is the detected new email data. Specifically, the server connects to the mail server once every hour to check for new emails and saves the detected new emails in a local database.

[0142] Step 2:

[0143] Content analysis

[0144] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy, NLTK). The input is the text data of the email, and the output is important keywords extracted through analysis. The server tokenizes the text and tags it with parts of speech. For example, the server extracts the keywords "project" and "progress" from an email that says, "Please tell us about the progress of the new project."

[0145] Step 3:

[0146] Information Search

[0147] Based on the extracted keywords, the server searches for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet search). The input is the extracted keywords and the output is related information. The server executes SQL queries to retrieve related information from the internal database and uses the Google Search API to retrieve related information from the Internet.

[0148] Step 4:

[0149] Answer generation

[0150] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT series) based on the search results. The input is the search results and prompt, and the output is the generated answer. The server generates a prompt, "Project progress is progressing smoothly and within budget," and inputs it into the generative AI model. This generates an answer, "Current project progress is proceeding as planned and within budget."

[0151] Step 5:

[0152] User verification and correction

[0153] The user terminal displays the generated answer to the user. The user checks this answer and makes corrections as necessary. The input is the generated answer, and the output is the corrected answer. The user interface has text boxes and editing tools, allowing the user to enter additional information, such as "The next milestone is the start of the test phase," and correct the answer.

[0154] Step 6:

[0155] Final submission of response

[0156] The user terminal confirms the final answer after the user has completed the corrections and sends it as an email. The input is the corrected answer text, and the output is the sent email. Specifically, the user confirms the corrected answer and clicks the send button. The user terminal uses the SMTP protocol to connect to the mail server and send the final answer.

[0157] (Application example 1)

[0158] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0159] Conventional email response systems require users to manually check emails, analyze their content, and create responses, resulting in significant time and effort. Furthermore, in today's world, where rapid responses to inquiries are required, there is a growing need for systems that can efficiently and appropriately respond. In particular, content distribution services require rapid and accurate responses because of the wide variety of user inquiries. The present invention aims to solve these problems and significantly automate the email response process.

[0160] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0161] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using a generation system based on the search results, means for displaying the generated answer on an output device so that a user can confirm and modify it by operating the output device, means for sending the generated answer as a reply to an inquiry, means for calling a generative AI model using a prompt sentence and generating an answer, and document generation means for automatically generating an appropriate answer to an inquiry. This allows users to efficiently automate the process from email content analysis to answer generation, enabling quick and accurate responses.

[0162] "New email" refers to new electronic mail that arrives in your inbox periodically.

[0163] "Content capture means" refers to the process or technology used to capture email content into the system.

[0164] "Means of content analysis" refers to the process of analyzing the content of captured emails using natural language processing technology to understand their intent and context.

[0165] "Important keywords" refer to words extracted from the content of emails that are important for extracting specific information.

[0166] "Internal sources" refers to information repositories maintained within internal systems, historical email history, intranets, and knowledge bases.

[0167] "External information sources" refers to information resources outside the company, such as the Internet.

[0168] "Means for searching related information" refers to the process of finding related information from internal and external sources based on the extracted keywords.

[0169] "Means for generating appropriate answers" refers to the process of using AI technology to create appropriate answers based on search results.

[0170] "Output device" refers to an appliance or device for displaying generated answers to a user.

[0171] "Means of invoking a generative AI model using a prompt sentence to generate an answer" refers to the process of running a generative AI model using a specified prompt sentence to construct an appropriate answer.

[0172] "Document generation means for automatically generating appropriate responses to inquiries" refers to a process that uses AI technology to automatically generate appropriate documents in response to user inquiries.

[0173] "Means for sending a reply to a query" refers to the technology or process for sending the generated answer back to the user.

[0174] This invention relates to a system for automating the process of responding to emails. This system automates everything from detecting new emails to generating and sending responses, thereby reducing the workload of users.

[0175] System Program Overview

[0176] The system mainly operates on a server and a user device. The server periodically detects new emails and analyzes their contents. Based on the analyzed important keywords, it searches for related information from internal and external sources. It then uses a generative AI model to generate an appropriate answer, which is displayed on the user's device. The user can then confirm and modify the answer, and send the final answer to the other party.

[0177] Hardware and software configuration

[0178] server:

[0179] The server uses the following software and libraries:

[0180] IMAP4 library: Used to retrieve new emails.

[0181] Python email module: Used to analyze the content of retrieved emails.

[0182] Generative AI models (OpenAI's GPT-3 / 4): Use natural language processing techniques to generate answers.

[0183] User device:

[0184] The user terminal uses the following devices and applications:

[0185] Output device (e.g. smartphone, tablet, PC): Displays the answers from the server and allows the user to check and correct them.

[0186] Data processing and calculation

[0187] The server processes data in the following steps: First, it periodically detects new emails using the IMAP4 library and analyzes their contents using the email module. This analysis extracts important keywords. Next, based on the extracted keywords, it searches for related information from internal sources (past emails, intranets, knowledge bases) and external sources (the Internet). It then uses a generative AI model to generate appropriate answers based on the search results. The generated answers are sent to the user's device, where they can be reviewed and revised by the user.

[0188] Specific examples

[0189] For example, consider a case where a user makes a query such as "What are the latest movie listings?"

[0190] The server detects new emails and analyzes the query content. Based on the extracted keywords "latest" and "movie list," it searches for related information from internal and external sources. Based on the results, it queries the generative AI model using the following prompt:

[0191] Example prompt sentence:

[0192] User Question: What is the latest movie list?

[0193] Generate the appropriate answer.

[0194] The generative AI model generates an appropriate answer based on this prompt, for example, "The latest movie list is as follows: 1. Inception 2. Avatar 2 3. Titanic."

[0195] The generated answer is displayed on the user's terminal, where the user can check the content and make corrections as necessary, after which the user can submit the final answer.

[0196] This invention enables users to respond to inquiries quickly and accurately, thereby improving work efficiency.

[0197] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0198] Step 1:

[0199] The server periodically checks for new emails. Using the IMAP4 library, it connects to the specified mail server and retrieves unread emails. At this point, the input is the mail server authentication information and the user's mailbox, and the output is a list of unread emails.

[0200] Step 2:

[0201] The server analyzes the contents of the retrieved email. It uses the Python email module to analyze the email content (body, subject, sender information, etc.). As a result of the analysis, it extracts important keywords from the subject and body of the email. At this point, the input is the content of the unread email, and the output is the extracted important keywords.

[0202] Step 3:

[0203] The server searches for related information based on the extracted keywords. It searches for related information from internal sources (past email history, intranet, knowledge base) and external sources (Internet). For example, it searches for related information using keywords such as "latest" and "movie list." At this point, the input is the extracted keywords, and the output is a list of related information.

[0204] Step 4:

[0205] The server uses a generation system to generate an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-3 or GPT-4), it inputs a prompt such as "User question: What are the latest movie listings? Please generate an appropriate answer." and receives the generated answer. At this point, the input is the prompt and related information, and the output is the generated answer.

[0206] Step 5:

[0207] The server sends the generated answer to the user terminal. The generated answer is displayed on the user terminal via the transmission system. At this point, the input is the generated answer, and the output is the display result on the user terminal.

[0208] Step 6:

[0209] The user checks the displayed answer and makes corrections if necessary. The user checks the answer using a device such as a smartphone or PC and makes corrections if necessary. The input at this point is the displayed answer and the user's corrections, and the output is the final answer after corrections.

[0210] Step 7:

[0211] The user sends the final answer to the other party. The user sends the revised answer from the user terminal to the system, and the system returns it to the other party. At this point, the input is the revised final answer, and the output is the sent answer.

[0212] The above steps automate the email response process, enabling fast and accurate responses.

[0213] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0214] An embodiment of the present invention relates to a system that automates the process of responding to emails and generates appropriate responses by recognizing user emotions. This system coordinates automatic email extraction, content analysis, information search, response generation, emotion recognition, and confirmation / sending via a user interface. The system of the present invention operates through the cooperation of a server, an emotion engine, and a user terminal.

[0215] Server Operation

[0216] Automatic email extraction

[0217] The server periodically accesses the user's mailbox to detect new emails and import them into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[0218] Content analysis

[0219] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words, and extracts important keywords.

[0220] Information Search

[0221] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[0222] Answer generation

[0223] The server uses a generative AI model to generate an appropriate answer based on the search results. For example, from an email about the progress of a new project, the server might generate an answer such as, "The current project is progressing on schedule and within budget."

[0224] Emotion Engine Operation

[0225] emotion recognition

[0226] The emotion engine analyzes emotions in real time based on user input and reactions, using keyboard input, mouse movements, and even biometric signals.

[0227] Emotion-based response adjustment

[0228] The emotion engine recognizes the user's emotions and adjusts the content of the generated responses based on the results. For example, if the user is feeling stressed, the responses can be tailored to be more kind and comforting.

[0229] Use of past emotion data

[0230] The emotion engine stores the user's past emotional data for a certain period of time and uses it to more accurately recognize emotions and respond, making it possible to respond based on the user's long-term emotional tendencies.

[0231] User terminal operation

[0232] User Interface

[0233] The user terminal presents the answer generated by the server to the user. The user can review the answer and modify it as necessary. For example, the user can modify the answer by adding a specific number. The final answer is then sent.

[0234] Specific examples

[0235] For example, consider the following email sent to a user:

[0236] Please tell me about the progress of your new project.

[0237] Server behavior:

[0238] The server periodically checks the mailbox to detect new mail and retrieves it.

[0239] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[0240] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[0241] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[0242] Emotion Engine in action:

[0243] The emotion engine analyzes the user's keyboard input and mouse movements to recognize when the user is feeling stressed.

[0244] Based on the recognition results, a comforting phrase such as "The current project is progressing very well. Please rest assured." is added to the generated answer.

[0245] User terminal behavior:

[0246] The user terminal displays the generated answer to the user.

[0247] The user checks the answers and makes corrections if necessary.

[0248] After making corrections, the user creates a final response and sends it to the other party.

[0249] This not only automates the process of responding to emails, but also enables appropriate responses that take into consideration the user's feelings, thereby achieving even higher levels of customer satisfaction.

[0250] The processing flow will be explained below.

[0251] Step 1:

[0252] The server periodically accesses the user's mailbox to check for new emails. For example, it accesses the mail server every five minutes to check if new emails have arrived. If there are any new emails, it imports them into the system.

[0253] Step 2:

[0254] The server analyzes the content of the emails it receives. Specifically, it uses natural language processing technology to divide the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords. For example, from an email that asks, "Please tell me about the progress of the new project," it extracts keywords such as "new project" and "progress."

[0255] Step 3:

[0256] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. For example, the server searches the company's intranet using the keywords "new project" and "progress" to retrieve relevant project plans and progress reports.

[0257] Step 4:

[0258] The server compiles the information it has acquired and uses a generative AI model to generate an appropriate response, such as "The current project is progressing on schedule and within budget."

[0259] Step 5:

[0260] The emotion engine analyzes the user's input and other biometric signals (e.g., keyboard typing speed and mouse movements) to recognize the user's emotional state. For example, if the user types quickly, it may determine that the user is stressed.

[0261] Step 6:

[0262] The emotion engine adjusts the generated answers based on the perceived emotion, for example, if it detects that the user is stressed, it will add a comforting element such as "Don't worry, progress on the current project is going very well."

[0263] Step 7:

[0264] The user terminal displays the generated answer received from the server to the user. The user can check the displayed answer and modify the content as necessary. For example, the user can modify the answer by adding specific progress data.

[0265] Step 8:

[0266] After checking and correcting the answer, the user clicks the send button to send it to the other party. At this time, the user's operation is sent to the server, and the final email is sent to the other party, completing the response.

[0267] By using this process flow, the present invention can automate the user's email response process and also realize an appropriate response that takes into consideration the user's feelings.

[0268] Example 2

[0269] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0270] Conventional email response systems require users to manually analyze the content of received emails and create replies, which requires a great deal of time and effort. Furthermore, the quality of responses can be poor because responses cannot take into account the user's feelings. The present invention aims to solve these problems by improving the efficiency of email response processes and providing high-quality responses that take into account the user's feelings.

[0271] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for periodically detecting new e-mails and capturing their contents, means for analyzing the contents of the captured e-mails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using an artificial intelligence model based on the search results, means for recognizing the user's emotions and adjusting the content of the generated answer based on the emotions, and means for displaying the generated answer on the user terminal so that the user can confirm and correct it. This makes it possible to automate the e-mail response process and provide an appropriate response that takes the user's emotions into consideration.

[0272] "Newly arrived email" refers to unread email that has been newly received by the user.

[0273] "Periodic detection" means that e-mails are automatically checked at preset time intervals.

[0274] "Means for retrieving" refers to the function of retrieving emails from an email server and storing them within the system.

[0275] "Content analysis" refers to analyzing the body and subject of an email using natural language processing techniques.

[0276] "Key word extraction" is the process of identifying and extracting specific tokens or phrases from the analyzed content.

[0277] "Internal sources" refer to documents, databases, and past communications records shared within an organization.

[0278] "External sources" refer to publicly available data on the Internet, information provided by other companies or organizations, etc.

[0279] "Searching for related information" means searching for relevant data from predefined information sources based on the extracted keywords.

[0280] "Artificial intelligence model" refers to an intelligent system that uses machine learning algorithms to generate output based on input data.

[0281] "Means for generating a response" refers to the function of constructing an appropriate reply based on the relevant information obtained.

[0282] "Recognizing user emotions" is the process of inferring a user's emotional state from their behavior and biosignals.

[0283] "Adjusting the content of the generated response based on emotion" refers to modifying the content and tone of the generated response according to the recognized emotion.

[0284] "User terminal" refers to a device such as a computer, smartphone, or tablet that is actually operated by a user.

[0285] "Means for checking and correcting" refers to a function that allows the user to check the automatically generated answers and manually correct them if necessary.

[0286] The present invention is a system that integrates an automatic email processing function with a user emotion recognition function. This system operates by linking a server, an emotion engine, and a user terminal.

[0287] Server Operation

[0288] Automatic email extraction

[0289] The server periodically accesses the user's email box to detect new emails, for example by connecting to a mail server such as Gmail or Outlook using the IMAP protocol to retrieve new emails.

[0290] Content analysis

[0291] The server then analyzes the email content using natural language processing techniques, using the Python Natural Language Toolkit (NLTK) library to tokenize the email body and subject line and extract key keywords.

[0292] Information Search

[0293] The server uses the extracted keywords to search for relevant information from internal sources (databases, knowledge bases, intranets) and external sources (the Internet) using SQL queries to the internal database and Google Search APIs.

[0294] Answer generation

[0295] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT-3) based on the search results. The generative AI model receives a prompt such as: "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." An example of a generated answer is, "Current project progress is on schedule and within budget."

[0296] Emotion Engine Operation

[0297] emotion recognition

[0298] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, and even biometric signals. Emotion recognition uses machine learning libraries such as TensorFlow and PyTorch to identify emotions such as stress or joy that the user is feeling.

[0299] Emotion-based response adjustment

[0300] The content of the generated answer is adjusted based on the results of emotion recognition. For example, if the user is feeling stressed, the answer obtained from the generative AI model will include comforting words such as "Please continue to watch over the project with peace of mind."

[0301] Use of past emotion data

[0302] The emotion engine stores the user's past emotional data for a certain period of time and uses this data to more accurately recognize and respond to emotions, enabling advanced responses based on the user's long-term emotional tendencies.

[0303] User terminal operation

[0304] User Interface

[0305] The user's device displays the answer generated by the server to the user. The user can review the answer and make corrections as necessary. For example, a front-end framework such as ReactJS or Vue.js can be used to display the answer in an easy-to-understand format. After the user makes corrections to the answer, such as adding specific numbers, the final answer can be submitted.

[0306] This system automates the email response process and enables appropriate responses that take into consideration the user's feelings, which contributes to increased customer satisfaction and dramatically improves business efficiency.

[0307] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0308] Step 1: Automatic email extraction

[0309] The server periodically accesses the user's mailbox to detect new emails. The server connects to the mail server using the IMAP protocol and retrieves new emails. The input includes the user's mailbox information and access frequency settings. The output is the data of new emails that are imported into the server. Specifically, for example, the server automatically synchronizes with the mail server every morning at 8:00 and detects new emails.

[0310] Step 2: Analyzing the email content

[0311] The server analyzes the content of the newly received emails using natural language processing technology. The input includes the data of the newly received emails. Specifically, it uses Python's NLTK library to tokenize the email body and subject, removes unnecessary words, and extracts important keywords using techniques such as TF-IDF. The output is the extracted keywords.

[0312] Step 3: Information search

[0313] The server searches for related information from internal information sources (databases, knowledge bases, intranets) and external information sources (the Internet) based on the extracted keywords. The input includes the extracted keywords. Specifically, it executes an SQL query to the internal database to retrieve the relevant data. It also retrieves related information from external information sources using the Google Search API. The output is the retrieved related information.

[0314] Step 4: Answer Generation

[0315] The server uses a generative AI model (such as GPT-3) to generate an appropriate answer based on the search results. The input includes the relevant information obtained as search results. An example of a prompt is, "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." Specifically, the prompt and search results are input into the generative AI model to generate an answer. The output is the generated answer.

[0316] Step 5: Emotion Recognition

[0317] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, biometric signals, etc. Inputs include user behavioral data (e.g., typing speed, mouse movements, etc.). Specifically, it uses machine learning models (TensorFlow or PyTorch) to identify the user's emotions. The output is the user's emotional state.

[0318] Step 6: Adjust your responses based on emotion

[0319] The emotion engine adjusts the content of the generated answer based on the analyzed user's emotions. The input includes the generated answer and the user's emotional state. For example, if the user is feeling stressed, the engine adds a phrase such as "Please continue to monitor the project with peace of mind." The output is an answer adjusted to take the user's emotions into account.

[0320] Step 7: Present and send the answer to the user

[0321] The user device presents the generated answer to the user and provides an interface that allows the user to review and modify it. The input includes the adjusted answer. Specific operations include displaying the answer in a user interface using a front-end framework such as ReactJS or Vue.js. The user can review the answer and modify it, such as adding a specific number. The user then creates and submits the final answer. The output is the modified final answer, which is then sent to the other party.

[0322] (Application example 2)

[0323] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0324] In conventional email response systems, the process of analyzing email content and generating appropriate responses required a lot of time and effort. Furthermore, if store clerks were unable to respond to customer inquiries immediately, customer satisfaction could decline. Furthermore, because responses were not flexibly adjusted based on customer emotions, there were limitations to how sensitive the system could be to customer emotions. Therefore, there is a need for a system that automates the processing of emails and inquiries and generates appropriate responses based on emotion recognition.

[0325] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer based on the search results, means for displaying the generated answer on a user terminal so that the user can confirm and modify it, means for adjusting the generated answer based on emotion recognition, and means for accepting customer inquiries in-store using QR codes. This makes it possible to automate the processing of emails and inquiries and provide flexible responses that take emotions into consideration.

[0326] "New email" is unread email that has just arrived in your inbox.

[0327] The "means for acquiring content" is a device or program that has the function of acquiring new e-mails into the system.

[0328] "Analyzing the content" means analyzing the text data of the imported email using natural language processing technology.

[0329] The "means for extracting important keywords" is a device or program that identifies and extracts keywords with specific meanings from the analyzed text data.

[0330] "Internal information sources" are information resources such as databases and knowledge bases within a company.

[0331] "External sources" are information resources outside the company, such as information on the Internet or other public databases.

[0332] The "means for searching related information" is a device or program that searches for relevant data from internal and external sources based on the extracted keywords.

[0333] The "means for generating an appropriate answer" is a device or program that generates an appropriate response to a query based on the search results.

[0334] A "user terminal" is a computer device or mobile device that can be operated by a user.

[0335] The "means for confirming and correcting" is a device or program that has the function of allowing the user to confirm the generated answer and make corrections as necessary.

[0336] The "means for adjusting based on emotion recognition" is a device or program that analyzes the user's input and response, recognizes the emotion, and modifies the response appropriately.

[0337] "Means for accepting inquiries in-store using QR codes" refers to a device or program that has the function of scanning QR codes and accepting inquiries electronically.

[0338] A system embodying the present invention automates email correspondence and can generate appropriate responses by recognizing the user's emotions. An embodiment of this system will now be described in detail.

[0339] System configuration

[0340] The system consists of a server, a user terminal, an emotion engine, and a QR code scanning device. The server has the following main functions:

[0341] Ability to periodically detect new emails and retrieve their contents

[0342] Ability to analyze email content and extract important keywords

[0343] Ability to search for related information from internal and external sources based on extracted keywords

[0344] Generate relevant answers based on search results

[0345] A function that displays the generated answers on the user's device so that the user can check and correct them.

[0346] The ability to tailor generated answers based on emotion recognition

[0347] A function to accept customer inquiries in-store using QR codes

[0348] Hardware and software used

[0349] Hardware: Servers, smartphones, smart glasses, QR code scanners

[0350] software:

[0351] Natural language processing: SpaCy(ja_core_news_sm)

[0352] Generative AI model: GPT-2 (transformers)

[0353] Email communication: smtplib, imaplib

[0354] Emotion recognition: keyboard input, mouse movement analysis software, and possibly biometric sensors

[0355] Operation process

[0356] Server Operation

[0357] The server periodically connects to the mail server to detect new emails. The detected emails are imported and their contents are analyzed to extract important keywords. Based on the extracted keywords, relevant information is searched from internal sources (past emails, internal intranet, internal knowledge base) and external sources (the Internet). Based on the search results, an appropriate answer is generated using a generative AI model (GPT-2). The following prompt sentences are used in this generation process:

[0358] Example prompt:

[0359] "Please answer the following questions: Project Status\n\nReply: "

[0360] Emotion Engine Operation

[0361] The emotion engine recognizes emotions by analyzing user input and reactions. For example, it can assess emotions in real time using keyboard typing speed and strength, mouse movements, and biometric sensors. The emotion engine adjusts the content of the generated response based on the analysis results. For example, if the user is feeling stressed, it will add comforting words such as "Don't worry" to the generated response.

[0362] User terminal operation

[0363] The generated answer is displayed on the user's device, where the user can review it and make any necessary corrections. The final answer is sent at the user's discretion. In-store inquiries are also accepted through a QR code scanner, which is also captured in the system and goes through the same process.

[0364] Specific examples

[0365] For example, if a user receives an email like this:

[0366] Please tell me about the progress of your new project.

[0367] The server detects this email, analyzes its contents, and extracts the keywords "project" and "progress." Based on this, a generative AI model is used to generate the answer, "The current project is progressing smoothly." If the emotion engine recognizes the user's stress, it further adjusts the answer to take their emotions into consideration, such as, "The current project is progressing smoothly. Please rest assured."

[0368] This allows for efficient handling of emails and in-store inquiries, and enables flexible, emotional responses.

[0369] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0370] Step 1:

[0371] The server periodically detects new emails and retrieves their contents. The server accesses the user's mailbox and retrieves unread emails. The input is the new email data from the mail server, and the output is the retrieved email contents.

[0372] Step 2:

[0373] The server analyzes the content of the captured email and extracts important keywords. The server uses natural language processing software (e.g., SpaCy) to split the text into tokens, filter out meaningless words, and extract important keywords. The input is the email content, and the output is the extracted keywords.

[0374] Step 3:

[0375] The server searches for related information from internal and external sources based on the extracted keywords. The server sends search queries to internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet) to collect related information. The input is the extracted keywords, and the output is the search results.

[0376] Step 4:

[0377] The server generates an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-2), the server inputs keywords and search results into a prepared prompt sentence to generate a natural language answer. The input is the search results, and the output is the generated answer. An example of a prompt sentence is as follows:

[0378] "Please answer the following questions: Project Status\n\nReply: "

[0379] Step 5:

[0380] The emotion engine analyzes user input and reactions to recognize emotions. The emotion engine analyzes keyboard input speed, keystroke strength, mouse movement, and, in some cases, data from biometric sensors. The input is user input and reaction data, and the output is emotional data that indicates the user's state.

[0381] Step 6:

[0382] The emotion engine adjusts the generated answer based on the recognized emotion data. For example, if it recognizes that the user is stressed, it adds comforting words such as "Don't worry" to the answer. The input is the generated answer and emotion data, and the output is the adjusted answer.

[0383] Step 7:

[0384] The user terminal displays the adjusted answer. The user reviews the adjusted answer, makes corrections if necessary, and finally submits the answer. The input is the adjusted answer, and the output is the final answer that has been reviewed and corrected by the user.

[0385] Step 8:

[0386] The server accepts inquiries from customers using QR codes in the store. The inquiry content obtained by scanning the QR code is imported into the system and goes through a process similar to that for email. The input is the inquiry content obtained from the QR code, and the output is the inquiry data imported into the system.

[0387] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0388] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0389] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0390] [Second embodiment]

[0391] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0392] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0393] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0394] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0395] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0396] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0397] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0398] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0399] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0400] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0401] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0402] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0403] An embodiment of the present invention relates to a system that automates the process of responding to emails. This system coordinates automatic email extraction, content analysis, information search, response generation, and confirmation / sending via a user interface. The system of the present invention operates primarily through the coordination of a server and a user terminal.

[0404] Server Operation

[0405] Automatic email extraction

[0406] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[0407] Content analysis

[0408] The server uses natural language processing technology to analyze the content of the email and extracts important keywords to understand the intent and context of the email. For example, the server can extract the keywords "project" and "progress" from an email with a subject such as "Regarding the progress of a new project."

[0409] Information Search

[0410] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[0411] Answer generation

[0412] The server uses a generative AI model to generate appropriate answers based on the search results. For example, based on the keywords above, it might generate an answer such as, "The current project progress is on schedule and within budget."

[0413] User terminal operation

[0414] User Interface

[0415] The user terminal presents the answer generated by the server to the user. The user can check this answer and modify it as necessary. For example, the user may modify the displayed answer by adding more specific progress information. The user then submits the final answer.

[0416] Specific examples

[0417] For example, consider the following email sent to a user:

[0418] Please tell me about the progress of your new project.

[0419] Server behavior:

[0420] The server periodically checks the mailbox to detect new mail and retrieves it.

[0421] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[0422] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[0423] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[0424] User terminal behavior:

[0425] The user terminal displays the generated answer to the user.

[0426] The user checks the answers and makes corrections, such as adding "more specific progress information."

[0427] After making corrections, the user creates a final response and sends it to the other party.

[0428] In this way, the present invention can significantly automate the process of responding to e-mails, reducing the workload of users and improving productivity.

[0429] The processing flow will be explained below.

[0430] Step 1:

[0431] The server periodically accesses the user's mailbox to check for new emails, and if there are any new emails, it imports them into the system.

[0432] Step 2:

[0433] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords.

[0434] Step 3:

[0435] The server searches for related information from internal and external sources based on the extracted keywords, utilizing past email history, the company intranet, the company knowledge base, the Internet, and other sources.

[0436] Step 4:

[0437] The server uses a generative AI model to generate appropriate answers based on the search results, and outputs the answers as syntactically and semantically consistent sentences using natural language processing technology.

[0438] Step 5:

[0439] The user terminal displays the generated answer sent from the server to the user, who can then check the displayed answer and modify it as necessary.

[0440] Step 6:

[0441] After the user has confirmed and corrected the answer, he or she clicks the send button to send the final answer. At this time, the user's operation details are sent to the server and then to the other party as an e-mail.

[0442] In this way, the process of responding to emails is automated, significantly reducing the amount of work required by the user.

[0443] Example 1

[0444] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0445] The traditional email response process requires a lot of manual work, which is time-consuming and labor-intensive, reducing the user's work efficiency. Furthermore, in order to understand the content of an email and respond appropriately, the user must analyze each piece of content and generate a response, which places a heavy burden on the user. Therefore, there is a need for a system that automates the email response process, reduces the user's workload, and improves productivity.

[0446] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0447] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords using natural language processing technology, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating appropriate answers based on the search results using a generative AI model, means for displaying the generated answers on a user terminal so that the user can confirm and modify them, and means for sending the final answers. This allows the email response process to be largely automated, reducing the user's workload and improving productivity.

[0448] "New email" is a new electronic message that has arrived in a user's mailbox.

[0449] "Periodic detection" means that new e-mails are automatically checked according to a preset time interval.

[0450] The "means for capturing" is a function for storing the contents of detected new e-mails in the system.

[0451] "Analyzing the content" refers to analyzing the text information of the acquired email using natural language processing technology.

[0452] "Natural language processing technology" is a general term for technologies that enable computers to understand, interpret, and generate human language.

[0453] "Extracting important keywords" means identifying key words and phrases from the content of an email that are necessary to understand its intent and context.

[0454] "Internal sources" refer to information repositories within an organization, including archived emails, internal intranets, and internal knowledge bases.

[0455] "External information sources" refer to repositories of information that can be obtained from outside the organization, primarily including the Internet.

[0456] "Searching for related information" refers to using the extracted keywords to research and obtain related information from internal and external sources.

[0457] A "generative AI model" refers to an algorithm or system that uses artificial intelligence technology to generate sentences in natural language.

[0458] "Means for generating answers" refers to a function that uses a generative AI model to create appropriate answers based on search results.

[0459] "User terminal" refers to equipment such as a computer or mobile device used by a user.

[0460] The "means for checking and correcting" is a function that displays the generated answer to the user, and allows the user to check the content and edit it as necessary.

[0461] The "means for sending the final answer" is a function for the user to send the answer that has been confirmed and corrected by the user to the other party as an email.

[0462] An embodiment of the present invention relates to a system for automating the process of responding to emails. This system operates mainly through cooperation between a server and a user terminal.

[0463] Server Operation

[0464] Automatic email extraction

[0465] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. Specifically, it periodically connects to the mail server using a mail protocol such as IMAP or POP3 to retrieve new emails. For example, the server connects to the mail server once every hour to check for new emails.

[0466] Content analysis

[0467] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy or NLTK). During this process, the email text is tokenized, tagged with parts of speech, and important keywords are extracted. For example, the keywords "project" and "progress" are extracted from an email that asks, "Please tell us about the progress of your new project."

[0468] Information Search

[0469] The server uses the extracted keywords to search for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet searches). Specifically, it executes SQL queries using the keywords to retrieve information from the internal database. It also uses external sources such as the Google Search API.

[0470] Answer generation

[0471] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT series) based on the search results. The search results and prompt are input into the generative AI model, which generates an appropriate answer in natural language. For example, the prompt "Project progress is progressing smoothly and within budget" is input.

[0472] User terminal operation

[0473] User Interface

[0474] The user terminal displays the generated answer to the user. The user can review the displayed answer and edit it if necessary. The user interface provides text boxes and editing tools to allow the user to easily modify the answer. For example, the user can enter additional information such as "The next milestone is the start of the testing phase."

[0475] Last sent

[0476] Once the user has finished making corrections, they can finalize the answer and send it to the other party as an email. The email is sent via the user's mail server using the SMTP protocol, making it easy for the user to send the final answer.

[0477] Specific examples

[0478] Consider the following example:

[0479] Please tell me about the progress of your new project.

[0480] Server behavior:

[0481] 1. The server connects to the user's mail server once every hour and checks for new email.

[0482] 2. Detect new emails and save their contents to a local database.

[0483] 3. Use Spacy to tokenize the text and extract the keywords "project" and "progress."

[0484] 4. Use SQL queries to retrieve relevant information from an internal knowledge base and use the Google Search API to retrieve information from the internet.

[0485] 5. The prompt sentence "Project progress is progressing smoothly and within budget" is input into the generative AI model, which generates the answer "Current project progress is progressing as planned and within budget."

[0486] User terminal behavior:

[0487] 1. The user's device displays the generated answer, and the user confirms the content.

[0488] 2. The user corrects their answer by adding additional information: "The next milestone is the start of the testing phase."

[0489] 3. The user confirms the revised answer and clicks the submit button.

[0490] 4. The final response is sent via the mail server using the SMTP protocol.

[0491] Prompt Sentence Examples

[0492] You receive an email asking, "How's your new project going?" Please generate an answer based on the following information:

[0493] Project name: AI system development project

[0494] Progress: 50% complete

[0495] Next Milestone: Testing Phase Begins

[0496] Budget status: On track and within budget

[0497] Using this prompt, the appropriate generative AI model will respond, generating the answer, "The AI ​​system development project is now 50% complete and is progressing towards the next milestone, the start of the testing phase. We are on track and within budget."

[0498] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0499] Step 1:

[0500] Automatic email extraction

[0501] The server connects to the user's mail server (e.g., using IMAP or POP3 protocol) at set intervals to detect and retrieve new emails. The input is the connection information to the user's mail server, and the output is the detected new email data. Specifically, the server connects to the mail server once every hour to check for new emails and saves the detected new emails in a local database.

[0502] Step 2:

[0503] Content analysis

[0504] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy, NLTK). The input is the text data of the email, and the output is important keywords extracted through analysis. The server tokenizes the text and tags it with parts of speech. For example, the server extracts the keywords "project" and "progress" from an email that says, "Please tell us about the progress of the new project."

[0505] Step 3:

[0506] Information Search

[0507] Based on the extracted keywords, the server searches for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet search). The input is the extracted keywords and the output is related information. The server executes SQL queries to retrieve related information from the internal database and uses the Google Search API to retrieve related information from the Internet.

[0508] Step 4:

[0509] Answer generation

[0510] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT series) based on the search results. The input is the search results and prompt, and the output is the generated answer. The server generates a prompt, "Project progress is progressing smoothly and within budget," and inputs it into the generative AI model. This generates an answer, "Current project progress is proceeding as planned and within budget."

[0511] Step 5:

[0512] User verification and correction

[0513] The user terminal displays the generated answer to the user. The user checks this answer and makes corrections as necessary. The input is the generated answer, and the output is the corrected answer. The user interface has text boxes and editing tools, allowing the user to enter additional information, such as "The next milestone is the start of the test phase," and correct the answer.

[0514] Step 6:

[0515] Final submission of response

[0516] The user terminal confirms the final answer after the user has completed the corrections and sends it as an email. The input is the corrected answer text, and the output is the sent email. Specifically, the user confirms the corrected answer and clicks the send button. The user terminal uses the SMTP protocol to connect to the mail server and send the final answer.

[0517] (Application example 1)

[0518] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0519] Conventional email response systems require users to manually check emails, analyze their content, and create responses, resulting in significant time and effort. Furthermore, in today's world, where rapid responses to inquiries are required, there is a growing need for systems that can efficiently and appropriately respond. In particular, content distribution services require rapid and accurate responses because of the wide variety of user inquiries. The present invention aims to solve these problems and significantly automate the email response process.

[0520] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0521] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using a generation system based on the search results, means for displaying the generated answer on an output device so that a user can confirm and modify it by operating the output device, means for sending the generated answer as a reply to an inquiry, means for calling a generative AI model using a prompt sentence and generating an answer, and document generation means for automatically generating an appropriate answer to an inquiry. This allows users to efficiently automate the process from email content analysis to answer generation, enabling quick and accurate responses.

[0522] "New email" refers to new electronic mail that arrives in your inbox periodically.

[0523] "Content capture means" refers to the process or technology used to capture email content into the system.

[0524] "Means of content analysis" refers to the process of analyzing the content of captured emails using natural language processing technology to understand their intent and context.

[0525] "Important keywords" refer to words extracted from the content of emails that are important for extracting specific information.

[0526] "Internal sources" refers to information repositories maintained within internal systems, historical email history, intranets, and knowledge bases.

[0527] "External information sources" refers to information resources outside the company, such as the Internet.

[0528] "Means for searching related information" refers to the process of finding related information from internal and external sources based on the extracted keywords.

[0529] "Means for generating appropriate answers" refers to the process of using AI technology to create appropriate answers based on search results.

[0530] "Output device" refers to an appliance or device for displaying generated answers to a user.

[0531] "Means of invoking a generative AI model using a prompt sentence to generate an answer" refers to the process of running a generative AI model using a specified prompt sentence to construct an appropriate answer.

[0532] "Document generation means for automatically generating appropriate responses to inquiries" refers to a process that uses AI technology to automatically generate appropriate documents in response to user inquiries.

[0533] "Means for sending a reply to a query" refers to the technology or process for sending the generated answer back to the user.

[0534] This invention relates to a system for automating the process of responding to emails. This system automates everything from detecting new emails to generating and sending responses, thereby reducing the workload of users.

[0535] System Program Overview

[0536] The system mainly operates on a server and a user device. The server periodically detects new emails and analyzes their contents. Based on the analyzed important keywords, it searches for related information from internal and external sources. It then uses a generative AI model to generate an appropriate answer, which is displayed on the user's device. The user can then confirm and modify the answer, and send the final answer to the other party.

[0537] Hardware and software configuration

[0538] server:

[0539] The server uses the following software and libraries:

[0540] IMAP4 library: Used to retrieve new emails.

[0541] Python email module: Used to analyze the content of retrieved emails.

[0542] Generative AI models (OpenAI's GPT-3 / 4): Use natural language processing techniques to generate answers.

[0543] User device:

[0544] The user terminal uses the following devices and applications:

[0545] Output device (e.g. smartphone, tablet, PC): Displays the answers from the server and allows the user to check and correct them.

[0546] Data processing and calculation

[0547] The server processes data in the following steps: First, it periodically detects new emails using the IMAP4 library and analyzes their contents using the email module. This analysis extracts important keywords. Next, based on the extracted keywords, it searches for related information from internal sources (past emails, intranets, knowledge bases) and external sources (the Internet). It then uses a generative AI model to generate appropriate answers based on the search results. The generated answers are sent to the user's device, where they can be reviewed and revised by the user.

[0548] Specific examples

[0549] For example, consider a case where a user makes a query such as "What are the latest movie listings?"

[0550] The server detects new emails and analyzes the query content. Based on the extracted keywords "latest" and "movie list," it searches for related information from internal and external sources. Based on the results, it queries the generative AI model using the following prompt:

[0551] Example prompt sentence:

[0552] User Question: What is the latest movie list?

[0553] Generate the appropriate answer.

[0554] The generative AI model generates an appropriate answer based on this prompt, for example, "The latest movie list is as follows: 1. Inception 2. Avatar 2 3. Titanic."

[0555] The generated answer is displayed on the user's terminal, where the user can check the content and make corrections as necessary, after which the user can submit the final answer.

[0556] This invention enables users to respond to inquiries quickly and accurately, thereby improving work efficiency.

[0557] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0558] Step 1:

[0559] The server periodically checks for new emails. Using the IMAP4 library, it connects to the specified mail server and retrieves unread emails. At this point, the input is the mail server authentication information and the user's mailbox, and the output is a list of unread emails.

[0560] Step 2:

[0561] The server analyzes the contents of the retrieved email. It uses the Python email module to analyze the email content (body, subject, sender information, etc.). As a result of the analysis, it extracts important keywords from the subject and body of the email. At this point, the input is the content of the unread email, and the output is the extracted important keywords.

[0562] Step 3:

[0563] The server searches for related information based on the extracted keywords. It searches for related information from internal sources (past email history, intranet, knowledge base) and external sources (Internet). For example, it searches for related information using keywords such as "latest" and "movie list." At this point, the input is the extracted keywords, and the output is a list of related information.

[0564] Step 4:

[0565] The server uses a generation system to generate an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-3 or GPT-4), it inputs a prompt such as "User question: What are the latest movie listings? Please generate an appropriate answer." and receives the generated answer. At this point, the input is the prompt and related information, and the output is the generated answer.

[0566] Step 5:

[0567] The server sends the generated answer to the user terminal. The generated answer is displayed on the user terminal via the transmission system. At this point, the input is the generated answer, and the output is the display result on the user terminal.

[0568] Step 6:

[0569] The user checks the displayed answer and makes corrections if necessary. The user checks the answer using a device such as a smartphone or PC and makes corrections if necessary. The input at this point is the displayed answer and the user's corrections, and the output is the final answer after corrections.

[0570] Step 7:

[0571] The user sends the final answer to the other party. The user sends the revised answer from the user terminal to the system, and the system returns it to the other party. At this point, the input is the revised final answer, and the output is the sent answer.

[0572] The above steps automate the email response process, enabling fast and accurate responses.

[0573] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0574] An embodiment of the present invention relates to a system that automates the process of responding to emails and generates appropriate responses by recognizing user emotions. This system coordinates automatic email extraction, content analysis, information search, response generation, emotion recognition, and confirmation / sending via a user interface. The system of the present invention operates through the cooperation of a server, an emotion engine, and a user terminal.

[0575] Server Operation

[0576] Automatic email extraction

[0577] The server periodically accesses the user's mailbox to detect new emails and import them into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[0578] Content analysis

[0579] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words, and extracts important keywords.

[0580] Information Search

[0581] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[0582] Answer generation

[0583] The server uses a generative AI model to generate an appropriate answer based on the search results. For example, from an email about the progress of a new project, the server might generate an answer such as, "The current project is progressing on schedule and within budget."

[0584] Emotion Engine Operation

[0585] emotion recognition

[0586] The emotion engine analyzes emotions in real time based on user input and reactions, using keyboard input, mouse movements, and even biometric signals.

[0587] Emotion-based response adjustment

[0588] The emotion engine recognizes the user's emotions and adjusts the content of the generated responses based on the results. For example, if the user is feeling stressed, the responses can be tailored to be more kind and comforting.

[0589] Use of past emotion data

[0590] The emotion engine stores the user's past emotional data for a certain period of time and uses it to more accurately recognize emotions and respond, making it possible to respond based on the user's long-term emotional tendencies.

[0591] User terminal operation

[0592] User Interface

[0593] The user terminal presents the answer generated by the server to the user. The user can review the answer and modify it as necessary. For example, the user can modify the answer by adding a specific number. The final answer is then sent.

[0594] Specific examples

[0595] For example, consider the following email sent to a user:

[0596] Please tell me about the progress of your new project.

[0597] Server behavior:

[0598] The server periodically checks the mailbox to detect new mail and retrieves it.

[0599] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[0600] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[0601] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[0602] Emotion Engine in action:

[0603] The emotion engine analyzes the user's keyboard input and mouse movements to recognize when the user is feeling stressed.

[0604] Based on the recognition results, a comforting phrase such as "The current project is progressing very well. Please rest assured." is added to the generated answer.

[0605] User terminal behavior:

[0606] The user terminal displays the generated answer to the user.

[0607] The user checks the answers and makes corrections if necessary.

[0608] After making corrections, the user creates a final response and sends it to the other party.

[0609] This not only automates the process of responding to emails, but also enables appropriate responses that take into consideration the user's feelings, thereby achieving even higher levels of customer satisfaction.

[0610] The processing flow will be explained below.

[0611] Step 1:

[0612] The server periodically accesses the user's mailbox to check for new emails. For example, it accesses the mail server every five minutes to check if new emails have arrived. If there are any new emails, it imports them into the system.

[0613] Step 2:

[0614] The server analyzes the content of the emails it receives. Specifically, it uses natural language processing technology to divide the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords. For example, from an email that asks, "Please tell me about the progress of the new project," it extracts keywords such as "new project" and "progress."

[0615] Step 3:

[0616] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. For example, the server searches the company's intranet using the keywords "new project" and "progress" to retrieve relevant project plans and progress reports.

[0617] Step 4:

[0618] The server compiles the information it has acquired and uses a generative AI model to generate an appropriate response, such as "The current project is progressing on schedule and within budget."

[0619] Step 5:

[0620] The emotion engine analyzes the user's input and other biometric signals (e.g., keyboard typing speed and mouse movements) to recognize the user's emotional state. For example, if the user types quickly, it may determine that the user is stressed.

[0621] Step 6:

[0622] The emotion engine adjusts the generated answers based on the perceived emotion, for example, if it detects that the user is stressed, it will add a comforting element such as "Don't worry, progress on the current project is going very well."

[0623] Step 7:

[0624] The user terminal displays the generated answer received from the server to the user. The user can check the displayed answer and modify the content as necessary. For example, the user can modify the answer by adding specific progress data.

[0625] Step 8:

[0626] After checking and correcting the answer, the user clicks the send button to send it to the other party. At this time, the user's operation is sent to the server, and the final email is sent to the other party, completing the response.

[0627] By using this process flow, the present invention can automate the user's email response process and also realize an appropriate response that takes into consideration the user's feelings.

[0628] Example 2

[0629] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0630] Conventional email response systems require users to manually analyze the content of received emails and create replies, which requires a great deal of time and effort. Furthermore, the quality of responses can be poor because responses cannot take into account the user's feelings. The present invention aims to solve these problems by improving the efficiency of email response processes and providing high-quality responses that take into account the user's feelings.

[0631] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for periodically detecting new e-mails and capturing their contents, means for analyzing the contents of the captured e-mails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using an artificial intelligence model based on the search results, means for recognizing the user's emotions and adjusting the content of the generated answer based on the emotions, and means for displaying the generated answer on the user terminal so that the user can confirm and correct it. This makes it possible to automate the e-mail response process and provide an appropriate response that takes the user's emotions into consideration.

[0632] "Newly arrived email" refers to unread email that has been newly received by the user.

[0633] "Periodic detection" means that e-mails are automatically checked at preset time intervals.

[0634] "Means for retrieving" refers to the function of retrieving emails from an email server and storing them within the system.

[0635] "Content analysis" refers to analyzing the body and subject of an email using natural language processing techniques.

[0636] "Key word extraction" is the process of identifying and extracting specific tokens or phrases from the analyzed content.

[0637] "Internal sources" refer to documents, databases, and past communications records shared within an organization.

[0638] "External sources" refer to publicly available data on the Internet, information provided by other companies or organizations, etc.

[0639] "Searching for related information" means searching for relevant data from predefined information sources based on the extracted keywords.

[0640] "Artificial intelligence model" refers to an intelligent system that uses machine learning algorithms to generate output based on input data.

[0641] "Means for generating a response" refers to the function of constructing an appropriate reply based on the relevant information obtained.

[0642] "Recognizing user emotions" is the process of inferring a user's emotional state from their behavior and biosignals.

[0643] "Adjusting the content of the generated response based on emotion" refers to modifying the content and tone of the generated response according to the recognized emotion.

[0644] "User terminal" refers to a device such as a computer, smartphone, or tablet that is actually operated by a user.

[0645] "Means for checking and correcting" refers to a function that allows the user to check the automatically generated answers and manually correct them if necessary.

[0646] The present invention is a system that integrates an automatic email processing function with a user emotion recognition function. This system operates by linking a server, an emotion engine, and a user terminal.

[0647] Server Operation

[0648] Automatic email extraction

[0649] The server periodically accesses the user's email box to detect new emails, for example by connecting to a mail server such as Gmail or Outlook using the IMAP protocol to retrieve new emails.

[0650] Content analysis

[0651] The server then analyzes the email content using natural language processing techniques, using the Python Natural Language Toolkit (NLTK) library to tokenize the email body and subject line and extract key keywords.

[0652] Information Search

[0653] The server uses the extracted keywords to search for relevant information from internal sources (databases, knowledge bases, intranets) and external sources (the Internet) using SQL queries to the internal database and Google Search APIs.

[0654] Answer generation

[0655] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT-3) based on the search results. The generative AI model receives a prompt such as: "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." An example of a generated answer is, "Current project progress is on schedule and within budget."

[0656] Emotion Engine Operation

[0657] emotion recognition

[0658] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, and even biometric signals. Emotion recognition uses machine learning libraries such as TensorFlow and PyTorch to identify emotions such as stress or joy that the user is feeling.

[0659] Emotion-based response adjustment

[0660] The content of the generated answer is adjusted based on the results of emotion recognition. For example, if the user is feeling stressed, the answer obtained from the generative AI model will include comforting words such as "Please continue to watch over the project with peace of mind."

[0661] Use of past emotion data

[0662] The emotion engine stores the user's past emotional data for a certain period of time and uses this data to more accurately recognize and respond to emotions, enabling advanced responses based on the user's long-term emotional tendencies.

[0663] User terminal operation

[0664] User Interface

[0665] The user's device displays the answer generated by the server to the user. The user can review the answer and make corrections as necessary. For example, a front-end framework such as ReactJS or Vue.js can be used to display the answer in an easy-to-understand format. After the user makes corrections to the answer, such as adding specific numbers, the final answer can be submitted.

[0666] This system automates the email response process and enables appropriate responses that take into consideration the user's feelings, which contributes to increased customer satisfaction and dramatically improves business efficiency.

[0667] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0668] Step 1: Automatic email extraction

[0669] The server periodically accesses the user's mailbox to detect new emails. The server connects to the mail server using the IMAP protocol and retrieves new emails. The input includes the user's mailbox information and access frequency settings. The output is the data of new emails that are imported into the server. Specifically, for example, the server automatically synchronizes with the mail server every morning at 8:00 and detects new emails.

[0670] Step 2: Analyzing the email content

[0671] The server analyzes the content of the newly received emails using natural language processing technology. The input includes the data of the newly received emails. Specifically, it uses Python's NLTK library to tokenize the email body and subject, removes unnecessary words, and extracts important keywords using techniques such as TF-IDF. The output is the extracted keywords.

[0672] Step 3: Information search

[0673] The server searches for related information from internal information sources (databases, knowledge bases, intranets) and external information sources (the Internet) based on the extracted keywords. The input includes the extracted keywords. Specifically, it executes an SQL query to the internal database to retrieve the relevant data. It also retrieves related information from external information sources using the Google Search API. The output is the retrieved related information.

[0674] Step 4: Answer Generation

[0675] The server uses a generative AI model (such as GPT-3) to generate an appropriate answer based on the search results. The input includes the relevant information obtained as search results. An example of a prompt is, "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." Specifically, the prompt and search results are input into the generative AI model to generate an answer. The output is the generated answer.

[0676] Step 5: Emotion Recognition

[0677] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, biometric signals, etc. Inputs include user behavioral data (e.g., typing speed, mouse movements, etc.). Specifically, it uses machine learning models (TensorFlow or PyTorch) to identify the user's emotions. The output is the user's emotional state.

[0678] Step 6: Adjust your responses based on emotion

[0679] The emotion engine adjusts the content of the generated answer based on the analyzed user's emotions. The input includes the generated answer and the user's emotional state. For example, if the user is feeling stressed, the engine adds a phrase such as "Please continue to monitor the project with peace of mind." The output is an answer adjusted to take the user's emotions into account.

[0680] Step 7: Present and send the answer to the user

[0681] The user device presents the generated answer to the user and provides an interface that allows the user to review and modify it. The input includes the adjusted answer. Specific operations include displaying the answer in a user interface using a front-end framework such as ReactJS or Vue.js. The user can review the answer and modify it, such as adding a specific number. The user then creates and submits the final answer. The output is the modified final answer, which is then sent to the other party.

[0682] (Application example 2)

[0683] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0684] In conventional email response systems, the process of analyzing email content and generating appropriate responses required a lot of time and effort. Furthermore, if store clerks were unable to respond to customer inquiries immediately, customer satisfaction could decline. Furthermore, because responses were not flexibly adjusted based on customer emotions, there were limitations to how sensitive the system could be to customer emotions. Therefore, there is a need for a system that automates the processing of emails and inquiries and generates appropriate responses based on emotion recognition.

[0685] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer based on the search results, means for displaying the generated answer on a user terminal so that the user can confirm and modify it, means for adjusting the generated answer based on emotion recognition, and means for accepting customer inquiries in-store using QR codes. This makes it possible to automate the processing of emails and inquiries and provide flexible responses that take emotions into consideration.

[0686] "New email" is unread email that has just arrived in your inbox.

[0687] The "means for acquiring content" is a device or program that has the function of acquiring new e-mails into the system.

[0688] "Analyzing the content" means analyzing the text data of the imported email using natural language processing technology.

[0689] The "means for extracting important keywords" is a device or program that identifies and extracts keywords with specific meanings from the analyzed text data.

[0690] "Internal information sources" are information resources such as databases and knowledge bases within a company.

[0691] "External sources" are information resources outside the company, such as information on the Internet or other public databases.

[0692] The "means for searching related information" is a device or program that searches for relevant data from internal and external sources based on the extracted keywords.

[0693] The "means for generating an appropriate answer" is a device or program that generates an appropriate response to a query based on the search results.

[0694] A "user terminal" is a computer device or mobile device that can be operated by a user.

[0695] The "means for confirming and correcting" is a device or program that has the function of allowing the user to confirm the generated answer and make corrections as necessary.

[0696] The "means for adjusting based on emotion recognition" is a device or program that analyzes the user's input and response, recognizes the emotion, and modifies the response appropriately.

[0697] "Means for accepting inquiries in-store using QR codes" refers to a device or program that has the function of scanning QR codes and accepting inquiries electronically.

[0698] A system embodying the present invention automates email correspondence and can generate appropriate responses by recognizing the user's emotions. An embodiment of this system will now be described in detail.

[0699] System configuration

[0700] The system consists of a server, a user terminal, an emotion engine, and a QR code scanning device. The server has the following main functions:

[0701] Ability to periodically detect new emails and retrieve their contents

[0702] Ability to analyze email content and extract important keywords

[0703] Ability to search for related information from internal and external sources based on extracted keywords

[0704] Generate relevant answers based on search results

[0705] A function that displays the generated answers on the user's device so that the user can check and correct them.

[0706] The ability to tailor generated answers based on emotion recognition

[0707] A function to accept customer inquiries in-store using QR codes

[0708] Hardware and software used

[0709] Hardware: Servers, smartphones, smart glasses, QR code scanners

[0710] software:

[0711] Natural language processing: SpaCy(ja_core_news_sm)

[0712] Generative AI model: GPT-2 (transformers)

[0713] Email communication: smtplib, imaplib

[0714] Emotion recognition: keyboard input, mouse movement analysis software, and possibly biometric sensors

[0715] Operation process

[0716] Server Operation

[0717] The server periodically connects to the mail server to detect new emails. The detected emails are imported and their contents are analyzed to extract important keywords. Based on the extracted keywords, relevant information is searched from internal sources (past emails, internal intranet, internal knowledge base) and external sources (the Internet). Based on the search results, an appropriate answer is generated using a generative AI model (GPT-2). The following prompt sentences are used in this generation process:

[0718] Example prompt:

[0719] "Please answer the following questions: Project Status\n\nReply: "

[0720] Emotion Engine Operation

[0721] The emotion engine recognizes emotions by analyzing user input and reactions. For example, it can assess emotions in real time using keyboard typing speed and strength, mouse movements, and biometric sensors. The emotion engine adjusts the content of the generated response based on the analysis results. For example, if the user is feeling stressed, it will add comforting words such as "Don't worry" to the generated response.

[0722] User terminal operation

[0723] The generated answer is displayed on the user's device, where the user can review it and make any necessary corrections. The final answer is sent at the user's discretion. In-store inquiries are also accepted through a QR code scanner, which is also captured in the system and goes through the same process.

[0724] Specific examples

[0725] For example, if a user receives an email like this:

[0726] Please tell me about the progress of your new project.

[0727] The server detects this email, analyzes its contents, and extracts the keywords "project" and "progress." Based on this, a generative AI model is used to generate the answer, "The current project is progressing smoothly." If the emotion engine recognizes the user's stress, it further adjusts the answer to take their emotions into consideration, such as, "The current project is progressing smoothly. Please rest assured."

[0728] This allows for efficient handling of emails and in-store inquiries, and enables flexible, emotional responses.

[0729] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0730] Step 1:

[0731] The server periodically detects new emails and retrieves their contents. The server accesses the user's mailbox and retrieves unread emails. The input is the new email data from the mail server, and the output is the retrieved email contents.

[0732] Step 2:

[0733] The server analyzes the content of the captured email and extracts important keywords. The server uses natural language processing software (e.g., SpaCy) to split the text into tokens, filter out meaningless words, and extract important keywords. The input is the email content, and the output is the extracted keywords.

[0734] Step 3:

[0735] The server searches for related information from internal and external sources based on the extracted keywords. The server sends search queries to internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet) to collect related information. The input is the extracted keywords, and the output is the search results.

[0736] Step 4:

[0737] The server generates an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-2), the server inputs keywords and search results into a prepared prompt sentence to generate a natural language answer. The input is the search results, and the output is the generated answer. An example of a prompt sentence is as follows:

[0738] "Please answer the following questions: Project Status\n\nReply: "

[0739] Step 5:

[0740] The emotion engine analyzes user input and reactions to recognize emotions. The emotion engine analyzes keyboard input speed, keystroke strength, mouse movement, and, in some cases, data from biometric sensors. The input is user input and reaction data, and the output is emotional data that indicates the user's state.

[0741] Step 6:

[0742] The emotion engine adjusts the generated answer based on the recognized emotion data. For example, if it recognizes that the user is stressed, it adds comforting words such as "Don't worry" to the answer. The input is the generated answer and emotion data, and the output is the adjusted answer.

[0743] Step 7:

[0744] The user terminal displays the adjusted answer. The user reviews the adjusted answer, makes corrections if necessary, and finally submits the answer. The input is the adjusted answer, and the output is the final answer that has been reviewed and corrected by the user.

[0745] Step 8:

[0746] The server accepts inquiries from customers using QR codes in the store. The inquiry content obtained by scanning the QR code is imported into the system and goes through a process similar to that for email. The input is the inquiry content obtained from the QR code, and the output is the inquiry data imported into the system.

[0747] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0748] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0749] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0750] [Third embodiment]

[0751] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0752] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0753] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0754] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0755] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0756] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0757] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0758] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0759] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0760] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0761] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0762] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0763] An embodiment of the present invention relates to a system that automates the process of responding to emails. This system coordinates automatic email extraction, content analysis, information search, response generation, and confirmation / sending via a user interface. The system of the present invention operates primarily through the coordination of a server and a user terminal.

[0764] Server Operation

[0765] Automatic email extraction

[0766] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[0767] Content analysis

[0768] The server uses natural language processing technology to analyze the content of the email and extracts important keywords to understand the intent and context of the email. For example, the server can extract the keywords "project" and "progress" from an email with a subject such as "Regarding the progress of a new project."

[0769] Information Search

[0770] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[0771] Answer generation

[0772] The server uses a generative AI model to generate appropriate answers based on the search results. For example, based on the keywords above, it might generate an answer such as, "The current project progress is on schedule and within budget."

[0773] User terminal operation

[0774] User Interface

[0775] The user terminal presents the answer generated by the server to the user. The user can check this answer and modify it as necessary. For example, the user may modify the displayed answer by adding more specific progress information. The user then submits the final answer.

[0776] Specific examples

[0777] For example, consider the following email sent to a user:

[0778] Please tell me about the progress of your new project.

[0779] Server behavior:

[0780] The server periodically checks the mailbox to detect new mail and retrieves it.

[0781] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[0782] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[0783] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[0784] User terminal behavior:

[0785] The user terminal displays the generated answer to the user.

[0786] The user checks the answers and makes corrections, such as adding "more specific progress information."

[0787] After making corrections, the user creates a final response and sends it to the other party.

[0788] In this way, the present invention can significantly automate the process of responding to e-mails, reducing the workload of users and improving productivity.

[0789] The processing flow will be explained below.

[0790] Step 1:

[0791] The server periodically accesses the user's mailbox to check for new emails, and if there are any new emails, it imports them into the system.

[0792] Step 2:

[0793] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords.

[0794] Step 3:

[0795] The server searches for related information from internal and external sources based on the extracted keywords, utilizing past email history, the company intranet, the company knowledge base, the Internet, and other sources.

[0796] Step 4:

[0797] The server uses a generative AI model to generate appropriate answers based on the search results, and outputs the answers as syntactically and semantically consistent sentences using natural language processing technology.

[0798] Step 5:

[0799] The user terminal displays the generated answer sent from the server to the user, who can then check the displayed answer and modify it as necessary.

[0800] Step 6:

[0801] After the user has confirmed and corrected the answer, he or she clicks the send button to send the final answer. At this time, the user's operation details are sent to the server and then to the other party as an e-mail.

[0802] In this way, the process of responding to emails is automated, significantly reducing the amount of work required by the user.

[0803] Example 1

[0804] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0805] The traditional email response process requires a lot of manual work, which is time-consuming and labor-intensive, reducing the user's work efficiency. Furthermore, in order to understand the content of an email and respond appropriately, the user must analyze each piece of content and generate a response, which places a heavy burden on the user. Therefore, there is a need for a system that automates the email response process, reduces the user's workload, and improves productivity.

[0806] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0807] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords using natural language processing technology, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating appropriate answers based on the search results using a generative AI model, means for displaying the generated answers on a user terminal so that the user can confirm and modify them, and means for sending the final answers. This allows the email response process to be largely automated, reducing the user's workload and improving productivity.

[0808] "New email" is a new electronic message that has arrived in a user's mailbox.

[0809] "Periodic detection" means that new e-mails are automatically checked according to a preset time interval.

[0810] The "means for capturing" is a function for storing the contents of detected new e-mails in the system.

[0811] "Analyzing the content" refers to analyzing the text information of the acquired email using natural language processing technology.

[0812] "Natural language processing technology" is a general term for technologies that enable computers to understand, interpret, and generate human language.

[0813] "Extracting important keywords" means identifying key words and phrases from the content of an email that are necessary to understand its intent and context.

[0814] "Internal sources" refer to information repositories within an organization, including archived emails, internal intranets, and internal knowledge bases.

[0815] "External information sources" refer to repositories of information that can be obtained from outside the organization, primarily including the Internet.

[0816] "Searching for related information" refers to using the extracted keywords to research and obtain related information from internal and external sources.

[0817] A "generative AI model" refers to an algorithm or system that uses artificial intelligence technology to generate sentences in natural language.

[0818] "Means for generating answers" refers to a function that uses a generative AI model to create appropriate answers based on search results.

[0819] "User terminal" refers to equipment such as a computer or mobile device used by a user.

[0820] The "means for checking and correcting" is a function that displays the generated answer to the user, and allows the user to check the content and edit it as necessary.

[0821] The "means for sending the final answer" is a function for the user to send the answer that has been confirmed and corrected by the user to the other party as an email.

[0822] An embodiment of the present invention relates to a system for automating the process of responding to emails. This system operates mainly through cooperation between a server and a user terminal.

[0823] Server Operation

[0824] Automatic email extraction

[0825] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. Specifically, it periodically connects to the mail server using a mail protocol such as IMAP or POP3 to retrieve new emails. For example, the server connects to the mail server once every hour to check for new emails.

[0826] Content analysis

[0827] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy or NLTK). During this process, the email text is tokenized, tagged with parts of speech, and important keywords are extracted. For example, the keywords "project" and "progress" are extracted from an email that asks, "Please tell us about the progress of your new project."

[0828] Information Search

[0829] The server uses the extracted keywords to search for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet searches). Specifically, it executes SQL queries using the keywords to retrieve information from the internal database. It also uses external sources such as the Google Search API.

[0830] Answer generation

[0831] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT series) based on the search results. The search results and prompt are input into the generative AI model, which generates an appropriate answer in natural language. For example, the prompt "Project progress is progressing smoothly and within budget" is input.

[0832] User terminal operation

[0833] User Interface

[0834] The user terminal displays the generated answer to the user. The user can review the displayed answer and edit it if necessary. The user interface provides text boxes and editing tools to allow the user to easily modify the answer. For example, the user can enter additional information such as "The next milestone is the start of the testing phase."

[0835] Last sent

[0836] Once the user has finished making corrections, they can finalize the answer and send it to the other party as an email. The email is sent via the user's mail server using the SMTP protocol, making it easy for the user to send the final answer.

[0837] Specific examples

[0838] Consider the following example:

[0839] Please tell me about the progress of your new project.

[0840] Server behavior:

[0841] 1. The server connects to the user's mail server once every hour and checks for new email.

[0842] 2. Detect new emails and save their contents to a local database.

[0843] 3. Use Spacy to tokenize the text and extract the keywords "project" and "progress."

[0844] 4. Use SQL queries to retrieve relevant information from an internal knowledge base and use the Google Search API to retrieve information from the internet.

[0845] 5. The prompt sentence "Project progress is progressing smoothly and within budget" is input into the generative AI model, which generates the answer "Current project progress is progressing as planned and within budget."

[0846] User terminal behavior:

[0847] 1. The user's device displays the generated answer, and the user confirms the content.

[0848] 2. The user corrects their answer by adding additional information: "The next milestone is the start of the testing phase."

[0849] 3. The user confirms the revised answer and clicks the submit button.

[0850] 4. The final response is sent via the mail server using the SMTP protocol.

[0851] Prompt Sentence Examples

[0852] You receive an email asking, "How's your new project going?" Please generate an answer based on the following information:

[0853] Project name: AI system development project

[0854] Progress: 50% complete

[0855] Next Milestone: Testing Phase Begins

[0856] Budget status: On track and within budget

[0857] Using this prompt, the appropriate generative AI model will respond, generating the answer, "The AI ​​system development project is now 50% complete and is progressing towards the next milestone, the start of the testing phase. We are on track and within budget."

[0858] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0859] Step 1:

[0860] Automatic email extraction

[0861] The server connects to the user's mail server (e.g., using IMAP or POP3 protocol) at set intervals to detect and retrieve new emails. The input is the connection information to the user's mail server, and the output is the detected new email data. Specifically, the server connects to the mail server once every hour to check for new emails and saves the detected new emails in a local database.

[0862] Step 2:

[0863] Content analysis

[0864] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy, NLTK). The input is the text data of the email, and the output is important keywords extracted through analysis. The server tokenizes the text and tags it with parts of speech. For example, the server extracts the keywords "project" and "progress" from an email that says, "Please tell us about the progress of the new project."

[0865] Step 3:

[0866] Information Search

[0867] Based on the extracted keywords, the server searches for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet search). The input is the extracted keywords and the output is related information. The server executes SQL queries to retrieve related information from the internal database and uses the Google Search API to retrieve related information from the Internet.

[0868] Step 4:

[0869] Answer generation

[0870] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT series) based on the search results. The input is the search results and prompt, and the output is the generated answer. The server generates a prompt, "Project progress is progressing smoothly and within budget," and inputs it into the generative AI model. This generates an answer, "Current project progress is proceeding as planned and within budget."

[0871] Step 5:

[0872] User verification and correction

[0873] The user terminal displays the generated answer to the user. The user checks this answer and makes corrections as necessary. The input is the generated answer, and the output is the corrected answer. The user interface has text boxes and editing tools, allowing the user to enter additional information, such as "The next milestone is the start of the test phase," and correct the answer.

[0874] Step 6:

[0875] Final submission of response

[0876] The user terminal confirms the final answer after the user has completed the corrections and sends it as an email. The input is the corrected answer text, and the output is the sent email. Specifically, the user confirms the corrected answer and clicks the send button. The user terminal uses the SMTP protocol to connect to the mail server and send the final answer.

[0877] (Application example 1)

[0878] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0879] Conventional email response systems require users to manually check emails, analyze their content, and create responses, resulting in significant time and effort. Furthermore, in today's world, where rapid responses to inquiries are required, there is a growing need for systems that can efficiently and appropriately respond. In particular, content distribution services require rapid and accurate responses because of the wide variety of user inquiries. The present invention aims to solve these problems and significantly automate the email response process.

[0880] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0881] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using a generation system based on the search results, means for displaying the generated answer on an output device so that a user can confirm and modify it by operating the output device, means for sending the generated answer as a reply to an inquiry, means for calling a generative AI model using a prompt sentence and generating an answer, and document generation means for automatically generating an appropriate answer to an inquiry. This allows users to efficiently automate the process from email content analysis to answer generation, enabling quick and accurate responses.

[0882] "New email" refers to new electronic mail that arrives in your inbox periodically.

[0883] "Content capture means" refers to the process or technology used to capture email content into the system.

[0884] "Means of content analysis" refers to the process of analyzing the content of captured emails using natural language processing technology to understand their intent and context.

[0885] "Important keywords" refer to words extracted from the content of emails that are important for extracting specific information.

[0886] "Internal sources" refers to information repositories maintained within internal systems, historical email history, intranets, and knowledge bases.

[0887] "External information sources" refers to information resources outside the company, such as the Internet.

[0888] "Means for searching related information" refers to the process of finding related information from internal and external sources based on the extracted keywords.

[0889] "Means for generating appropriate answers" refers to the process of using AI technology to create appropriate answers based on search results.

[0890] "Output device" refers to an appliance or device for displaying generated answers to a user.

[0891] "Means of invoking a generative AI model using a prompt sentence to generate an answer" refers to the process of running a generative AI model using a specified prompt sentence to construct an appropriate answer.

[0892] "Document generation means for automatically generating appropriate responses to inquiries" refers to a process that uses AI technology to automatically generate appropriate documents in response to user inquiries.

[0893] "Means for sending a reply to a query" refers to the technology or process for sending the generated answer back to the user.

[0894] This invention relates to a system for automating the process of responding to emails. This system automates everything from detecting new emails to generating and sending responses, thereby reducing the workload of users.

[0895] System Program Overview

[0896] The system mainly operates on a server and a user device. The server periodically detects new emails and analyzes their contents. Based on the analyzed important keywords, it searches for related information from internal and external sources. It then uses a generative AI model to generate an appropriate answer, which is displayed on the user's device. The user can then confirm and modify the answer, and send the final answer to the other party.

[0897] Hardware and software configuration

[0898] server:

[0899] The server uses the following software and libraries:

[0900] IMAP4 library: Used to retrieve new emails.

[0901] Python email module: Used to analyze the content of retrieved emails.

[0902] Generative AI models (OpenAI's GPT-3 / 4): Use natural language processing techniques to generate answers.

[0903] User device:

[0904] The user terminal uses the following devices and applications:

[0905] Output device (e.g. smartphone, tablet, PC): Displays the answers from the server and allows the user to check and correct them.

[0906] Data processing and calculation

[0907] The server processes data in the following steps: First, it periodically detects new emails using the IMAP4 library and analyzes their contents using the email module. This analysis extracts important keywords. Next, based on the extracted keywords, it searches for related information from internal sources (past emails, intranets, knowledge bases) and external sources (the Internet). It then uses a generative AI model to generate appropriate answers based on the search results. The generated answers are sent to the user's device, where they can be reviewed and revised by the user.

[0908] Specific examples

[0909] For example, consider a case where a user makes a query such as "What are the latest movie listings?"

[0910] The server detects new emails and analyzes the query content. Based on the extracted keywords "latest" and "movie list," it searches for related information from internal and external sources. Based on the results, it queries the generative AI model using the following prompt:

[0911] Example prompt sentence:

[0912] User Question: What is the latest movie list?

[0913] Generate the appropriate answer.

[0914] The generative AI model generates an appropriate answer based on this prompt, for example, "The latest movie list is as follows: 1. Inception 2. Avatar 2 3. Titanic."

[0915] The generated answer is displayed on the user's terminal, where the user can check the content and make corrections as necessary, after which the user can submit the final answer.

[0916] This invention enables users to respond to inquiries quickly and accurately, thereby improving work efficiency.

[0917] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0918] Step 1:

[0919] The server periodically checks for new emails. Using the IMAP4 library, it connects to the specified mail server and retrieves unread emails. At this point, the input is the mail server authentication information and the user's mailbox, and the output is a list of unread emails.

[0920] Step 2:

[0921] The server analyzes the contents of the retrieved email. It uses the Python email module to analyze the email content (body, subject, sender information, etc.). As a result of the analysis, it extracts important keywords from the subject and body of the email. At this point, the input is the content of the unread email, and the output is the extracted important keywords.

[0922] Step 3:

[0923] The server searches for related information based on the extracted keywords. It searches for related information from internal sources (past email history, intranet, knowledge base) and external sources (Internet). For example, it searches for related information using keywords such as "latest" and "movie list." At this point, the input is the extracted keywords, and the output is a list of related information.

[0924] Step 4:

[0925] The server uses a generation system to generate an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-3 or GPT-4), it inputs a prompt such as "User question: What are the latest movie listings? Please generate an appropriate answer." and receives the generated answer. At this point, the input is the prompt and related information, and the output is the generated answer.

[0926] Step 5:

[0927] The server sends the generated answer to the user terminal. The generated answer is displayed on the user terminal via the transmission system. At this point, the input is the generated answer, and the output is the display result on the user terminal.

[0928] Step 6:

[0929] The user checks the displayed answer and makes corrections if necessary. The user checks the answer using a device such as a smartphone or PC and makes corrections if necessary. The input at this point is the displayed answer and the user's corrections, and the output is the final answer after corrections.

[0930] Step 7:

[0931] The user sends the final answer to the other party. The user sends the revised answer from the user terminal to the system, and the system returns it to the other party. At this point, the input is the revised final answer, and the output is the sent answer.

[0932] The above steps automate the email response process, enabling fast and accurate responses.

[0933] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0934] An embodiment of the present invention relates to a system that automates the process of responding to emails and generates appropriate responses by recognizing user emotions. This system coordinates automatic email extraction, content analysis, information search, response generation, emotion recognition, and confirmation / sending via a user interface. The system of the present invention operates through the cooperation of a server, an emotion engine, and a user terminal.

[0935] Server Operation

[0936] Automatic email extraction

[0937] The server periodically accesses the user's mailbox to detect new emails and import them into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[0938] Content analysis

[0939] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words, and extracts important keywords.

[0940] Information Search

[0941] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[0942] Answer generation

[0943] The server uses a generative AI model to generate an appropriate answer based on the search results. For example, from an email about the progress of a new project, the server might generate an answer such as, "The current project is progressing on schedule and within budget."

[0944] Emotion Engine Operation

[0945] emotion recognition

[0946] The emotion engine analyzes emotions in real time based on user input and reactions, using keyboard input, mouse movements, and even biometric signals.

[0947] Emotion-based response adjustment

[0948] The emotion engine recognizes the user's emotions and adjusts the content of the generated responses based on the results. For example, if the user is feeling stressed, the responses can be tailored to be more kind and comforting.

[0949] Use of past emotion data

[0950] The emotion engine stores the user's past emotional data for a certain period of time and uses it to more accurately recognize emotions and respond, making it possible to respond based on the user's long-term emotional tendencies.

[0951] User terminal operation

[0952] User Interface

[0953] The user terminal presents the answer generated by the server to the user. The user can review the answer and modify it as necessary. For example, the user can modify the answer by adding a specific number. The final answer is then sent.

[0954] Specific examples

[0955] For example, consider the following email sent to a user:

[0956] Please tell me about the progress of your new project.

[0957] Server behavior:

[0958] The server periodically checks the mailbox to detect new mail and retrieves it.

[0959] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[0960] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[0961] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[0962] Emotion Engine in action:

[0963] The emotion engine analyzes the user's keyboard input and mouse movements to recognize when the user is feeling stressed.

[0964] Based on the recognition results, a comforting phrase such as "The current project is progressing very well. Please rest assured." is added to the generated answer.

[0965] User terminal behavior:

[0966] The user terminal displays the generated answer to the user.

[0967] The user checks the answers and makes corrections if necessary.

[0968] After making corrections, the user creates a final response and sends it to the other party.

[0969] This not only automates the process of responding to emails, but also enables appropriate responses that take into consideration the user's feelings, thereby achieving even higher levels of customer satisfaction.

[0970] The processing flow will be explained below.

[0971] Step 1:

[0972] The server periodically accesses the user's mailbox to check for new emails. For example, it accesses the mail server every five minutes to check if new emails have arrived. If there are any new emails, it imports them into the system.

[0973] Step 2:

[0974] The server analyzes the content of the emails it receives. Specifically, it uses natural language processing technology to divide the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords. For example, from an email that asks, "Please tell me about the progress of the new project," it extracts keywords such as "new project" and "progress."

[0975] Step 3:

[0976] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. For example, the server searches the company's intranet using the keywords "new project" and "progress" to retrieve relevant project plans and progress reports.

[0977] Step 4:

[0978] The server compiles the information it has acquired and uses a generative AI model to generate an appropriate response, such as "The current project is progressing on schedule and within budget."

[0979] Step 5:

[0980] The emotion engine analyzes the user's input and other biometric signals (e.g., keyboard typing speed and mouse movements) to recognize the user's emotional state. For example, if the user types quickly, it may determine that the user is stressed.

[0981] Step 6:

[0982] The emotion engine adjusts the generated answers based on the perceived emotion, for example, if it detects that the user is stressed, it will add a comforting element such as "Don't worry, progress on the current project is going very well."

[0983] Step 7:

[0984] The user terminal displays the generated answer received from the server to the user. The user can check the displayed answer and modify the content as necessary. For example, the user can modify the answer by adding specific progress data.

[0985] Step 8:

[0986] After checking and correcting the answer, the user clicks the send button to send it to the other party. At this time, the user's operation is sent to the server, and the final email is sent to the other party, completing the response.

[0987] By using this process flow, the present invention can automate the user's email response process and also realize an appropriate response that takes into consideration the user's feelings.

[0988] Example 2

[0989] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0990] Conventional email response systems require users to manually analyze the content of received emails and create replies, which requires a great deal of time and effort. Furthermore, the quality of responses can be poor because responses cannot take into account the user's feelings. The present invention aims to solve these problems by improving the efficiency of email response processes and providing high-quality responses that take into account the user's feelings.

[0991] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for periodically detecting new e-mails and capturing their contents, means for analyzing the contents of the captured e-mails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using an artificial intelligence model based on the search results, means for recognizing the user's emotions and adjusting the content of the generated answer based on the emotions, and means for displaying the generated answer on the user terminal so that the user can confirm and correct it. This makes it possible to automate the e-mail response process and provide an appropriate response that takes the user's emotions into consideration.

[0992] "Newly arrived email" refers to unread email that has been newly received by the user.

[0993] "Periodic detection" means that e-mails are automatically checked at preset time intervals.

[0994] "Means for retrieving" refers to the function of retrieving emails from an email server and storing them within the system.

[0995] "Content analysis" refers to analyzing the body and subject of an email using natural language processing techniques.

[0996] "Key word extraction" is the process of identifying and extracting specific tokens or phrases from the analyzed content.

[0997] "Internal sources" refer to documents, databases, and past communications records shared within an organization.

[0998] "External sources" refer to publicly available data on the Internet, information provided by other companies or organizations, etc.

[0999] "Searching for related information" means searching for relevant data from predefined information sources based on the extracted keywords.

[1000] "Artificial intelligence model" refers to an intelligent system that uses machine learning algorithms to generate output based on input data.

[1001] "Means for generating a response" refers to the function of constructing an appropriate reply based on the relevant information obtained.

[1002] "Recognizing user emotions" is the process of inferring a user's emotional state from their behavior and biosignals.

[1003] "Adjusting the content of the generated response based on emotion" refers to modifying the content and tone of the generated response according to the recognized emotion.

[1004] "User terminal" refers to a device such as a computer, smartphone, or tablet that is actually operated by a user.

[1005] "Means for checking and correcting" refers to a function that allows the user to check the automatically generated answers and manually correct them if necessary.

[1006] The present invention is a system that integrates an automatic email processing function with a user emotion recognition function. This system operates by linking a server, an emotion engine, and a user terminal.

[1007] Server Operation

[1008] Automatic email extraction

[1009] The server periodically accesses the user's email box to detect new emails, for example by connecting to a mail server such as Gmail or Outlook using the IMAP protocol to retrieve new emails.

[1010] Content analysis

[1011] The server then analyzes the email content using natural language processing techniques, using the Python Natural Language Toolkit (NLTK) library to tokenize the email body and subject line and extract key keywords.

[1012] Information Search

[1013] The server uses the extracted keywords to search for relevant information from internal sources (databases, knowledge bases, intranets) and external sources (the Internet) using SQL queries to the internal database and Google Search APIs.

[1014] Answer generation

[1015] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT-3) based on the search results. The generative AI model receives a prompt such as: "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." An example of a generated answer is, "Current project progress is on schedule and within budget."

[1016] Emotion Engine Operation

[1017] emotion recognition

[1018] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, and even biometric signals. Emotion recognition uses machine learning libraries such as TensorFlow and PyTorch to identify emotions such as stress or joy that the user is feeling.

[1019] Emotion-based response adjustment

[1020] The content of the generated answer is adjusted based on the results of emotion recognition. For example, if the user is feeling stressed, the answer obtained from the generative AI model will include comforting words such as "Please continue to watch over the project with peace of mind."

[1021] Use of past emotion data

[1022] The emotion engine stores the user's past emotional data for a certain period of time and uses this data to more accurately recognize and respond to emotions, enabling advanced responses based on the user's long-term emotional tendencies.

[1023] User terminal operation

[1024] User Interface

[1025] The user's device displays the answer generated by the server to the user. The user can review the answer and make corrections as necessary. For example, a front-end framework such as ReactJS or Vue.js can be used to display the answer in an easy-to-understand format. After the user makes corrections to the answer, such as adding specific numbers, the final answer can be submitted.

[1026] This system automates the email response process and enables appropriate responses that take into consideration the user's feelings, which contributes to increased customer satisfaction and dramatically improves business efficiency.

[1027] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1028] Step 1: Automatic email extraction

[1029] The server periodically accesses the user's mailbox to detect new emails. The server connects to the mail server using the IMAP protocol and retrieves new emails. The input includes the user's mailbox information and access frequency settings. The output is the data of new emails that are imported into the server. Specifically, for example, the server automatically synchronizes with the mail server every morning at 8:00 and detects new emails.

[1030] Step 2: Analyzing the email content

[1031] The server analyzes the content of the newly received emails using natural language processing technology. The input includes the data of the newly received emails. Specifically, it uses Python's NLTK library to tokenize the email body and subject, removes unnecessary words, and extracts important keywords using techniques such as TF-IDF. The output is the extracted keywords.

[1032] Step 3: Information search

[1033] The server searches for related information from internal information sources (databases, knowledge bases, intranets) and external information sources (the Internet) based on the extracted keywords. The input includes the extracted keywords. Specifically, it executes an SQL query to the internal database to retrieve the relevant data. It also retrieves related information from external information sources using the Google Search API. The output is the retrieved related information.

[1034] Step 4: Answer Generation

[1035] The server uses a generative AI model (such as GPT-3) to generate an appropriate answer based on the search results. The input includes the relevant information obtained as search results. An example of a prompt is, "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." Specifically, the prompt and search results are input into the generative AI model to generate an answer. The output is the generated answer.

[1036] Step 5: Emotion Recognition

[1037] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, biometric signals, etc. Inputs include user behavioral data (e.g., typing speed, mouse movements, etc.). Specifically, it uses machine learning models (TensorFlow or PyTorch) to identify the user's emotions. The output is the user's emotional state.

[1038] Step 6: Adjust your responses based on emotion

[1039] The emotion engine adjusts the content of the generated answer based on the analyzed user's emotions. The input includes the generated answer and the user's emotional state. For example, if the user is feeling stressed, the engine adds a phrase such as "Please continue to monitor the project with peace of mind." The output is an answer adjusted to take the user's emotions into account.

[1040] Step 7: Present and send the answer to the user

[1041] The user device presents the generated answer to the user and provides an interface that allows the user to review and modify it. The input includes the adjusted answer. Specific operations include displaying the answer in a user interface using a front-end framework such as ReactJS or Vue.js. The user can review the answer and modify it, such as adding a specific number. The user then creates and submits the final answer. The output is the modified final answer, which is then sent to the other party.

[1042] (Application example 2)

[1043] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1044] In conventional email response systems, the process of analyzing email content and generating appropriate responses required a lot of time and effort. Furthermore, if store clerks were unable to respond to customer inquiries immediately, customer satisfaction could decline. Furthermore, because responses were not flexibly adjusted based on customer emotions, there were limitations to how sensitive the system could be to customer emotions. Therefore, there is a need for a system that automates the processing of emails and inquiries and generates appropriate responses based on emotion recognition.

[1045] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer based on the search results, means for displaying the generated answer on a user terminal so that the user can confirm and modify it, means for adjusting the generated answer based on emotion recognition, and means for accepting customer inquiries in-store using QR codes. This makes it possible to automate the processing of emails and inquiries and provide flexible responses that take emotions into consideration.

[1046] "New email" is unread email that has just arrived in your inbox.

[1047] The "means for acquiring content" is a device or program that has the function of acquiring new e-mails into the system.

[1048] "Analyzing the content" means analyzing the text data of the imported email using natural language processing technology.

[1049] The "means for extracting important keywords" is a device or program that identifies and extracts keywords with specific meanings from the analyzed text data.

[1050] "Internal information sources" are information resources such as databases and knowledge bases within a company.

[1051] "External sources" are information resources outside the company, such as information on the Internet or other public databases.

[1052] The "means for searching related information" is a device or program that searches for relevant data from internal and external sources based on the extracted keywords.

[1053] The "means for generating an appropriate answer" is a device or program that generates an appropriate response to a query based on the search results.

[1054] A "user terminal" is a computer device or mobile device that can be operated by a user.

[1055] The "means for confirming and correcting" is a device or program that has the function of allowing the user to confirm the generated answer and make corrections as necessary.

[1056] The "means for adjusting based on emotion recognition" is a device or program that analyzes the user's input and response, recognizes the emotion, and modifies the response appropriately.

[1057] "Means for accepting inquiries in-store using QR codes" refers to a device or program that has the function of scanning QR codes and accepting inquiries electronically.

[1058] A system embodying the present invention automates email correspondence and can generate appropriate responses by recognizing the user's emotions. An embodiment of this system will now be described in detail.

[1059] System configuration

[1060] The system consists of a server, a user terminal, an emotion engine, and a QR code scanning device. The server has the following main functions:

[1061] Ability to periodically detect new emails and retrieve their contents

[1062] Ability to analyze email content and extract important keywords

[1063] Ability to search for related information from internal and external sources based on extracted keywords

[1064] Generate relevant answers based on search results

[1065] A function that displays the generated answers on the user's device so that the user can check and correct them.

[1066] The ability to tailor generated answers based on emotion recognition

[1067] A function to accept customer inquiries in-store using QR codes

[1068] Hardware and software used

[1069] Hardware: Servers, smartphones, smart glasses, QR code scanners

[1070] software:

[1071] Natural language processing: SpaCy(ja_core_news_sm)

[1072] Generative AI model: GPT-2 (transformers)

[1073] Email communication: smtplib, imaplib

[1074] Emotion recognition: keyboard input, mouse movement analysis software, and possibly biometric sensors

[1075] Operation process

[1076] Server Operation

[1077] The server periodically connects to the mail server to detect new emails. The detected emails are imported and their contents are analyzed to extract important keywords. Based on the extracted keywords, relevant information is searched from internal sources (past emails, internal intranet, internal knowledge base) and external sources (the Internet). Based on the search results, an appropriate answer is generated using a generative AI model (GPT-2). The following prompt sentences are used in this generation process:

[1078] Example prompt:

[1079] "Please answer the following questions: Project Status\n\nReply: "

[1080] Emotion Engine Operation

[1081] The emotion engine recognizes emotions by analyzing user input and reactions. For example, it can assess emotions in real time using keyboard typing speed and strength, mouse movements, and biometric sensors. The emotion engine adjusts the content of the generated response based on the analysis results. For example, if the user is feeling stressed, it will add comforting words such as "Don't worry" to the generated response.

[1082] User terminal operation

[1083] The generated answer is displayed on the user's device, where the user can review it and make any necessary corrections. The final answer is sent at the user's discretion. In-store inquiries are also accepted through a QR code scanner, which is also captured in the system and goes through the same process.

[1084] Specific examples

[1085] For example, if a user receives an email like this:

[1086] Please tell me about the progress of your new project.

[1087] The server detects this email, analyzes its contents, and extracts the keywords "project" and "progress." Based on this, a generative AI model is used to generate the answer, "The current project is progressing smoothly." If the emotion engine recognizes the user's stress, it further adjusts the answer to take their emotions into consideration, such as, "The current project is progressing smoothly. Please rest assured."

[1088] This allows for efficient handling of emails and in-store inquiries, and enables flexible, emotional responses.

[1089] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1090] Step 1:

[1091] The server periodically detects new emails and retrieves their contents. The server accesses the user's mailbox and retrieves unread emails. The input is the new email data from the mail server, and the output is the retrieved email contents.

[1092] Step 2:

[1093] The server analyzes the content of the captured email and extracts important keywords. The server uses natural language processing software (e.g., SpaCy) to split the text into tokens, filter out meaningless words, and extract important keywords. The input is the email content, and the output is the extracted keywords.

[1094] Step 3:

[1095] The server searches for related information from internal and external sources based on the extracted keywords. The server sends search queries to internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet) to collect related information. The input is the extracted keywords, and the output is the search results.

[1096] Step 4:

[1097] The server generates an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-2), the server inputs keywords and search results into a prepared prompt sentence to generate a natural language answer. The input is the search results, and the output is the generated answer. An example of a prompt sentence is as follows:

[1098] "Please answer the following questions: Project Status\n\nReply: "

[1099] Step 5:

[1100] The emotion engine analyzes user input and reactions to recognize emotions. The emotion engine analyzes keyboard input speed, keystroke strength, mouse movement, and, in some cases, data from biometric sensors. The input is user input and reaction data, and the output is emotional data that indicates the user's state.

[1101] Step 6:

[1102] The emotion engine adjusts the generated answer based on the recognized emotion data. For example, if it recognizes that the user is stressed, it adds comforting words such as "Don't worry" to the answer. The input is the generated answer and emotion data, and the output is the adjusted answer.

[1103] Step 7:

[1104] The user terminal displays the adjusted answer. The user reviews the adjusted answer, makes corrections if necessary, and finally submits the answer. The input is the adjusted answer, and the output is the final answer that has been reviewed and corrected by the user.

[1105] Step 8:

[1106] The server accepts inquiries from customers using QR codes in the store. The inquiry content obtained by scanning the QR code is imported into the system and goes through a process similar to that for email. The input is the inquiry content obtained from the QR code, and the output is the inquiry data imported into the system.

[1107] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1108] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1109] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1110] [Fourth embodiment]

[1111] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1112] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1113] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1114] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1115] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1116] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1117] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1118] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1119] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1120] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1121] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1122] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1123] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1124] An embodiment of the present invention relates to a system that automates the process of responding to emails. This system coordinates automatic email extraction, content analysis, information search, response generation, and confirmation / sending via a user interface. The system of the present invention operates primarily through the coordination of a server and a user terminal.

[1125] Server Operation

[1126] Automatic email extraction

[1127] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[1128] Content analysis

[1129] The server uses natural language processing technology to analyze the content of the email and extracts important keywords to understand the intent and context of the email. For example, the server can extract the keywords "project" and "progress" from an email with a subject such as "Regarding the progress of a new project."

[1130] Information Search

[1131] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[1132] Answer generation

[1133] The server uses a generative AI model to generate appropriate answers based on the search results. For example, based on the keywords above, it might generate an answer such as, "The current project progress is on schedule and within budget."

[1134] User terminal operation

[1135] User Interface

[1136] The user terminal presents the answer generated by the server to the user. The user can check this answer and modify it as necessary. For example, the user may modify the displayed answer by adding more specific progress information. The user then submits the final answer.

[1137] Specific examples

[1138] For example, consider the following email sent to a user:

[1139] Please tell me about the progress of your new project.

[1140] Server behavior:

[1141] The server periodically checks the mailbox to detect new mail and retrieves it.

[1142] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[1143] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[1144] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[1145] User terminal behavior:

[1146] The user terminal displays the generated answer to the user.

[1147] The user checks the answers and makes corrections, such as adding "more specific progress information."

[1148] After making corrections, the user creates a final response and sends it to the other party.

[1149] In this way, the present invention can significantly automate the process of responding to e-mails, reducing the workload of users and improving productivity.

[1150] The processing flow will be explained below.

[1151] Step 1:

[1152] The server periodically accesses the user's mailbox to check for new emails, and if there are any new emails, it imports them into the system.

[1153] Step 2:

[1154] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords.

[1155] Step 3:

[1156] The server searches for related information from internal and external sources based on the extracted keywords, utilizing past email history, the company intranet, the company knowledge base, the Internet, and other sources.

[1157] Step 4:

[1158] The server uses a generative AI model to generate appropriate answers based on the search results, and outputs the answers as syntactically and semantically consistent sentences using natural language processing technology.

[1159] Step 5:

[1160] The user terminal displays the generated answer sent from the server to the user, who can then check the displayed answer and modify it as necessary.

[1161] Step 6:

[1162] After the user has confirmed and corrected the answer, he or she clicks the send button to send the final answer. At this time, the user's operation details are sent to the server and then to the other party as an e-mail.

[1163] In this way, the process of responding to emails is automated, significantly reducing the amount of work required by the user.

[1164] Example 1

[1165] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1166] The traditional email response process requires a lot of manual work, which is time-consuming and labor-intensive, reducing the user's work efficiency. Furthermore, in order to understand the content of an email and respond appropriately, the user must analyze each piece of content and generate a response, which places a heavy burden on the user. Therefore, there is a need for a system that automates the email response process, reduces the user's workload, and improves productivity.

[1167] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1168] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords using natural language processing technology, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating appropriate answers based on the search results using a generative AI model, means for displaying the generated answers on a user terminal so that the user can confirm and modify them, and means for sending the final answers. This allows the email response process to be largely automated, reducing the user's workload and improving productivity.

[1169] "New email" is a new electronic message that has arrived in a user's mailbox.

[1170] "Periodic detection" means that new e-mails are automatically checked according to a preset time interval.

[1171] The "means for capturing" is a function for storing the contents of detected new e-mails in the system.

[1172] "Analyzing the content" refers to analyzing the text information of the acquired email using natural language processing technology.

[1173] "Natural language processing technology" is a general term for technologies that enable computers to understand, interpret, and generate human language.

[1174] "Extracting important keywords" means identifying key words and phrases from the content of an email that are necessary to understand its intent and context.

[1175] "Internal sources" refer to information repositories within an organization, including archived emails, internal intranets, and internal knowledge bases.

[1176] "External information sources" refer to repositories of information that can be obtained from outside the organization, primarily including the Internet.

[1177] "Searching for related information" refers to using the extracted keywords to research and obtain related information from internal and external sources.

[1178] A "generative AI model" refers to an algorithm or system that uses artificial intelligence technology to generate sentences in natural language.

[1179] "Means for generating answers" refers to a function that uses a generative AI model to create appropriate answers based on search results.

[1180] "User terminal" refers to equipment such as a computer or mobile device used by a user.

[1181] The "means for checking and correcting" is a function that displays the generated answer to the user, and allows the user to check the content and edit it as necessary.

[1182] The "means for sending the final answer" is a function for the user to send the answer that has been confirmed and corrected by the user to the other party as an email.

[1183] An embodiment of the present invention relates to a system for automating the process of responding to emails. This system operates mainly through cooperation between a server and a user terminal.

[1184] Server Operation

[1185] Automatic email extraction

[1186] The server periodically accesses the user's mailbox to detect new emails and import their contents into the system. Specifically, it periodically connects to the mail server using a mail protocol such as IMAP or POP3 to retrieve new emails. For example, the server connects to the mail server once every hour to check for new emails.

[1187] Content analysis

[1188] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy or NLTK). During this process, the email text is tokenized, tagged with parts of speech, and important keywords are extracted. For example, the keywords "project" and "progress" are extracted from an email that asks, "Please tell us about the progress of your new project."

[1189] Information Search

[1190] The server uses the extracted keywords to search for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet searches). Specifically, it executes SQL queries using the keywords to retrieve information from the internal database. It also uses external sources such as the Google Search API.

[1191] Answer generation

[1192] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT series) based on the search results. The search results and prompt are input into the generative AI model, which generates an appropriate answer in natural language. For example, the prompt "Project progress is progressing smoothly and within budget" is input.

[1193] User terminal operation

[1194] User Interface

[1195] The user terminal displays the generated answer to the user. The user can review the displayed answer and edit it if necessary. The user interface provides text boxes and editing tools to allow the user to easily modify the answer. For example, the user can enter additional information such as "The next milestone is the start of the testing phase."

[1196] Last sent

[1197] Once the user has finished making corrections, they can finalize the answer and send it to the other party as an email. The email is sent via the user's mail server using the SMTP protocol, making it easy for the user to send the final answer.

[1198] Specific examples

[1199] Consider the following example:

[1200] Please tell me about the progress of your new project.

[1201] Server behavior:

[1202] 1. The server connects to the user's mail server once every hour and checks for new email.

[1203] 2. Detect new emails and save their contents to a local database.

[1204] 3. Use Spacy to tokenize the text and extract the keywords "project" and "progress."

[1205] 4. Use SQL queries to retrieve relevant information from an internal knowledge base and use the Google Search API to retrieve information from the internet.

[1206] 5. The prompt sentence "Project progress is progressing smoothly and within budget" is input into the generative AI model, which generates the answer "Current project progress is progressing as planned and within budget."

[1207] User terminal behavior:

[1208] 1. The user's device displays the generated answer, and the user confirms the content.

[1209] 2. The user corrects their answer by adding additional information: "The next milestone is the start of the testing phase."

[1210] 3. The user confirms the revised answer and clicks the submit button.

[1211] 4. The final response is sent via the mail server using the SMTP protocol.

[1212] Prompt Sentence Examples

[1213] You receive an email asking, "How's your new project going?" Please generate an answer based on the following information:

[1214] Project name: AI system development project

[1215] Progress: 50% complete

[1216] Next Milestone: Testing Phase Begins

[1217] Budget status: On track and within budget

[1218] Using this prompt, the appropriate generative AI model will respond, generating the answer, "The AI ​​system development project is now 50% complete and is progressing towards the next milestone, the start of the testing phase. We are on track and within budget."

[1219] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1220] Step 1:

[1221] Automatic email extraction

[1222] The server connects to the user's mail server (e.g., using IMAP or POP3 protocol) at set intervals to detect and retrieve new emails. The input is the connection information to the user's mail server, and the output is the detected new email data. Specifically, the server connects to the mail server once every hour to check for new emails and saves the detected new emails in a local database.

[1223] Step 2:

[1224] Content analysis

[1225] The server analyzes the content of the imported email using natural language processing technology (e.g., Spacy, NLTK). The input is the text data of the email, and the output is important keywords extracted through analysis. The server tokenizes the text and tags it with parts of speech. For example, the server extracts the keywords "project" and "progress" from an email that says, "Please tell us about the progress of the new project."

[1226] Step 3:

[1227] Information Search

[1228] Based on the extracted keywords, the server searches for related information from internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet search). The input is the extracted keywords and the output is related information. The server executes SQL queries to retrieve related information from the internal database and uses the Google Search API to retrieve related information from the Internet.

[1229] Step 4:

[1230] Answer generation

[1231] The server generates an appropriate answer using a generative AI model (e.g., OpenAI GPT series) based on the search results. The input is the search results and prompt, and the output is the generated answer. The server generates a prompt, "Project progress is progressing smoothly and within budget," and inputs it into the generative AI model. This generates an answer, "Current project progress is proceeding as planned and within budget."

[1232] Step 5:

[1233] User verification and correction

[1234] The user terminal displays the generated answer to the user. The user checks this answer and makes corrections as necessary. The input is the generated answer, and the output is the corrected answer. The user interface has text boxes and editing tools, allowing the user to enter additional information, such as "The next milestone is the start of the test phase," and correct the answer.

[1235] Step 6:

[1236] Final submission of response

[1237] The user terminal confirms the final answer after the user has completed the corrections and sends it as an email. The input is the corrected answer text, and the output is the sent email. Specifically, the user confirms the corrected answer and clicks the send button. The user terminal uses the SMTP protocol to connect to the mail server and send the final answer.

[1238] (Application example 1)

[1239] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1240] Conventional email response systems require users to manually check emails, analyze their content, and create responses, resulting in significant time and effort. Furthermore, in today's world, where rapid responses to inquiries are required, there is a growing need for systems that can efficiently and appropriately respond. In particular, content distribution services require rapid and accurate responses because of the wide variety of user inquiries. The present invention aims to solve these problems and significantly automate the email response process.

[1241] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1242] In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using a generation system based on the search results, means for displaying the generated answer on an output device so that a user can confirm and modify it by operating the output device, means for sending the generated answer as a reply to an inquiry, means for calling a generative AI model using a prompt sentence and generating an answer, and document generation means for automatically generating an appropriate answer to an inquiry. This allows users to efficiently automate the process from email content analysis to answer generation, enabling quick and accurate responses.

[1243] "New email" refers to new electronic mail that arrives in your inbox periodically.

[1244] "Content capture means" refers to the process or technology used to capture email content into the system.

[1245] "Means of content analysis" refers to the process of analyzing the content of captured emails using natural language processing technology to understand their intent and context.

[1246] "Important keywords" refer to words extracted from the content of emails that are important for extracting specific information.

[1247] "Internal sources" refers to information repositories maintained within internal systems, historical email history, intranets, and knowledge bases.

[1248] "External information sources" refers to information resources outside the company, such as the Internet.

[1249] "Means for searching related information" refers to the process of finding related information from internal and external sources based on the extracted keywords.

[1250] "Means for generating appropriate answers" refers to the process of using AI technology to create appropriate answers based on search results.

[1251] "Output device" refers to an appliance or device for displaying generated answers to a user.

[1252] "Means of invoking a generative AI model using a prompt sentence to generate an answer" refers to the process of running a generative AI model using a specified prompt sentence to construct an appropriate answer.

[1253] "Document generation means for automatically generating appropriate responses to inquiries" refers to a process that uses AI technology to automatically generate appropriate documents in response to user inquiries.

[1254] "Means for sending a reply to a query" refers to the technology or process for sending the generated answer back to the user.

[1255] This invention relates to a system for automating the process of responding to emails. This system automates everything from detecting new emails to generating and sending responses, thereby reducing the workload of users.

[1256] System Program Overview

[1257] The system mainly operates on a server and a user device. The server periodically detects new emails and analyzes their contents. Based on the analyzed important keywords, it searches for related information from internal and external sources. It then uses a generative AI model to generate an appropriate answer, which is displayed on the user's device. The user can then confirm and modify the answer, and send the final answer to the other party.

[1258] Hardware and software configuration

[1259] server:

[1260] The server uses the following software and libraries:

[1261] IMAP4 library: Used to retrieve new emails.

[1262] Python email module: Used to analyze the content of retrieved emails.

[1263] Generative AI models (OpenAI's GPT-3 / 4): Use natural language processing techniques to generate answers.

[1264] User device:

[1265] The user terminal uses the following devices and applications:

[1266] Output device (e.g. smartphone, tablet, PC): Displays the answers from the server and allows the user to check and correct them.

[1267] Data processing and calculation

[1268] The server processes data in the following steps: First, it periodically detects new emails using the IMAP4 library and analyzes their contents using the email module. This analysis extracts important keywords. Next, based on the extracted keywords, it searches for related information from internal sources (past emails, intranets, knowledge bases) and external sources (the Internet). It then uses a generative AI model to generate appropriate answers based on the search results. The generated answers are sent to the user's device, where they can be reviewed and revised by the user.

[1269] Specific examples

[1270] For example, consider a case where a user makes a query such as "What are the latest movie listings?"

[1271] The server detects new emails and analyzes the query content. Based on the extracted keywords "latest" and "movie list," it searches for related information from internal and external sources. Based on the results, it queries the generative AI model using the following prompt:

[1272] Example prompt sentence:

[1273] User Question: What is the latest movie list?

[1274] Generate the appropriate answer.

[1275] The generative AI model generates an appropriate answer based on this prompt, for example, "The latest movie list is as follows: 1. Inception 2. Avatar 2 3. Titanic."

[1276] The generated answer is displayed on the user's terminal, where the user can check the content and make corrections as necessary, after which the user can submit the final answer.

[1277] This invention enables users to respond to inquiries quickly and accurately, thereby improving work efficiency.

[1278] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1279] Step 1:

[1280] The server periodically checks for new emails. Using the IMAP4 library, it connects to the specified mail server and retrieves unread emails. At this point, the input is the mail server authentication information and the user's mailbox, and the output is a list of unread emails.

[1281] Step 2:

[1282] The server analyzes the contents of the retrieved email. It uses the Python email module to analyze the email content (body, subject, sender information, etc.). As a result of the analysis, it extracts important keywords from the subject and body of the email. At this point, the input is the content of the unread email, and the output is the extracted important keywords.

[1283] Step 3:

[1284] The server searches for related information based on the extracted keywords. It searches for related information from internal sources (past email history, intranet, knowledge base) and external sources (Internet). For example, it searches for related information using keywords such as "latest" and "movie list." At this point, the input is the extracted keywords, and the output is a list of related information.

[1285] Step 4:

[1286] The server uses a generation system to generate an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-3 or GPT-4), it inputs a prompt such as "User question: What are the latest movie listings? Please generate an appropriate answer." and receives the generated answer. At this point, the input is the prompt and related information, and the output is the generated answer.

[1287] Step 5:

[1288] The server sends the generated answer to the user terminal. The generated answer is displayed on the user terminal via the transmission system. At this point, the input is the generated answer, and the output is the display result on the user terminal.

[1289] Step 6:

[1290] The user checks the displayed answer and makes corrections if necessary. The user checks the answer using a device such as a smartphone or PC and makes corrections if necessary. The input at this point is the displayed answer and the user's corrections, and the output is the final answer after corrections.

[1291] Step 7:

[1292] The user sends the final answer to the other party. The user sends the revised answer from the user terminal to the system, and the system returns it to the other party. At this point, the input is the revised final answer, and the output is the sent answer.

[1293] The above steps automate the email response process, enabling fast and accurate responses.

[1294] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1295] An embodiment of the present invention relates to a system that automates the process of responding to emails and generates appropriate responses by recognizing user emotions. This system coordinates automatic email extraction, content analysis, information search, response generation, emotion recognition, and confirmation / sending via a user interface. The system of the present invention operates through the cooperation of a server, an emotion engine, and a user terminal.

[1296] Server Operation

[1297] Automatic email extraction

[1298] The server periodically accesses the user's mailbox to detect new emails and import them into the system. For example, the server connects to the mail server at specific time intervals set by the user to retrieve new emails.

[1299] Content analysis

[1300] The server analyzes the content of the email using natural language processing technology. Specifically, it divides the email body and subject line into tokens, removes unnecessary words, and extracts important keywords.

[1301] Information Search

[1302] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. Based on the search results, relevant information is obtained.

[1303] Answer generation

[1304] The server uses a generative AI model to generate an appropriate answer based on the search results. For example, from an email about the progress of a new project, the server might generate an answer such as, "The current project is progressing on schedule and within budget."

[1305] Emotion Engine Operation

[1306] emotion recognition

[1307] The emotion engine analyzes emotions in real time based on user input and reactions, using keyboard input, mouse movements, and even biometric signals.

[1308] Emotion-based response adjustment

[1309] The emotion engine recognizes the user's emotions and adjusts the content of the generated responses based on the results. For example, if the user is feeling stressed, the responses can be tailored to be more kind and comforting.

[1310] Use of past emotion data

[1311] The emotion engine stores the user's past emotional data for a certain period of time and uses it to more accurately recognize emotions and respond, making it possible to respond based on the user's long-term emotional tendencies.

[1312] User terminal operation

[1313] User Interface

[1314] The user terminal presents the answer generated by the server to the user. The user can review the answer and modify it as necessary. For example, the user can modify the answer by adding a specific number. The final answer is then sent.

[1315] Specific examples

[1316] For example, consider the following email sent to a user:

[1317] Please tell me about the progress of your new project.

[1318] Server behavior:

[1319] The server periodically checks the mailbox to detect new mail and retrieves it.

[1320] The content of the imported emails is analyzed and keywords such as "project" and "progress" are extracted.

[1321] Based on the extracted keywords, related information is searched for from the company intranet, the company knowledge base, and the Internet.

[1322] Based on the search results, a generative AI model is used to generate the answer, "Current project progress is on schedule and within budget."

[1323] Emotion Engine in action:

[1324] The emotion engine analyzes the user's keyboard input and mouse movements to recognize when the user is feeling stressed.

[1325] Based on the recognition results, a comforting phrase such as "The current project is progressing very well. Please rest assured." is added to the generated answer.

[1326] User terminal behavior:

[1327] The user terminal displays the generated answer to the user.

[1328] The user checks the answers and makes corrections if necessary.

[1329] After making corrections, the user creates a final response and sends it to the other party.

[1330] This not only automates the process of responding to emails, but also enables appropriate responses that take into consideration the user's feelings, thereby achieving even higher levels of customer satisfaction.

[1331] The processing flow will be explained below.

[1332] Step 1:

[1333] The server periodically accesses the user's mailbox to check for new emails. For example, it accesses the mail server every five minutes to check if new emails have arrived. If there are any new emails, it imports them into the system.

[1334] Step 2:

[1335] The server analyzes the content of the emails it receives. Specifically, it uses natural language processing technology to divide the email body and subject line into tokens, removes unnecessary words (stop words), and extracts important keywords. For example, from an email that asks, "Please tell me about the progress of the new project," it extracts keywords such as "new project" and "progress."

[1336] Step 3:

[1337] The server searches for relevant information from internal and external sources based on the extracted keywords. Internal sources include past email history, the company's intranet, and the company's knowledge base, while external sources include the Internet. For example, the server searches the company's intranet using the keywords "new project" and "progress" to retrieve relevant project plans and progress reports.

[1338] Step 4:

[1339] The server compiles the information it has acquired and uses a generative AI model to generate an appropriate response, such as "The current project is progressing on schedule and within budget."

[1340] Step 5:

[1341] The emotion engine analyzes the user's input and other biometric signals (e.g., keyboard typing speed and mouse movements) to recognize the user's emotional state. For example, if the user types quickly, it may determine that the user is stressed.

[1342] Step 6:

[1343] The emotion engine adjusts the generated answers based on the perceived emotion, for example, if it detects that the user is stressed, it will add a comforting element such as "Don't worry, progress on the current project is going very well."

[1344] Step 7:

[1345] The user terminal displays the generated answer received from the server to the user. The user can check the displayed answer and modify the content as necessary. For example, the user can modify the answer by adding specific progress data.

[1346] Step 8:

[1347] After checking and correcting the answer, the user clicks the send button to send it to the other party. At this time, the user's operation is sent to the server, and the final email is sent to the other party, completing the response.

[1348] By using this process flow, the present invention can automate the user's email response process and also realize an appropriate response that takes into consideration the user's feelings.

[1349] Example 2

[1350] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1351] Conventional email response systems require users to manually analyze the content of received emails and create replies, which requires a great deal of time and effort. Furthermore, the quality of responses can be poor because responses cannot take into account the user's feelings. The present invention aims to solve these problems by improving the efficiency of email response processes and providing high-quality responses that take into account the user's feelings.

[1352] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for periodically detecting new e-mails and capturing their contents, means for analyzing the contents of the captured e-mails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer using an artificial intelligence model based on the search results, means for recognizing the user's emotions and adjusting the content of the generated answer based on the emotions, and means for displaying the generated answer on the user terminal so that the user can confirm and correct it. This makes it possible to automate the e-mail response process and provide an appropriate response that takes the user's emotions into consideration.

[1353] "Newly arrived email" refers to unread email that has been newly received by the user.

[1354] "Periodic detection" means that e-mails are automatically checked at preset time intervals.

[1355] "Means for retrieving" refers to the function of retrieving emails from an email server and storing them within the system.

[1356] "Content analysis" refers to analyzing the body and subject of an email using natural language processing techniques.

[1357] "Key word extraction" is the process of identifying and extracting specific tokens or phrases from the analyzed content.

[1358] "Internal sources" refer to documents, databases, and past communications records shared within an organization.

[1359] "External sources" refer to publicly available data on the Internet, information provided by other companies or organizations, etc.

[1360] "Searching for related information" means searching for relevant data from predefined information sources based on the extracted keywords.

[1361] "Artificial intelligence model" refers to an intelligent system that uses machine learning algorithms to generate output based on input data.

[1362] "Means for generating a response" refers to the function of constructing an appropriate reply based on the relevant information obtained.

[1363] "Recognizing user emotions" is the process of inferring a user's emotional state from their behavior and biosignals.

[1364] "Adjusting the content of the generated response based on emotion" refers to modifying the content and tone of the generated response according to the recognized emotion.

[1365] "User terminal" refers to a device such as a computer, smartphone, or tablet that is actually operated by a user.

[1366] "Means for checking and correcting" refers to a function that allows the user to check the automatically generated answers and manually correct them if necessary.

[1367] The present invention is a system that integrates an automatic email processing function with a user emotion recognition function. This system operates by linking a server, an emotion engine, and a user terminal.

[1368] Server Operation

[1369] Automatic email extraction

[1370] The server periodically accesses the user's email box to detect new emails, for example by connecting to a mail server such as Gmail or Outlook using the IMAP protocol to retrieve new emails.

[1371] Content analysis

[1372] The server then analyzes the email content using natural language processing techniques, using the Python Natural Language Toolkit (NLTK) library to tokenize the email body and subject line and extract key keywords.

[1373] Information Search

[1374] The server uses the extracted keywords to search for relevant information from internal sources (databases, knowledge bases, intranets) and external sources (the Internet) using SQL queries to the internal database and Google Search APIs.

[1375] Answer generation

[1376] The server generates an appropriate answer using a generative AI model (e.g., OpenAI's GPT-3) based on the search results. The generative AI model receives a prompt such as: "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." An example of a generated answer is, "Current project progress is on schedule and within budget."

[1377] Emotion Engine Operation

[1378] emotion recognition

[1379] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, and even biometric signals. Emotion recognition uses machine learning libraries such as TensorFlow and PyTorch to identify emotions such as stress or joy that the user is feeling.

[1380] Emotion-based response adjustment

[1381] The content of the generated answer is adjusted based on the results of emotion recognition. For example, if the user is feeling stressed, the answer obtained from the generative AI model will include comforting words such as "Please continue to watch over the project with peace of mind."

[1382] Use of past emotion data

[1383] The emotion engine stores the user's past emotional data for a certain period of time and uses this data to more accurately recognize and respond to emotions, enabling advanced responses based on the user's long-term emotional tendencies.

[1384] User terminal operation

[1385] User Interface

[1386] The user's device displays the answer generated by the server to the user. The user can review the answer and make corrections as necessary. For example, a front-end framework such as ReactJS or Vue.js can be used to display the answer in an easy-to-understand format. After the user makes corrections to the answer, such as adding specific numbers, the final answer can be submitted.

[1387] This system automates the email response process and enables appropriate responses that take into consideration the user's feelings, which contributes to increased customer satisfaction and dramatically improves business efficiency.

[1388] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1389] Step 1: Automatic email extraction

[1390] The server periodically accesses the user's mailbox to detect new emails. The server connects to the mail server using the IMAP protocol and retrieves new emails. The input includes the user's mailbox information and access frequency settings. The output is the data of new emails that are imported into the server. Specifically, for example, the server automatically synchronizes with the mail server every morning at 8:00 and detects new emails.

[1391] Step 2: Analyzing the email content

[1392] The server analyzes the content of the newly received emails using natural language processing technology. The input includes the data of the newly received emails. Specifically, it uses Python's NLTK library to tokenize the email body and subject, removes unnecessary words, and extracts important keywords using techniques such as TF-IDF. The output is the extracted keywords.

[1393] Step 3: Information search

[1394] The server searches for related information from internal information sources (databases, knowledge bases, intranets) and external information sources (the Internet) based on the extracted keywords. The input includes the extracted keywords. Specifically, it executes an SQL query to the internal database to retrieve the relevant data. It also retrieves related information from external information sources using the Google Search API. The output is the retrieved related information.

[1395] Step 4: Answer Generation

[1396] The server uses a generative AI model (such as GPT-3) to generate an appropriate answer based on the search results. The input includes the relevant information obtained as search results. An example of a prompt is, "In response to an email about the progress of a new project, please generate an answer that provides current progress and reassurance." Specifically, the prompt and search results are input into the generative AI model to generate an answer. The output is the generated answer.

[1397] Step 5: Emotion Recognition

[1398] The emotion engine analyzes emotions in real time using the user's keyboard input, mouse movements, biometric signals, etc. Inputs include user behavioral data (e.g., typing speed, mouse movements, etc.). Specifically, it uses machine learning models (TensorFlow or PyTorch) to identify the user's emotions. The output is the user's emotional state.

[1399] Step 6: Adjust your responses based on emotion

[1400] The emotion engine adjusts the content of the generated answer based on the analyzed user's emotions. The input includes the generated answer and the user's emotional state. For example, if the user is feeling stressed, the engine adds a phrase such as "Please continue to monitor the project with peace of mind." The output is an answer adjusted to take the user's emotions into account.

[1401] Step 7: Present and send the answer to the user

[1402] The user device presents the generated answer to the user and provides an interface that allows the user to review and modify it. The input includes the adjusted answer. Specific operations include displaying the answer in a user interface using a front-end framework such as ReactJS or Vue.js. The user can review the answer and modify it, such as adding a specific number. The user then creates and submits the final answer. The output is the modified final answer, which is then sent to the other party.

[1403] (Application example 2)

[1404] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1405] In conventional email response systems, the process of analyzing email content and generating appropriate responses required a lot of time and effort. Furthermore, if store clerks were unable to respond to customer inquiries immediately, customer satisfaction could decline. Furthermore, because responses were not flexibly adjusted based on customer emotions, there were limitations to how sensitive the system could be to customer emotions. Therefore, there is a need for a system that automates the processing of emails and inquiries and generates appropriate responses based on emotion recognition.

[1406] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically detecting new emails and capturing their contents, means for analyzing the contents of the captured emails and extracting important keywords, means for searching for related information from internal and external information sources based on the extracted keywords, means for generating an appropriate answer based on the search results, means for displaying the generated answer on a user terminal so that the user can confirm and modify it, means for adjusting the generated answer based on emotion recognition, and means for accepting customer inquiries in-store using QR codes. This makes it possible to automate the processing of emails and inquiries and provide flexible responses that take emotions into consideration.

[1407] "New email" is unread email that has just arrived in your inbox.

[1408] The "means for acquiring content" is a device or program that has the function of acquiring new e-mails into the system.

[1409] "Analyzing the content" means analyzing the text data of the imported email using natural language processing technology.

[1410] The "means for extracting important keywords" is a device or program that identifies and extracts keywords with specific meanings from the analyzed text data.

[1411] "Internal information sources" are information resources such as databases and knowledge bases within a company.

[1412] "External sources" are information resources outside the company, such as information on the Internet or other public databases.

[1413] The "means for searching related information" is a device or program that searches for relevant data from internal and external sources based on the extracted keywords.

[1414] The "means for generating an appropriate answer" is a device or program that generates an appropriate response to a query based on the search results.

[1415] A "user terminal" is a computer device or mobile device that can be operated by a user.

[1416] The "means for confirming and correcting" is a device or program that has the function of allowing the user to confirm the generated answer and make corrections as necessary.

[1417] The "means for adjusting based on emotion recognition" is a device or program that analyzes the user's input and response, recognizes the emotion, and modifies the response appropriately.

[1418] "Means for accepting inquiries in-store using QR codes" refers to a device or program that has the function of scanning QR codes and accepting inquiries electronically.

[1419] A system embodying the present invention automates email correspondence and can generate appropriate responses by recognizing the user's emotions. An embodiment of this system will now be described in detail.

[1420] System configuration

[1421] The system consists of a server, a user terminal, an emotion engine, and a QR code scanning device. The server has the following main functions:

[1422] Ability to periodically detect new emails and retrieve their contents

[1423] Ability to analyze email content and extract important keywords

[1424] Ability to search for related information from internal and external sources based on extracted keywords

[1425] Generate relevant answers based on search results

[1426] A function that displays the generated answers on the user's device so that the user can check and correct them.

[1427] The ability to tailor generated answers based on emotion recognition

[1428] A function to accept customer inquiries in-store using QR codes

[1429] Hardware and software used

[1430] Hardware: Servers, smartphones, smart glasses, QR code scanners

[1431] software:

[1432] Natural language processing: SpaCy(ja_core_news_sm)

[1433] Generative AI model: GPT-2 (transformers)

[1434] Email communication: smtplib, imaplib

[1435] Emotion recognition: keyboard input, mouse movement analysis software, and possibly biometric sensors

[1436] Operation process

[1437] Server Operation

[1438] The server periodically connects to the mail server to detect new emails. The detected emails are imported and their contents are analyzed to extract important keywords. Based on the extracted keywords, relevant information is searched from internal sources (past emails, internal intranet, internal knowledge base) and external sources (the Internet). Based on the search results, an appropriate answer is generated using a generative AI model (GPT-2). The following prompt sentences are used in this generation process:

[1439] Example prompt:

[1440] "Please answer the following questions: Project Status\n\nReply: "

[1441] Emotion Engine Operation

[1442] The emotion engine recognizes emotions by analyzing user input and reactions. For example, it can assess emotions in real time using keyboard typing speed and strength, mouse movements, and biometric sensors. The emotion engine adjusts the content of the generated response based on the analysis results. For example, if the user is feeling stressed, it will add comforting words such as "Don't worry" to the generated response.

[1443] User terminal operation

[1444] The generated answer is displayed on the user's device, where the user can review it and make any necessary corrections. The final answer is sent at the user's discretion. In-store inquiries are also accepted through a QR code scanner, which is also captured in the system and goes through the same process.

[1445] Specific examples

[1446] For example, if a user receives an email like this:

[1447] Please tell me about the progress of your new project.

[1448] The server detects this email, analyzes its contents, and extracts the keywords "project" and "progress." Based on this, a generative AI model is used to generate the answer, "The current project is progressing smoothly." If the emotion engine recognizes the user's stress, it further adjusts the answer to take their emotions into consideration, such as, "The current project is progressing smoothly. Please rest assured."

[1449] This allows for efficient handling of emails and in-store inquiries, and enables flexible, emotional responses.

[1450] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1451] Step 1:

[1452] The server periodically detects new emails and retrieves their contents. The server accesses the user's mailbox and retrieves unread emails. The input is the new email data from the mail server, and the output is the retrieved email contents.

[1453] Step 2:

[1454] The server analyzes the content of the captured email and extracts important keywords. The server uses natural language processing software (e.g., SpaCy) to split the text into tokens, filter out meaningless words, and extract important keywords. The input is the email content, and the output is the extracted keywords.

[1455] Step 3:

[1456] The server searches for related information from internal and external sources based on the extracted keywords. The server sends search queries to internal sources (past emails, internal intranet, internal knowledge base) and external sources (Internet) to collect related information. The input is the extracted keywords, and the output is the search results.

[1457] Step 4:

[1458] The server generates an appropriate answer based on the search results. Using a generative AI model (e.g., GPT-2), the server inputs keywords and search results into a prepared prompt sentence to generate a natural language answer. The input is the search results, and the output is the generated answer. An example of a prompt sentence is as follows:

[1459] "Please answer the following questions: Project Status\n\nReply: "

[1460] Step 5:

[1461] The emotion engine analyzes user input and reactions to recognize emotions. The emotion engine analyzes keyboard input speed, keystroke strength, mouse movement, and, in some cases, data from biometric sensors. The input is user input and reaction data, and the output is emotional data that indicates the user's state.

[1462] Step 6:

[1463] The emotion engine adjusts the generated answer based on the recognized emotion data. For example, if it recognizes that the user is stressed, it adds comforting words such as "Don't worry" to the answer. The input is the generated answer and emotion data, and the output is the adjusted answer.

[1464] Step 7:

[1465] The user terminal displays the adjusted answer. The user reviews the adjusted answer, makes corrections if necessary, and finally submits the answer. The input is the adjusted answer, and the output is the final answer that has been reviewed and corrected by the user.

[1466] Step 8:

[1467] The server accepts inquiries from customers using QR codes in the store. The inquiry content obtained by scanning the QR code is imported into the system and goes through a process similar to that for email. The input is the inquiry content obtained from the QR code, and the output is the inquiry data imported into the system.

[1468] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1469] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1470] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1471] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1472] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1473] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1474] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1475] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1476] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1477] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1478] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1479] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1480] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1481] 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.

[1482] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1483] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1484] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1485] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1486] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1487] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1488] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1489] The following is further disclosed regarding the above embodiment.

[1490] (Claim 1)

[1491] a means for periodically detecting new email and capturing its contents;

[1492] A means of analyzing the content of the captured email and extracting important keywords;

[1493] A means of searching for related information from internal and external sources based on the extracted keywords;

[1494] a means for generating an appropriate answer based on the search results;

[1495] A means for displaying the generated answer on a user terminal so that the user can confirm and correct the answer;

[1496] A system including:

[1497] (Claim 2)

[1498] 10. The system of claim 1, further comprising means for automatically generating the generated answers using natural language processing.

[1499] (Claim 3)

[1500] 10. The system of claim 1, further comprising means for using past emails, company intranets, and company knowledge bases as internal information sources.

[1501] "Example 1"

[1502] (Claim 1)

[1503] a means for periodically detecting new email and capturing its contents;

[1504] A means for analyzing the content of the captured email and extracting important keywords using natural language processing technology;

[1505] a means for searching for related information from internal and external sources based on the extracted keywords;

[1506] A means of generating appropriate answers based on search results using an AI model; and

[1507] A means for displaying the generated answer on a user terminal so that the user can confirm and correct the answer;

[1508] a means for transmitting a final response;

[1509] A system including:

[1510] (Claim 2)

[1511] 10. The system of claim 1, further comprising means for automatically generating the generated answers using natural language processing.

[1512] (Claim 3)

[1513] 10. The system of claim 1, further comprising means for using historical emails, an internal intranet, and an internal knowledge base as internal information sources.

[1514] "Application Example 1"

[1515] (Claim 1)

[1516] a means for periodically detecting new email and capturing its contents;

[1517] A means of analyzing the content of the captured email and extracting important keywords;

[1518] a means for searching for related information from internal and external sources based on the extracted keywords;

[1519] a means for generating an appropriate answer based on the search results using a generation system;

[1520] a means for displaying the generated answer on an output device, and allowing the user to confirm and correct the answer by operating the output device;

[1521] means for transmitting the generated response in response to the inquiry;

[1522] A means for invoking a generative AI model using a prompt to generate an answer;

[1523] a document generation means for automatically generating an appropriate response to the inquiry;

[1524] A system including:

[1525] (Claim 2)

[1526] 10. The system of claim 1, further comprising means for automatically generating the generated answers using natural language processing.

[1527] (Claim 3)

[1528] 10. The system of claim 1, further comprising means for using historical emails, internal intranets, and internal knowledge bases as internal information sources.

[1529] "Example 2: Combining Emotion Engines"

[1530] (Claim 1)

[1531] a means for periodically detecting new email and capturing its contents;

[1532] A means of analyzing the content of the captured email and extracting important keywords;

[1533] A means of searching for related information from internal and external sources based on the extracted keywords;

[1534] a means for generating an appropriate answer using an artificial intelligence model based on the search results;

[1535] means for recognizing a user's emotion and adjusting the content of the generated answer based on the emotion;

[1536] A means for displaying the generated answer on a user terminal so that the user can confirm and correct the answer;

[1537] A system including:

[1538] (Claim 2)

[1539] 10. The system of claim 1, further comprising means for automatically generating the generated answers using natural language processing.

[1540] (Claim 3)

[1541] 10. The system of claim 1, further comprising means for using past emails, company intranets, and company knowledge bases as internal information sources.

[1542] "Application example 2 when combining emotion engines"

[1543] (Claim 1)

[1544] a means for periodically detecting new email and capturing its contents;

[1545] A means of analyzing the content of the captured email and extracting important keywords;

[1546] a means for searching for related information from internal and external sources based on the extracted keywords;

[1547] a means for generating an appropriate answer based on the search results;

[1548] A means for displaying the generated answer on a user terminal so that the user can confirm and correct the answer;

[1549] means for adjusting the generated answers based on emotion recognition;

[1550] A means of accepting customer inquiries in-store using QR codes,

[1551] A system including:

[1552] (Claim 2)

[1553] 10. The system of claim 1, further comprising means for automatically generating the generated answers using natural language processing.

[1554] (Claim 3)

[1555] 10. The system of claim 1, further comprising means for using historical emails, internal intranets, and internal knowledge bases as internal information sources. [Explanation of symbols]

[1556] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for periodically detecting new email and capturing its contents; A means of analyzing the content of the captured email and extracting important keywords; A means of searching for related information from internal and external sources based on the extracted keywords; a means for generating an appropriate answer based on the search results; A means for displaying the generated answer on a user terminal so that the user can confirm and correct the answer; A system including:

2. The system of claim 1 , further comprising means for automatically generating the generated answers using natural language processing.

3. 10. The system of claim 1, further comprising means for using past emails, a company intranet, and a company knowledge base as internal information sources.

Images