system

Abstract

Provide a system. A means for acquiring new emails; A means for analyzing the content of emails using natural language processing technology; A means for predicting reply content based on the analyzed content; A means of embedding predicted reply content into templates; means for presenting the template to a user; and means for sending the reply content that has been edited and reviewed by the user.

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] While replying to emails is a routine task for many employees, it requires a great deal of time and effort. As a result, it becomes difficult to allocate sufficient time to other important tasks, resulting in a decline in overall productivity. Furthermore, in situations where a quick and accurate reply is required, the time required to manually create a reply can be a problem. Furthermore, it is difficult to maintain a consistent response. To solve these issues, a system is needed to streamline email replies and quickly generate standardized, appropriate replies. [Means for solving the problem]

[0005] The present invention provides a system that uses generative AI and natural language processing technology to streamline email reply work and improve productivity. The present invention first includes a means for acquiring new emails. Next, the system includes a means for analyzing the acquired emails using natural language processing technology. Based on the analysis results, the system includes a means for predicting appropriate reply content using AI technology. The system further includes a means for embedding the predicted reply content into a template, and a means for presenting this template to the user. Finally, the system includes a means for sending reply content that the user has edited and confirmed. This enables employees to reply to emails quickly and accurately, improving overall productivity.

[0006] "New email" refers to email that has just arrived in the user's email account.

[0007] "Natural language processing technology" is a general term for technologies used by computers to understand, analyze, and generate human language.

[0008] "Analyzing" means converting the body of an email into structured data and extracting its intent and important information.

[0009] "Predicting reply content" means using AI technology to generate an appropriate reply based on the analysis results.

[0010] A "template" is a standard reply phrase that formalizes the reply content, and is a framework for embedding predicted reply content.

[0011] "Presenting to the user" means displaying the generated reply template on the user's screen so that the user can check and edit it.

[0012] "Edit and confirm" means that the user modifies the content of the presented reply as necessary and finally approves the content.

[0013] "Sending" means sending the edited and confirmed reply mail to the other party via the email server.

[0014] "Generative AI" is a type of artificial intelligence technology that can generate new data or text based on input data. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[0037] composition

[0038] The system has the following main functions:

[0039] 1. Receiving new emails

[0040] The server periodically checks the user's mail account to retrieve new emails, eliminating the need for the user to manually check their mail.

[0041] 2. Email Content Analysis

[0042] The server analyzes the content of the email using natural language processing technology, which extracts the email's intent and important information.

[0043] 3. Predicting replies

[0044] The server uses AI technology to predict the reply content based on the analyzed email content, using data from past emails and business context.

[0045] 4. Generate a reply template

[0046] The server embeds the predicted reply content into a predefined template and presents the completed reply email to the user.

[0047] 5. User review and editing

[0048] The terminal displays the template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[0049] 6. Sending emails

[0050] The server sends a reply email that the user finally checks and corrects.

[0051] Operation details

[0052] 1. Server Operation

[0053] The server connects to the user's email account and checks for new emails. If there are new emails, they are received and added to a processing queue for analysis. Next, natural language processing techniques are used to analyze the email's content and extract its intent (e.g., confirming a meeting date). Based on the results of this analysis, the server uses an AI model to predict an appropriate reply (e.g., "How about next Tuesday?").

[0054] 2. Device Operation

[0055] The terminal receives the predicted reply content from the server, embeds it in a template, and presents it to the user. An interface is provided for the user to view and edit the reply template.

[0056] 3. User Actions

[0057] The user reviews the proposed reply and edits it if necessary, and the final, reviewed reply is sent to the server, which then sends it to the email recipient.

[0058] Specific examples

[0059] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[0060] Server operation: The server receives the email and analyzes its contents. As a result of the analysis, the intent of "confirming the meeting date" is extracted. Based on this intent, the AI ​​model predicts the reply content, such as "How about next Tuesday?" This is then embedded in a template and presented to the user.

[0061] Terminal behavior: The terminal displays a template ("Dear [Recipient],\n\nHow about next Tuesday?\n\nBest regards,\n[Your Name]") to the user. The user edits this template, for example, to "How about next Tuesday at 3:00 PM?"

[0062] User action: User reviews the modified reply and sends it to the server, which finally sends this email.

[0063] In this way, the system of the present invention makes replying to e-mails more efficient and improves user productivity.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] Receive new emails

[0067] The server periodically checks the user's mail account to see if there is any new email. If new mail is found, it is retrieved from the mail server and added to a processing queue on the server.

[0068] Step 2:

[0069] Email content analysis

[0070] The server analyzes the body of newly arrived emails using natural language processing technology. Specifically, the body of the email is fed into an NLP (natural language processing) model to extract the email's intent and important information. The results of this analysis are used as input for subsequent processing.

[0071] Step 3:

[0072] Response prediction

[0073] The server then uses an AI model to predict the appropriate reply based on the analyzed email content. The AI ​​model incorporates data from past emails and business context, resulting in highly accurate replies.

[0074] Step 4:

[0075] Generate a reply template

[0076] The server embeds the predicted reply content into a standard template, which has fixed parts (e.g., greetings and signatures) and variable parts (reply content generated by AI), resulting in consistent reply emails.

[0077] Step 5:

[0078] Show reply content to user

[0079] The terminal displays the generated reply template to the user, and the user interface provides a form that allows the user to review and edit the reply content, allowing the user to easily review and modify the content.

[0080] Step 6:

[0081] User review and editing

[0082] The user checks the email reply template using the terminal. If necessary, the user edits the content of the template and performs a final check. Once editing is complete, the user clicks the send button to finalize the reply email.

[0083] Step 7:

[0084] Sending emails

[0085] The server receives the reply email that has been finally confirmed by the user. It then sends it to the recipient's email address via the mail server. Once the sending is complete, it is recorded in the sending history.

[0086] Through the above series of processing steps, the user can quickly and efficiently reply to e-mails, allowing the user to concentrate on other important tasks.

[0087] Example 1

[0088] 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."

[0089] In conventional email processing systems, users had to manually check new emails, analyze their content, and think up a reply, which required a great deal of time and effort. Particularly in busy work situations where quick and accurate responses are required, these manual tasks significantly reduced user productivity.

[0090] 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.

[0091] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content using AI technology based on the analyzed content, means for automatically embedding the predicted reply content into a template, means for presenting the template to the user and allowing the user to edit and confirm it, and means for sending the reply content edited and confirmed by the user. This enables the user to automate a series of processes from checking newly arrived emails, generating reply content, embedding it into a template, and sending the final email, quickly and efficiently.

[0092] "New email" is email that has just been received and that the user has not yet opened.

[0093] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes text tokenization, keyword extraction, and intent analysis.

[0094] "AI technology" is a technology that uses artificial intelligence to analyze data and generate appropriate answers and predictions.

[0095] "Means for predicting reply content" refers to functions and algorithms for generating appropriate reply content based on the content of received email.

[0096] A "template" is a model of a document created according to a specific format, and is designed to be easy for a user to edit.

[0097] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task (e.g., generating email replies).

[0098] A "prompt" is text data input to a generative AI model that provides instructions and context for the model to generate a reply.

[0099] "Automation" refers to the process by which machines or software automatically perform tasks that humans would otherwise perform manually.

[0100] "User review and editing" refers to the process in which the user reviews the email reply content generated by the system and makes corrections as necessary.

[0101] This invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[0102] Server behavior:

[0103] The server periodically connects to the user's email account and checks for new emails. If new emails are available, the server receives them using the IMAP or POP3 protocol. The server then analyzes the email content using natural language processing techniques (e.g., Python's NLTK or SpaCy). This analysis extracts the email's intent and important keywords. Next, the server uses a generative AI model (e.g., GPT-4 (registered trademark)) to predict an appropriate reply based on the analyzed content.

[0104] As a concrete example, if an email is received saying "Please confirm the date of the next meeting," the server analyzes the content of the email and extracts the intent of "confirm the date of the meeting." Based on that intent, the server inputs the prompt "Please confirm the date of the next meeting" into the generative AI model and obtains the reply text (for example, "How about next Tuesday?") obtained from the AI ​​model.

[0105] Terminal behavior:

[0106] The server sends the predicted reply content to the terminal to present to the user. The terminal embeds the predicted reply content in a template and displays it to the user. For example, this template has the format "Dear [Recipient],\n\n[How about next Tuesday?]\n\nBest regards,\n[Your Name]". The user checks the template presented to them through the terminal and edits the content as necessary. Once editing is complete, the user checks the content and finally sends it to the server.

[0107] User Action:

[0108] The user creates the final reply based on the template that has been checked and edited on the terminal. If any corrections are necessary, the template "How about next Tuesday?" can be edited to "How about next Tuesday at 3:00 PM?". Once editing is complete, the user sends the reply to the server. The server uses the SMTP protocol to send the final email to the recipient.

[0109] In this way, users can efficiently complete a series of automated processes, from checking new emails to creating, editing, and sending replies. This significantly reduces the time and effort required for users to process emails, improving productivity. This system ensures speed and accuracy in responding to emails, contributing to work efficiency.

[0110] As a concrete example, the following prompt sentences are fed into a generative AI model to generate a reply:

[0111] "Generate an appropriate response to an email asking to confirm the next meeting date."

[0112] Please use this information to create a polite business reply email.

[0113] This allows the generative AI model to automatically generate appropriate reply content.

[0114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0115] Step 1: Check for new emails

[0116] The server periodically connects to the user's email account and checks for new email. The server connects to the mail server using the IMAP or POP3 protocol and checks for new email. If new email is available, the server accepts it and proceeds to the next step.

[0117] Specific behavior:

[0118] Input: Email account information, IMAP / POP3 protocol

[0119] Data processing: Query the mail server and check for new mail.

[0120] Output: List of new emails (email ID, sender, subject, body)

[0121] Step 2: Analyzing the email content

[0122] The server acquires the content of newly received emails and analyzes them using natural language processing technology. Tools such as NLTK and SpaCy are used for the analysis. This analysis extracts the intent of the email and important keywords.

[0123] Specific behavior:

[0124] Input: Body of new email

[0125] Data processing: Using natural language processing tools to tokenize the email body and extract key keywords and intent

[0126] Output: Analysis results (intention, keywords)

[0127] Step 3: Anticipate replies

[0128] Based on the parsed content, the server inputs prompts into a generative AI model to predict appropriate responses, taking into account past email data and business context in the process.

[0129] Specific behavior:

[0130] Input: Analysis results (intention, keywords), prompt

[0131] Data calculation: Enter a prompt into the generative AI model and generate a reply.

[0132] Output: Predicted reply

[0133] Step 4: Generate a reply template

[0134] The server generates a reply template by embedding the predicted reply content into a predefined template, which is then generated in a format that is easy for the user to edit.

[0135] Specific behavior:

[0136] Input: predicted reply

[0137] Data processing: Embed reply content in a template

[0138] Output: Reply template (e.g., "Dear [Recipient],\n\n[Reply]\n\nBest regards,\n[Your Name]")

[0139] Step 5: User review and editing

[0140] The terminal presents the reply template received from the server to the user, who is then provided with an interface to review the template and edit its contents as necessary.

[0141] Specific behavior:

[0142] Input: Reply template

[0143] Data processing: Displaying a template on the user interface

[0144] Output: User edits

[0145] Step 6: Sending the final email

[0146] The server receives the final reply, edited and reviewed by the user, and sends it to the email recipient using the SMTP protocol.

[0147] Specific behavior:

[0148] Input: Reply confirmed by the user

[0149] Data Processing: Sending email using the SMTP protocol

[0150] Output: Sending result log

[0151] In this way, all steps are automated, allowing users to reply to emails quickly and accurately with minimal effort, improving user productivity and significantly increasing work efficiency.

[0152] (Application example 1)

[0153] 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."

[0154] For companies and organizations, responding to security-related emails is extremely important, but manual response requires a great deal of time and effort. Furthermore, it is not easy to respond appropriately and promptly to security alerts, which require a rapid response. Therefore, there is a need for a system that automates the entire process of receiving emails, analyzing them, generating replies, and sending them, thereby streamlining responses to security-related emails in particular.

[0155] 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.

[0156] In this invention, the server includes a means for acquiring newly arrived emails, a means for analyzing the content of the emails using natural language processing technology, a means for analyzing security-related emails and generating reply content, a means for verifying that the generated reply content is in accordance with the security policy, and a means for providing an interface designed to enable a prompt response to security alerts, thereby enabling prompt and appropriate responses to security-related emails in companies and organizations.

[0157] "Newly arrived email" refers to email with the most recent received date and time.

[0158] "Natural language processing technology" refers to computer technology for analyzing and understanding human language.

[0159] "Reply content prediction" refers to the process of estimating what should be replied to based on the content of the received email.

[0160] "Template embedding" refers to incorporating predicted reply content into a predetermined format or framework.

[0161] "Security-related email" refers to emails that contain content related to security threats such as phishing, malware, and unauthorized access.

[0162] A "security policy" refers to a guideline that describes the security measures and response policies established by an organization.

[0163] "Interface" refers to the screen and operation method used by users to operate the system.

[0164] A "generative AI model" refers to an artificial intelligence model that uses machine learning to generate information for specific tasks.

[0165] A "prompt" is a piece of text that describes instructions or questions to be input into a generative AI model.

[0166] A "server" refers to a computer system that performs a series of processes such as receiving, analyzing, generating a reply, and sending an email.

[0167] To implement this invention, it is necessary to build a system that automates security-related email correspondence. This system has the following main functions:

[0168] First, the server connects to the company or organization's mail server and retrieves new emails. This is done using the IMAP protocol. When new emails are received, their contents are stored in a database. The software used in this step is an email sending and receiving library (e.g., IMAP, SMTP).

[0169] Next, the server analyzes the content of the received email using natural language processing technology. This analysis uses natural language processing libraries such as spaCy and NLTK. As a result of the analysis, information particularly related to security is extracted. For example, keywords such as "phishing," "malware," and "anomaly detection" are identified.

[0170] The server then predicts the reply content based on the analysis results. This is where an AI model (e.g., GPT-4) is used. The AI ​​model references past data and security policies to generate the optimal reply content. This simultaneously verifies that the generated reply content complies with the security policy.

[0171] The generated reply is embedded in a pre-prepared template and presented to the user. The template uses the following prompt:

[0172] "Generate an appropriate response to the following phishing report email: Subject: Phishing Report Content: User reporting phishing activity Example of expected response: "Dear [reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter.""

[0173] The terminal displays this template to the user and provides an interface that allows the user to view and edit the content. This interface is designed to allow the user to select appropriate actions based on security policies.

[0174] After the user has finished editing and checking, the reply is finally sent to the server, which then sends it to the recipient using the SMTP protocol.

[0175] Through the above process, this system will enable companies and organizations to streamline security-related email responses and achieve prompt and appropriate responses.

[0176] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0177] Step 1:

[0178] The server connects to the mail server of the company or organization using the IMAP protocol to receive new emails and stores them in a database.

[0179] Input: New email from the company's email server

[0180] Output: New emails stored in the database

[0181] Specific operation: Connect to the mail server using the IMAP library, retrieve unread emails, and store their contents (body, subject, sender information, etc.) in a database.

[0182] Step 2:

[0183] The server analyzes the content of the stored emails using natural language processing techniques (e.g., spaCy or NLTK), particularly to extract security-related information (e.g., phishing, malware, unauthorized access).

[0184] Input: Contents of new emails stored in the database

[0185] Output: Parsed email content (security-related information)

[0186] Specific operations: Tokenize the email content, extract keywords, and analyze sentence structure to identify security-related information.

[0187] Step 3:

[0188] The server predicts the reply content based on the analysis results. Here, a generative AI model (e.g., GPT-4) is used. The AI ​​model generates the optimal reply content by referring to past data and security policies.

[0189] Input: Parsed email content (security-related information)

[0190] Output: Predicted reply

[0191] Specific operation: The analyzed information is input into the AI ​​model, and prompts are used to generate predicted replies. Example: "Generate an appropriate reply to the following phishing report email: Subject: Phishing Report Content: User's report of phishing activity Example of predicted reply: "Dear [Reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter."

[0192] Step 4:

[0193] The server embeds the generated reply content into a template based on a pre-defined format to create data to present to the user.

[0194] Input: predicted reply

[0195] Output: Reply content embedded in template

[0196] Specific operation: The predicted reply content is automatically embedded in the defined template and converted into a format that can be presented to the user.

[0197] Step 5:

[0198] The terminal displays the reply content embedded in the template to the user, and provides an interface for the user to view and edit the content.

[0199] Input: Reply content embedded in template

[0200] Output: A screen that the user can see and edit

[0201] What it does: Displays a form through the user interface to allow the user to review and edit the reply content embedded in the template, and waits for the user to finish editing.

[0202] Step 6:

[0203] The user then sends the reply content after reviewing and editing it to the server, which receives it and sends the final reply email to the email recipient.

[0204] Input: The reply that the user has reviewed and edited.

[0205] Output: Reply email sent

[0206] Specific behavior: Receives the content that the user has completed editing and sends a final reply email using the SMTP protocol.

[0207] The above steps create a system that automates security-related email responses and allows them to be handled efficiently and quickly.

[0208] 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.

[0209] The present invention relates to a system that automates the entire process from receiving new emails, analyzing them, generating reply content, embedding them in templates, and finally sending them, and combines it with an emotion engine that recognizes the user's emotions and optimizes the reply content.

[0210] composition

[0211] The system has the following main functions:

[0212] 1. Receiving new emails

[0213] The server periodically checks the user's mail account and receives new email, saving the user the trouble of having to manually check their mail.

[0214] 2. Email Content Analysis

[0215] The server analyzes the content of the email received using natural language processing technology to extract the intent of the email and important information.

[0216] 3. Emotional Recognition

[0217] The emotion engine recognizes the user's emotions and analyzes the user's emotional state from the email text and the user's reading.

[0218] 4. Predicting replies

[0219] The server uses AI technology to predict the appropriate reply based on the analysis results and data from the emotion engine. The emotional information from the emotion engine is reflected in the tone and expression of the reply.

[0220] 5. Generate a reply template

[0221] The server embeds the predicted reply content into a standard template and presents the completed reply email to the user, generating a consistent reply document.

[0222] 6. User Review and Editing

[0223] The terminal displays the reply template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[0224] 7. Sending emails

[0225] The server receives the reply email that has been finally confirmed by the user and sends it to the recipient. During this process, the sending history is recorded.

[0226] Operation details

[0227] 1. Server Operation

[0228] The server connects to the user's email account and receives new emails. The received emails are added to a processing queue within the server. Next, natural language processing technology is used to analyze the content of the email and extract its intent and important information. In parallel, an emotion engine recognizes the user's emotions and processes them by combining them with the analysis results.

[0229] 2. Device Operation

[0230] Based on the analysis data and emotion data received by the device from the server, the AI ​​model generates an appropriate reply. The generated reply is automatically embedded in a template and presented to the user. The user can then review the reply through the device interface and edit it as necessary.

[0231] 3. User Actions

[0232] The user reviews the reply displayed on the device, especially the tone and expressions based on emotions, and edits the reply if necessary. Once the final review is complete, the reply email is sent.

[0233] Specific examples

[0234] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[0235] Server operation: The server receives the email, analyzes its contents, and extracts the intent of "confirming the meeting date." At the same time, the emotion engine recognizes the user's current emotional state (e.g., high stress). Based on this, the AI ​​model predicts the appropriate reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[0236] What happens on the device: The device embeds the above reply content into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[0237] User action: The user checks the revised reply and gives a final confirmation before sending it to the server, which then sends the reply to the recipient.

[0238] This system allows users to reply to e-mails quickly and efficiently, and improves the quality of communication by taking emotional information into account when responding.

[0239] The processing flow will be explained below.

[0240] Step 1:

[0241] Receive new emails

[0242] The server periodically connects to the user's mail account to check for new emails. If new emails are found, they are retrieved from the mail server and added to a processing queue on the server.

[0243] Step 2:

[0244] Email content analysis

[0245] The server analyzes the body of the newly received email using natural language processing technology. During the analysis, the intent and important information are extracted from the email content. This data becomes the input information required for subsequent processing.

[0246] Step 3:

[0247] Emotion recognition

[0248] The emotion engine installed on the server collects the user's emotional data, analyzes the current emotional state (e.g., stress, joy, anger, etc.) from the text of the email, the user's past behavioral history, reading, etc., and generates quantified emotional data.

[0249] Step 4:

[0250] Response prediction

[0251] The server uses an AI model to predict an appropriate reply based on the analysis data obtained through natural language processing and the emotional data recognized by the emotion engine. For example, if the user is feeling stressed in response to an email asking, "Please confirm the date of the next meeting," the server will generate a gentler reply such as, "How about next Tuesday? We can also arrange other dates, so please let us know if that's convenient for you."

[0252] Step 5:

[0253] Generate a reply template

[0254] The server embeds predicted replies into standard templates, which have fixed parts (such as a greeting or signature) and variable parts (the predicted replies), resulting in consistent replies.

[0255] Step 6:

[0256] Show reply content to user

[0257] The terminal displays the generated reply template to the user, and the user interface provides a form for reviewing and editing the reply content, allowing the user to easily modify the content.

[0258] Step 7:

[0259] User review and editing

[0260] The user uses the device to check the displayed reply template. The user reviews the content and edits the template as needed. For example, the user can add a specific date, such as "How about 3:00 PM next Tuesday?"

[0261] Step 8:

[0262] Sending emails

[0263] The user finally sends the reply email that has been confirmed and edited to the server. The server receives and confirms this reply email, and sends it to the recipient's email address. Once the sending is complete, the history is recorded.

[0264] This allows users to reply to emails efficiently and appropriately according to their emotions, which not only improves work productivity but also the quality of communication.

[0265] Example 2

[0266] 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."

[0267] Conventional email reply systems require a lot of time and effort for manually checking emails and creating replies, making it difficult to reply quickly. Furthermore, the system generates a uniform reply without considering the user's feelings, which results in a decline in the quality of communication. The objective of this invention is to solve these problems and provide a system that realizes efficient, emotionally appropriate automatic replies.

[0268] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0269] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for recognizing the user's emotions using an emotion recognition engine based on the analyzed content, means for predicting reply content based on the analysis result and emotion data using a generative AI model, means for embedding the predicted reply content in a template, means for presenting the template to the user, and means for sending the reply content edited and confirmed by the user. This enables fast and efficient email replies and realizes high-quality communication that takes the user's emotions into consideration.

[0270] The "means for obtaining new e-mails" is a function for periodically accessing the user's e-mail account, checking whether new e-mails have arrived, and receiving them.

[0271] "Natural language processing technology" is a set of technologies that allow computers to understand, analyze, and generate human language, and is a language analysis technology used to understand the content of emails.

[0272] An "emotion recognition engine" is a system for recognizing a user's emotional state, and is a technology that estimates a user's emotions based on emails and past data.

[0273] A "generative AI model" is a model that uses artificial intelligence techniques to generate appropriate responses or documents based on given data or information.

[0274] "Analysis results" refers to information obtained when analyzing the content of an email using natural language processing technology, and includes the intent and important information.

[0275] "Emotion data" is information obtained as a result of analyzing the user's emotional state using an emotion recognition engine.

[0276] "Means for predicting reply content" is a function that uses a generative AI model to generate appropriate reply content based on the analysis results and emotional data.

[0277] The "means for embedding in a template" is a function for incorporating predicted reply content into a standard format prepared in advance.

[0278] The "means for presenting to the user" is a function for displaying the reply content embedded in the template to the user on the terminal.

[0279] The "means for sending reply content edited and confirmed by the user" is a function for sending reply content that the user has confirmed and edited to the recipient, and is necessary to complete the email sending process.

[0280] This invention relates to a system that automates the entire process from receiving new e-mails, analyzing their contents, generating reply content, embedding them in templates, and finally sending them, and further optimizes the reply content by recognizing the user's emotions. This system operates in cooperation with a server, a terminal, and a user.

[0281] Server Operation

[0282] The server periodically connects to the user's mail account to receive new emails using the IMAP or POP3 protocols, which are then added to a processing queue on the server.

[0283] Next, the server analyzes the email content using natural language processing (NLP) technology. The NLP model used here may be "Google® BERT" or "OpenAI® GPT." The server uses NLP technology to extract the main intent and important information from the email.

[0284] Furthermore, an emotion recognition engine is activated to recognize the user's emotions. This emotion recognition is performed using tools such as IBM Watson (registered trademark) Tone Analyzer. The analyzed emotion data is stored on the server and used to generate subsequent reply content.

[0285] The server uses a generative AI model (e.g., OpenAI GPT-3 (registered trademark)) to predict an appropriate reply based on the analysis results and emotional data. The reply is generated in a tone that reflects the user's emotions.

[0286] The predicted reply content is embedded in a standard template, which has a predefined format based on which a reply document is automatically generated.

[0287] Device behavior

[0288] The terminal displays the templated reply content received from the server to the user. The user can check the reply content on the terminal and edit it as necessary. The interface is user-friendly, allowing for easy editing.

[0289] User Actions

[0290] The user checks the reply displayed on their device, particularly the tone and expressions reflected by the emotion engine, and makes any necessary corrections. Once the corrections are complete, the user makes a final check and sends the reply email to the server. The server then sends this reply email to the recipient and saves the sending history.

[0291] Specific examples

[0292] For example, when a user receives an email message saying "Please confirm the date of the next meeting," the process proceeds as follows:

[0293] The server receives the email and analyzes its contents. At the same time, the emotion engine recognizes the user's emotional state and determines that they are in a "high stress state." Based on this, the AI ​​model generates a reply such as, "How about next Tuesday at 3:00 PM? Please let me know if other dates are also convenient."

[0294] The terminal embeds this reply content in a template and presents it to the user. The user then modifies it to "Can you make it 4:00 PM next Tuesday instead of 3:00 PM?" and presses the confirmation button. The server then sends the modified reply content to the other party and saves it as a transmission history.

[0295] Prompt Sentence Examples

[0296] To generate a reply to the email "Please confirm the date of the next meeting," the prompt would look like this:

[0297] The email read: "Please confirm the date of our next meeting."

[0298] User Emotion: "High Stress"

[0299] Example reply: "How about next Tuesday? Tuesday at 3:00 PM? We can also arrange other dates, so please let us know if that's convenient for you."

[0300] This allows the system to reply to emails quickly and efficiently, and also enables responses that take into account the user's feelings, improving the quality of communication.

[0301] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0302] Program processing flow

[0303] Step 1: Receive new email

[0304] Input: User's email account information

[0305] What happens: The server connects to the user's email account (for example, a mail server using the IMAP or POP3 protocol).

[0306] Data processing / calculation: Periodically check for new emails.

[0307] Output: Receives any new mail and adds it to the processing queue.

[0308] Step 2: Email Content Analysis

[0309] Input: New email content

[0310] Specific behavior: Obtains the contents of emails received by the server.

[0311] Data processing / calculation: Analyze email content using natural language processing (NLP) techniques, using NLP models such as "Google BERT" and "OpenAI GPT."

[0312] Output: The intent and important information of the email is extracted.

[0313] Step 3: Recognize emotions

[0314] Input: Parsed email content

[0315] Specific operation: The server launches an emotion recognition engine (e.g., "IBM Watson Tone Analyzer").

[0316] Data processing / calculation: Recognize the user's emotional state based on the text of the email and past email history.

[0317] Output: User emotion data (e.g. stress, joy, anger, etc.).

[0318] Step 4: Anticipate replies

[0319] Input: Analysis results and emotion data

[0320] Specific operation: The server generates the reply content using a generative AI model (e.g., "OpenAI GPT-3").

[0321] Data processing / calculation: Predicting the appropriate tone and content of a reply based on analysis results and sentiment data.

[0322] Output: The predicted reply.

[0323] Step 5: Generate a reply template

[0324] Input: Predicted reply

[0325] What happens: The server embeds the predicted reply content into a standard template.

[0326] Data processing / calculation: Assembling the reply content into a predetermined format.

[0327] Output: A templated reply email.

[0328] Step 6: User review and editing

[0329] Input: Templated reply email

[0330] Specific behavior: The device displays the templated reply email to the user.

[0331] Data processing / calculation: The user checks the reply content on the terminal and edits it as necessary. Editing is done using the interface.

[0332] Output: The final reply that the user reviewed and edited.

[0333] Step 7: Sending an email

[0334] Input: The last reply the user confirmed

[0335] Specific action: The server receives the final confirmed reply email.

[0336] Data processing / calculation: Sends email to the recipient using the SMTP protocol.

[0337] Output: Sent emails and their sending history are recorded.

[0338] Based on the above steps, the system automatically and efficiently analyzes and replies to emails, allowing users to reply quickly and achieve appropriate communication according to their emotional state.

[0339] (Application example 2)

[0340] 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."

[0341] Currently, there are many systems that automate email replies, but they lack the ability to generate replies and content that take the user's emotional state into account. This makes it difficult to respond appropriately to the user's emotions, and improving the quality of communication is a challenge.

[0342] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0343] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content based on the analyzed content, means for identifying the user's emotional state using emotion analysis technology, means for generating optimal content based on the user's emotional state, means for automatically embedding the content in a template, means for presenting the template to the user, and means for sending reply content edited and confirmed by the user, thereby enabling the generation of appropriate replies and content that take the user's emotional state into consideration.

[0344] "New email" refers to new, unread email that has arrived in a user's email account.

[0345] "Natural language processing technology" refers to technology that allows computers to understand, analyze, and generate human language.

[0346] "Emotion analysis technology" is a technology that identifies and analyzes a user's emotional state from text and voice data.

[0347] The "reply content" is a sentence that indicates an appropriate answer or response to the received email.

[0348] A "template" is a model for automatically filling in content based on a defined format or structure.

[0349] "Content" refers to the information and entertainment content provided to users.

[0350] A "generative AI model" refers to an algorithm or program that uses artificial intelligence technology to generate new information or text from data.

[0351] "User interface" refers to the screens and operating means that allow users to interact with the system.

[0352] "Emotional state" refers to the mental or emotional state a user is in at a particular moment.

[0353] An "information processing system" is a computer system that collects, analyzes, generates, and communicates data.

[0354] This invention provides a specific implementation method for an emotion-aware media recommendation system, which includes processes for receiving new emails, analyzing their contents, predicting and generating reply messages, and analyzing the user's emotional state to provide optimal content.

[0355] Hardware and Software Configuration

[0356] 1. Hardware:

[0357] Smartphone

[0358] Smart Glasses

[0359] head-mounted display

[0360] 2. Software:

[0361] Python

[0362] TextBlob library (for natural language processing techniques)

[0363] OpenAI API (for generative AI models)

[0364] Processing Details

[0365] The server first retrieves new emails. The server periodically checks the user's email account and receives newly arrived unread emails. These new emails are added to a processing queue and their contents are analyzed using natural language processing techniques.

[0366] The analyzed content extracts the email's subject, intent, and key information, and then uses sentiment analysis technology to identify the user's emotional state, which is then classified as positive, neutral, or negative.

[0367] Using a generative AI model, the system generates replies and content appropriate to the user's emotional state based on the analysis results and data from emotion analysis technology. For example, if a user types "I'm a little tired" and TextBlob identifies this as a negative emotion, the system will use the OpenAI API to recommend relaxing movies.

[0368] The generated reply and content are automatically embedded in a template and presented to the user, who can then view the template displayed on their smartphone, smart glasses, or head-mounted display and edit it as needed.

[0369] Finally, the reply or content that the user has confirmed and edited is sent to the other party, allowing the user to provide the most appropriate reply or content that suits their current emotional state.

[0370] Specific examples

[0371] When a user receives an email saying "Please confirm the date of your next meeting," the process goes like this:

[0372] Server operation: The server receives the email, analyzes its content, and extracts the intent of "confirming the meeting date." At the same time, emotion analysis technology identifies the user's current emotional state (e.g., high stress). Based on this, the generative AI model predicts the reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[0373] What happens on the device: The device automatically embeds the above reply into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[0374] User action: The user reviews the revised reply, gives a final confirmation, and then sends the email. This process records the sending history.

[0375] Prompt Sentence Examples

[0376] "Can you recommend a relaxing movie?"

[0377] In this way, by using this system, users can receive content that is optimal for their emotional state, thereby improving the quality of communication.

[0378] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0379] Step 1:

[0380] The server periodically checks the user's mail account to retrieve new emails.

[0381] Input: User account information on the mail server

[0382] Output: New emails received

[0383] What happens: The server accesses the user's mailbox and checks if there is new mail. If there is new mail, it adds it to the processing queue.

[0384] Step 2:

[0385] The server analyzes the content of newly arrived emails using natural language processing technology.

[0386] Input: Text data of newly arrived email

[0387] Output: Important information and intent of the email (e.g., meeting schedule confirmation)

[0388] What it does: The server uses a natural language processing library such as TextBlob to extract key keywords and intent from the text of the email.

[0389] Step 3:

[0390] The server uses emotion analysis techniques to identify the user's emotional state.

[0391] Input: Text data such as user input comments and emails

[0392] Output: User's emotional state (e.g. positive, neutral, negative)

[0393] What it does: The server uses TextBlob to calculate the emotional polarity of the input text and identifies the user's emotional state based on that.

[0394] Step 4:

[0395] Generative AI models are used to generate content based on analysis results and emotional data.

[0396] Input: Email analysis results, user emotional state

[0397] Output: Content recommendation or reply

[0398] How it works: The server uses the OpenAI API to send a prompt (e.g., "Tell me a relaxing movie") and have the AI ​​model generate appropriate content.

[0399] Step 5:

[0400] The server automatically embeds the generated content into a template and sends it to the terminal.

[0401] Input: Generated content or reply content

[0402] Output: Content embedded in the template

[0403] Specific operation: The server embeds the generated content into a predefined template and sends the template to the terminal.

[0404] Step 6:

[0405] The terminal displays the template to the user, who then checks and edits it.

[0406] Input: Content embedded in the template

[0407] Output: Reply edited by the user

[0408] Specific behavior: The device displays the template in the user interface, allowing the user to review the content and make any necessary modifications.

[0409] Step 7:

[0410] The reply content that the user has confirmed and edited is sent to the server and then transmitted to the other party.

[0411] Input: Reply content confirmed and edited by the user

[0412] Output: Reply email sent to the recipient

[0413] Specific operation: After the user makes a final confirmation, the reply is sent to the server, which then sends it to the other party and records the sending history.

[0414] These processing steps allow users to receive replies and content that are efficient, emotionally relevant, and relevant.

[0415] 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.

[0416] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0417] 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.

[0418] [Second embodiment]

[0419] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0420] 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.

[0421] 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).

[0422] 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.

[0423] 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.

[0424] 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).

[0425] 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.

[0426] 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.

[0427] 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.

[0428] 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.

[0429] In the smart glasses 214, 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.

[0430] 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."

[0431] The present invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[0432] composition

[0433] The system has the following main functions:

[0434] 1. Receiving new emails

[0435] The server periodically checks the user's mail account to retrieve new emails, eliminating the need for the user to manually check their mail.

[0436] 2. Email Content Analysis

[0437] The server analyzes the content of the email using natural language processing technology, which extracts the email's intent and important information.

[0438] 3. Predicting replies

[0439] The server uses AI technology to predict the reply content based on the analyzed email content, using data from past emails and business context.

[0440] 4. Generate a reply template

[0441] The server embeds the predicted reply content into a predefined template and presents the completed reply email to the user.

[0442] 5. User review and editing

[0443] The terminal displays the template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[0444] 6. Sending emails

[0445] The server sends a reply email that the user finally checks and corrects.

[0446] Operation details

[0447] 1. Server Operation

[0448] The server connects to the user's email account and checks for new emails. If there are new emails, they are received and added to a processing queue for analysis. Next, natural language processing techniques are used to analyze the email's content and extract its intent (e.g., confirming a meeting date). Based on the results of this analysis, the server uses an AI model to predict an appropriate reply (e.g., "How about next Tuesday?").

[0449] 2. Device Operation

[0450] The terminal receives the predicted reply content from the server, embeds it in a template, and presents it to the user. An interface is provided for the user to view and edit the reply template.

[0451] 3. User Actions

[0452] The user reviews the proposed reply and edits it if necessary, and the final, reviewed reply is sent to the server, which then sends it to the email recipient.

[0453] Specific examples

[0454] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[0455] Server operation: The server receives the email and analyzes its contents. As a result of the analysis, the intent of "confirming the meeting date" is extracted. Based on this intent, the AI ​​model predicts the reply content, such as "How about next Tuesday?" This is then embedded in a template and presented to the user.

[0456] Terminal behavior: The terminal displays a template ("Dear [Recipient],\n\nHow about next Tuesday?\n\nBest regards,\n[Your Name]") to the user. The user edits this template, for example, to "How about next Tuesday at 3:00 PM?"

[0457] User action: User reviews the modified reply and sends it to the server, which finally sends this email.

[0458] In this way, the system of the present invention makes replying to e-mails more efficient and improves user productivity.

[0459] The processing flow will be explained below.

[0460] Step 1:

[0461] Receive new emails

[0462] The server periodically checks the user's mail account to see if there is any new email. If new mail is found, it is retrieved from the mail server and added to a processing queue on the server.

[0463] Step 2:

[0464] Email content analysis

[0465] The server analyzes the body of newly arrived emails using natural language processing technology. Specifically, the body of the email is fed into an NLP (natural language processing) model to extract the email's intent and important information. The results of this analysis are used as input for subsequent processing.

[0466] Step 3:

[0467] Response prediction

[0468] The server then uses an AI model to predict the appropriate reply based on the analyzed email content. The AI ​​model incorporates data from past emails and business context, resulting in highly accurate replies.

[0469] Step 4:

[0470] Generate a reply template

[0471] The server embeds the predicted reply content into a standard template, which has fixed parts (e.g., greetings and signatures) and variable parts (reply content generated by AI), resulting in consistent reply emails.

[0472] Step 5:

[0473] Show reply content to user

[0474] The terminal displays the generated reply template to the user, and the user interface provides a form that allows the user to review and edit the reply content, allowing the user to easily review and modify the content.

[0475] Step 6:

[0476] User review and editing

[0477] The user checks the email reply template using the terminal. If necessary, the user edits the content of the template and performs a final check. Once editing is complete, the user clicks the send button to finalize the reply email.

[0478] Step 7:

[0479] Sending emails

[0480] The server receives the reply email that has been finally confirmed by the user. It then sends it to the recipient's email address via the mail server. Once the sending is complete, it is recorded in the sending history.

[0481] Through the above series of processing steps, the user can quickly and efficiently reply to e-mails, allowing the user to concentrate on other important tasks.

[0482] Example 1

[0483] 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."

[0484] In conventional email processing systems, users had to manually check new emails, analyze their content, and think up a reply, which required a great deal of time and effort. Particularly in busy work situations where quick and accurate responses are required, these manual tasks significantly reduced user productivity.

[0485] 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.

[0486] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content using AI technology based on the analyzed content, means for automatically embedding the predicted reply content into a template, means for presenting the template to the user and allowing the user to edit and confirm it, and means for sending the reply content edited and confirmed by the user. This enables the user to automate a series of processes from checking newly arrived emails, generating reply content, embedding it into a template, and sending the final email, quickly and efficiently.

[0487] "New email" is email that has just been received and that the user has not yet opened.

[0488] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes text tokenization, keyword extraction, and intent analysis.

[0489] "AI technology" is a technology that uses artificial intelligence to analyze data and generate appropriate answers and predictions.

[0490] "Means for predicting reply content" refers to functions and algorithms for generating appropriate reply content based on the content of received email.

[0491] A "template" is a model of a document created according to a specific format, and is designed to be easy for a user to edit.

[0492] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task (e.g., generating email replies).

[0493] A "prompt" is text data input to a generative AI model that provides instructions and context for the model to generate a reply.

[0494] "Automation" refers to the process by which machines or software automatically perform tasks that humans would otherwise perform manually.

[0495] "User review and editing" refers to the process in which the user reviews the email reply content generated by the system and makes corrections as necessary.

[0496] This invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[0497] Server behavior:

[0498] The server periodically connects to the user's email account and checks for new emails. If new emails are available, the server receives them using the IMAP or POP3 protocol. The server then analyzes the email content using natural language processing techniques (e.g., Python's NLTK or SpaCy). This analysis extracts the email's intent and important keywords. Next, the server uses a generative AI model (e.g., GPT-4) to predict an appropriate reply based on the analyzed content.

[0499] As a concrete example, if an email is received saying "Please confirm the date of the next meeting," the server analyzes the content of the email and extracts the intent of "confirm the date of the meeting." Based on that intent, the server inputs the prompt "Please confirm the date of the next meeting" into the generative AI model and obtains the reply text (for example, "How about next Tuesday?") obtained from the AI ​​model.

[0500] Terminal behavior:

[0501] The server sends the predicted reply content to the terminal to present to the user. The terminal embeds the predicted reply content in a template and displays it to the user. For example, this template has the format "Dear [Recipient],\n\n[How about next Tuesday?]\n\nBest regards,\n[Your Name]". The user checks the template presented to them through the terminal and edits the content as necessary. Once editing is complete, the user checks the content and finally sends it to the server.

[0502] User Action:

[0503] The user creates the final reply based on the template that has been checked and edited on the terminal. If any corrections are necessary, the template "How about next Tuesday?" can be edited to "How about next Tuesday at 3:00 PM?". Once editing is complete, the user sends the reply to the server. The server uses the SMTP protocol to send the final email to the recipient.

[0504] In this way, users can efficiently complete a series of automated processes, from checking new emails to creating, editing, and sending replies. This significantly reduces the time and effort required for users to process emails, improving productivity. This system ensures speed and accuracy in responding to emails, contributing to work efficiency.

[0505] As a concrete example, the following prompt sentences are fed into a generative AI model to generate a reply:

[0506] "Generate an appropriate response to an email asking to confirm the next meeting date."

[0507] Please use this information to create a polite business reply email.

[0508] This allows the generative AI model to automatically generate appropriate reply content.

[0509] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0510] Step 1: Check for new emails

[0511] The server periodically connects to the user's email account and checks for new email. The server connects to the mail server using the IMAP or POP3 protocol and checks for new email. If new email is available, the server accepts it and proceeds to the next step.

[0512] Specific behavior:

[0513] Input: Email account information, IMAP / POP3 protocol

[0514] Data processing: Query the mail server and check for new mail.

[0515] Output: List of new emails (email ID, sender, subject, body)

[0516] Step 2: Analyzing the email content

[0517] The server acquires the content of newly received emails and analyzes them using natural language processing technology. Tools such as NLTK and SpaCy are used for the analysis. This analysis extracts the intent of the email and important keywords.

[0518] Specific behavior:

[0519] Input: Body of new email

[0520] Data processing: Using natural language processing tools to tokenize the email body and extract key keywords and intent

[0521] Output: Analysis results (intention, keywords)

[0522] Step 3: Anticipate replies

[0523] Based on the parsed content, the server inputs prompts into a generative AI model to predict appropriate responses, taking into account past email data and business context in the process.

[0524] Specific behavior:

[0525] Input: Analysis results (intention, keywords), prompt

[0526] Data calculation: Enter a prompt into the generative AI model and generate a reply.

[0527] Output: Predicted reply

[0528] Step 4: Generate a reply template

[0529] The server generates a reply template by embedding the predicted reply content into a predefined template, which is then generated in a format that is easy for the user to edit.

[0530] Specific behavior:

[0531] Input: predicted reply

[0532] Data processing: Embed reply content in a template

[0533] Output: Reply template (e.g., "Dear [Recipient],\n\n[Reply]\n\nBest regards,\n[Your Name]")

[0534] Step 5: User review and editing

[0535] The terminal presents the reply template received from the server to the user, who is then provided with an interface to review the template and edit its contents as necessary.

[0536] Specific behavior:

[0537] Input: Reply template

[0538] Data processing: Displaying a template on the user interface

[0539] Output: User edits

[0540] Step 6: Sending the final email

[0541] The server receives the final reply, edited and reviewed by the user, and sends it to the email recipient using the SMTP protocol.

[0542] Specific behavior:

[0543] Input: Reply confirmed by the user

[0544] Data Processing: Sending email using the SMTP protocol

[0545] Output: Sending result log

[0546] In this way, all steps are automated, allowing users to reply to emails quickly and accurately with minimal effort, improving user productivity and significantly increasing work efficiency.

[0547] (Application example 1)

[0548] 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."

[0549] For companies and organizations, responding to security-related emails is extremely important, but manual response requires a great deal of time and effort. Furthermore, it is not easy to respond appropriately and promptly to security alerts, which require a rapid response. Therefore, there is a need for a system that automates the entire process of receiving emails, analyzing them, generating replies, and sending them, thereby streamlining responses to security-related emails in particular.

[0550] 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.

[0551] In this invention, the server includes a means for acquiring newly arrived emails, a means for analyzing the content of the emails using natural language processing technology, a means for analyzing security-related emails and generating reply content, a means for verifying that the generated reply content is in accordance with the security policy, and a means for providing an interface designed to enable a prompt response to security alerts, thereby enabling prompt and appropriate responses to security-related emails in companies and organizations.

[0552] "Newly arrived email" refers to email with the most recent received date and time.

[0553] "Natural language processing technology" refers to computer technology for analyzing and understanding human language.

[0554] "Reply content prediction" refers to the process of estimating what should be replied to based on the content of the received email.

[0555] "Template embedding" refers to incorporating predicted reply content into a predetermined format or framework.

[0556] "Security-related email" refers to emails that contain content related to security threats such as phishing, malware, and unauthorized access.

[0557] A "security policy" refers to a guideline that describes the security measures and response policies established by an organization.

[0558] "Interface" refers to the screen and operation method used by users to operate the system.

[0559] A "generative AI model" refers to an artificial intelligence model that uses machine learning to generate information for specific tasks.

[0560] A "prompt" is a piece of text that describes instructions or questions to be input into a generative AI model.

[0561] A "server" refers to a computer system that performs a series of processes such as receiving, analyzing, generating a reply, and sending an email.

[0562] To implement this invention, it is necessary to build a system that automates security-related email correspondence. This system has the following main functions:

[0563] First, the server connects to the company or organization's mail server and retrieves new emails. This is done using the IMAP protocol. When new emails are received, their contents are stored in a database. The software used in this step is an email sending and receiving library (e.g., IMAP, SMTP).

[0564] Next, the server analyzes the content of the received email using natural language processing technology. This analysis uses natural language processing libraries such as spaCy and NLTK. As a result of the analysis, information particularly related to security is extracted. For example, keywords such as "phishing," "malware," and "anomaly detection" are identified.

[0565] The server then predicts the reply content based on the analysis results. This is where an AI model (e.g., GPT-4) is used. The AI ​​model references past data and security policies to generate the optimal reply content. This simultaneously verifies that the generated reply content complies with the security policy.

[0566] The generated reply is embedded in a pre-prepared template and presented to the user. The template uses the following prompt:

[0567] "Generate an appropriate response to the following phishing report email: Subject: Phishing Report Content: User reporting phishing activity Example of expected response: "Dear [reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter.""

[0568] The terminal displays this template to the user and provides an interface that allows the user to view and edit the content. This interface is designed to allow the user to select appropriate actions based on security policies.

[0569] After the user has finished editing and checking, the reply is finally sent to the server, which then sends it to the recipient using the SMTP protocol.

[0570] Through the above process, this system will enable companies and organizations to streamline security-related email responses and achieve prompt and appropriate responses.

[0571] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0572] Step 1:

[0573] The server connects to the mail server of the company or organization using the IMAP protocol to receive new emails and stores them in a database.

[0574] Input: New email from the company's email server

[0575] Output: New emails stored in the database

[0576] Specific operation: Connect to the mail server using the IMAP library, retrieve unread emails, and store their contents (body, subject, sender information, etc.) in a database.

[0577] Step 2:

[0578] The server analyzes the content of the stored emails using natural language processing techniques (e.g., spaCy or NLTK), particularly to extract security-related information (e.g., phishing, malware, unauthorized access).

[0579] Input: Contents of new emails stored in the database

[0580] Output: Parsed email content (security-related information)

[0581] Specific operations: Tokenize the email content, extract keywords, and analyze sentence structure to identify security-related information.

[0582] Step 3:

[0583] The server predicts the reply content based on the analysis results. Here, a generative AI model (e.g., GPT-4) is used. The AI ​​model generates the optimal reply content by referring to past data and security policies.

[0584] Input: Parsed email content (security-related information)

[0585] Output: Predicted reply

[0586] Specific operation: The analyzed information is input into the AI ​​model, and prompts are used to generate predicted replies. Example: "Generate an appropriate reply to the following phishing report email: Subject: Phishing Report Content: User's report of phishing activity Example of predicted reply: "Dear [Reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter."

[0587] Step 4:

[0588] The server embeds the generated reply content into a template based on a pre-defined format to create data to present to the user.

[0589] Input: predicted reply

[0590] Output: Reply content embedded in template

[0591] Specific operation: The predicted reply content is automatically embedded in the defined template and converted into a format that can be presented to the user.

[0592] Step 5:

[0593] The terminal displays the reply content embedded in the template to the user, and provides an interface for the user to view and edit the content.

[0594] Input: Reply content embedded in template

[0595] Output: A screen that the user can see and edit

[0596] What it does: Displays a form through the user interface to allow the user to review and edit the reply content embedded in the template, and waits for the user to finish editing.

[0597] Step 6:

[0598] The user then sends the reply content after reviewing and editing it to the server, which receives it and sends the final reply email to the email recipient.

[0599] Input: The reply that the user has reviewed and edited.

[0600] Output: Reply email sent

[0601] Specific behavior: Receives the content that the user has completed editing and sends a final reply email using the SMTP protocol.

[0602] The above steps create a system that automates security-related email responses and allows them to be handled efficiently and quickly.

[0603] 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.

[0604] The present invention relates to a system that automates the entire process from receiving new emails, analyzing them, generating reply content, embedding them in templates, and finally sending them, and combines it with an emotion engine that recognizes the user's emotions and optimizes the reply content.

[0605] composition

[0606] The system has the following main functions:

[0607] 1. Receiving new emails

[0608] The server periodically checks the user's mail account and receives new email, saving the user the trouble of having to manually check their mail.

[0609] 2. Email Content Analysis

[0610] The server analyzes the content of the email received using natural language processing technology to extract the intent of the email and important information.

[0611] 3. Emotional Recognition

[0612] The emotion engine recognizes the user's emotions and analyzes the user's emotional state from the email text and the user's reading.

[0613] 4. Predicting replies

[0614] The server uses AI technology to predict the appropriate reply based on the analysis results and data from the emotion engine. The emotional information from the emotion engine is reflected in the tone and expression of the reply.

[0615] 5. Generate a reply template

[0616] The server embeds the predicted reply content into a standard template and presents the completed reply email to the user, generating a consistent reply document.

[0617] 6. User Review and Editing

[0618] The terminal displays the reply template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[0619] 7. Sending emails

[0620] The server receives the reply email that has been finally confirmed by the user and sends it to the recipient. During this process, the sending history is recorded.

[0621] Operation details

[0622] 1. Server Operation

[0623] The server connects to the user's email account and receives new emails. The received emails are added to a processing queue within the server. Next, natural language processing technology is used to analyze the content of the email and extract its intent and important information. In parallel, an emotion engine recognizes the user's emotions and processes them by combining them with the analysis results.

[0624] 2. Device Operation

[0625] Based on the analysis data and emotion data received by the device from the server, the AI ​​model generates an appropriate reply. The generated reply is automatically embedded in a template and presented to the user. The user can then review the reply through the device interface and edit it as necessary.

[0626] 3. User Actions

[0627] The user reviews the reply displayed on the device, especially the tone and expressions based on emotions, and edits the reply if necessary. Once the final review is complete, the reply email is sent.

[0628] Specific examples

[0629] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[0630] Server operation: The server receives the email, analyzes its contents, and extracts the intent of "confirming the meeting date." At the same time, the emotion engine recognizes the user's current emotional state (e.g., high stress). Based on this, the AI ​​model predicts the appropriate reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[0631] What happens on the device: The device embeds the above reply content into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[0632] User action: The user checks the revised reply and gives a final confirmation before sending it to the server, which then sends the reply to the recipient.

[0633] This system allows users to reply to e-mails quickly and efficiently, and improves the quality of communication by taking emotional information into account when responding.

[0634] The processing flow will be explained below.

[0635] Step 1:

[0636] Receive new emails

[0637] The server periodically connects to the user's mail account to check for new emails. If new emails are found, they are retrieved from the mail server and added to a processing queue on the server.

[0638] Step 2:

[0639] Email content analysis

[0640] The server analyzes the body of the newly received email using natural language processing technology. During the analysis, the intent and important information are extracted from the email content. This data becomes the input information required for subsequent processing.

[0641] Step 3:

[0642] Emotion recognition

[0643] The emotion engine installed on the server collects the user's emotional data, analyzes the current emotional state (e.g., stress, joy, anger, etc.) from the text of the email, the user's past behavioral history, reading, etc., and generates quantified emotional data.

[0644] Step 4:

[0645] Response prediction

[0646] The server uses an AI model to predict an appropriate reply based on the analysis data obtained through natural language processing and the emotional data recognized by the emotion engine. For example, if the user is feeling stressed in response to an email asking, "Please confirm the date of the next meeting," the server will generate a gentler reply such as, "How about next Tuesday? We can also arrange other dates, so please let us know if that's convenient for you."

[0647] Step 5:

[0648] Generate a reply template

[0649] The server embeds predicted replies into standard templates, which have fixed parts (such as a greeting or signature) and variable parts (the predicted replies), resulting in consistent replies.

[0650] Step 6:

[0651] Show reply content to user

[0652] The terminal displays the generated reply template to the user, and the user interface provides a form for reviewing and editing the reply content, allowing the user to easily modify the content.

[0653] Step 7:

[0654] User review and editing

[0655] The user uses the device to check the displayed reply template. The user reviews the content and edits the template as needed. For example, the user can add a specific date, such as "How about 3:00 PM next Tuesday?"

[0656] Step 8:

[0657] Sending emails

[0658] The user finally sends the reply email that has been confirmed and edited to the server. The server receives and confirms this reply email, and sends it to the recipient's email address. Once the sending is complete, the history is recorded.

[0659] This allows users to reply to emails efficiently and appropriately according to their emotions, which not only improves work productivity but also the quality of communication.

[0660] Example 2

[0661] 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."

[0662] Conventional email reply systems require a lot of time and effort for manually checking emails and creating replies, making it difficult to reply quickly. Furthermore, the system generates a uniform reply without considering the user's feelings, which results in a decline in the quality of communication. The objective of this invention is to solve these problems and provide a system that realizes efficient, emotionally appropriate automatic replies.

[0663] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0664] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for recognizing the user's emotions using an emotion recognition engine based on the analyzed content, means for predicting reply content based on the analysis result and emotion data using a generative AI model, means for embedding the predicted reply content in a template, means for presenting the template to the user, and means for sending the reply content edited and confirmed by the user. This enables fast and efficient email replies and realizes high-quality communication that takes the user's emotions into consideration.

[0665] The "means for obtaining new e-mails" is a function for periodically accessing the user's e-mail account, checking whether new e-mails have arrived, and receiving them.

[0666] "Natural language processing technology" is a set of technologies that allow computers to understand, analyze, and generate human language, and is a language analysis technology used to understand the content of emails.

[0667] An "emotion recognition engine" is a system for recognizing a user's emotional state, and is a technology that estimates a user's emotions based on emails and past data.

[0668] A "generative AI model" is a model that uses artificial intelligence techniques to generate appropriate responses or documents based on given data or information.

[0669] "Analysis results" refers to information obtained when analyzing the content of an email using natural language processing technology, and includes the intent and important information.

[0670] "Emotion data" is information obtained as a result of analyzing the user's emotional state using an emotion recognition engine.

[0671] "Means for predicting reply content" is a function that uses a generative AI model to generate appropriate reply content based on the analysis results and emotional data.

[0672] The "means for embedding in a template" is a function for incorporating predicted reply content into a standard format prepared in advance.

[0673] The "means for presenting to the user" is a function for displaying the reply content embedded in the template to the user on the terminal.

[0674] The "means for sending reply content edited and confirmed by the user" is a function for sending reply content that the user has confirmed and edited to the recipient, and is necessary to complete the email sending process.

[0675] This invention relates to a system that automates the entire process from receiving new e-mails, analyzing their contents, generating reply content, embedding them in templates, and finally sending them, and further optimizes the reply content by recognizing the user's emotions. This system operates in cooperation with a server, a terminal, and a user.

[0676] Server Operation

[0677] The server periodically connects to the user's mail account to receive new emails using the IMAP or POP3 protocols, which are then added to a processing queue on the server.

[0678] The server then analyzes the email content using natural language processing (NLP) techniques, such as Google BERT or OpenAI GPT. The server then uses NLP techniques to extract the main intent and important information from the email.

[0679] Furthermore, an emotion recognition engine is activated to recognize the user's emotions. This emotion recognition is performed using tools such as the IBM Watson Tone Analyzer. The analyzed emotion data is stored on the server and used to generate subsequent reply content.

[0680] The server uses a generative AI model (such as OpenAI GPT-3) to predict an appropriate reply based on the analysis results and emotional data. The reply is generated in a tone that reflects the user's emotions.

[0681] The predicted reply content is embedded in a standard template, which has a predefined format based on which a reply document is automatically generated.

[0682] Device behavior

[0683] The terminal displays the templated reply content received from the server to the user. The user can check the reply content on the terminal and edit it as necessary. The interface is user-friendly, allowing for easy editing.

[0684] User Actions

[0685] The user checks the reply displayed on their device, particularly the tone and expressions reflected by the emotion engine, and makes any necessary corrections. Once the corrections are complete, the user makes a final check and sends the reply email to the server. The server then sends this reply email to the recipient and saves the sending history.

[0686] Specific examples

[0687] For example, when a user receives an email message saying "Please confirm the date of the next meeting," the process proceeds as follows:

[0688] The server receives the email and analyzes its contents. At the same time, the emotion engine recognizes the user's emotional state and determines that they are in a "high stress state." Based on this, the AI ​​model generates a reply such as, "How about next Tuesday at 3:00 PM? Please let me know if other dates are also convenient."

[0689] The terminal embeds this reply content in a template and presents it to the user. The user then modifies it to "Can you make it 4:00 PM next Tuesday instead of 3:00 PM?" and presses the confirmation button. The server then sends the modified reply content to the other party and saves it as a transmission history.

[0690] Prompt Sentence Examples

[0691] To generate a reply to the email "Please confirm the date of the next meeting," the prompt would look like this:

[0692] The email read: "Please confirm the date of our next meeting."

[0693] User Emotion: "High Stress"

[0694] Example reply: "How about next Tuesday? Tuesday at 3:00 PM? We can also arrange other dates, so please let us know if that's convenient for you."

[0695] This allows the system to reply to emails quickly and efficiently, and also enables responses that take into account the user's feelings, improving the quality of communication.

[0696] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0697] Program processing flow

[0698] Step 1: Receive new email

[0699] Input: User's email account information

[0700] What happens: The server connects to the user's email account (for example, a mail server using the IMAP or POP3 protocol).

[0701] Data processing / calculation: Periodically check for new emails.

[0702] Output: Receives any new mail and adds it to the processing queue.

[0703] Step 2: Email Content Analysis

[0704] Input: New email content

[0705] Specific behavior: Obtains the contents of emails received by the server.

[0706] Data processing / calculation: Analyze email content using natural language processing (NLP) techniques, using NLP models such as "Google BERT" and "OpenAI GPT."

[0707] Output: The intent and important information of the email is extracted.

[0708] Step 3: Recognize emotions

[0709] Input: Parsed email content

[0710] Specific operation: The server launches an emotion recognition engine (e.g., "IBM Watson Tone Analyzer").

[0711] Data processing / calculation: Recognize the user's emotional state based on the text of the email and past email history.

[0712] Output: User emotion data (e.g. stress, joy, anger, etc.).

[0713] Step 4: Anticipate replies

[0714] Input: Analysis results and emotion data

[0715] Specific operation: The server generates the reply content using a generative AI model (e.g., "OpenAI GPT-3").

[0716] Data processing / calculation: Predicting the appropriate tone and content of a reply based on analysis results and sentiment data.

[0717] Output: The predicted reply.

[0718] Step 5: Generate a reply template

[0719] Input: Predicted reply

[0720] What happens: The server embeds the predicted reply content into a standard template.

[0721] Data processing / calculation: Assembling the reply content into a predetermined format.

[0722] Output: A templated reply email.

[0723] Step 6: User review and editing

[0724] Input: Templated reply email

[0725] Specific behavior: The device displays the templated reply email to the user.

[0726] Data processing / calculation: The user checks the reply content on the terminal and edits it as necessary. Editing is done using the interface.

[0727] Output: The final reply that the user reviewed and edited.

[0728] Step 7: Sending an email

[0729] Input: The last reply the user confirmed

[0730] Specific action: The server receives the final confirmed reply email.

[0731] Data processing / calculation: Sends email to the recipient using the SMTP protocol.

[0732] Output: Sent emails and their sending history are recorded.

[0733] Based on the above steps, the system automatically and efficiently analyzes and replies to emails, allowing users to reply quickly and achieve appropriate communication according to their emotional state.

[0734] (Application example 2)

[0735] 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."

[0736] Currently, there are many systems that automate email replies, but they lack the ability to generate replies and content that take the user's emotional state into account. This makes it difficult to respond appropriately to the user's emotions, and improving the quality of communication is a challenge.

[0737] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0738] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content based on the analyzed content, means for identifying the user's emotional state using emotion analysis technology, means for generating optimal content based on the user's emotional state, means for automatically embedding the content in a template, means for presenting the template to the user, and means for sending reply content edited and confirmed by the user, thereby enabling the generation of appropriate replies and content that take the user's emotional state into consideration.

[0739] "New email" refers to new, unread email that has arrived in a user's email account.

[0740] "Natural language processing technology" refers to technology that allows computers to understand, analyze, and generate human language.

[0741] "Emotion analysis technology" is a technology that identifies and analyzes a user's emotional state from text and voice data.

[0742] The "reply content" is a sentence that indicates an appropriate answer or response to the received email.

[0743] A "template" is a model for automatically filling in content based on a defined format or structure.

[0744] "Content" refers to the information and entertainment content provided to users.

[0745] A "generative AI model" refers to an algorithm or program that uses artificial intelligence technology to generate new information or text from data.

[0746] "User interface" refers to the screens and operating means that allow users to interact with the system.

[0747] "Emotional state" refers to the mental or emotional state a user is in at a particular moment.

[0748] An "information processing system" is a computer system that collects, analyzes, generates, and communicates data.

[0749] This invention provides a specific implementation method for an emotion-aware media recommendation system, which includes processes for receiving new emails, analyzing their contents, predicting and generating reply messages, and analyzing the user's emotional state to provide optimal content.

[0750] Hardware and Software Configuration

[0751] 1. Hardware:

[0752] Smartphone

[0753] Smart Glasses

[0754] head-mounted display

[0755] 2. Software:

[0756] Python

[0757] TextBlob library (for natural language processing techniques)

[0758] OpenAI API (for generative AI models)

[0759] Processing Details

[0760] The server first retrieves new emails. The server periodically checks the user's email account and receives newly arrived unread emails. These new emails are added to a processing queue and their contents are analyzed using natural language processing techniques.

[0761] The analyzed content extracts the email's subject, intent, and key information, and then uses sentiment analysis technology to identify the user's emotional state, which is then classified as positive, neutral, or negative.

[0762] Using a generative AI model, the system generates replies and content appropriate to the user's emotional state based on the analysis results and data from emotion analysis technology. For example, if a user types "I'm a little tired" and TextBlob identifies this as a negative emotion, the system will use the OpenAI API to recommend relaxing movies.

[0763] The generated reply and content are automatically embedded in a template and presented to the user, who can then view the template displayed on their smartphone, smart glasses, or head-mounted display and edit it as needed.

[0764] Finally, the reply or content that the user has confirmed and edited is sent to the other party, allowing the user to provide the most appropriate reply or content that suits their current emotional state.

[0765] Specific examples

[0766] When a user receives an email saying "Please confirm the date of your next meeting," the process goes like this:

[0767] Server operation: The server receives the email, analyzes its content, and extracts the intent of "confirming the meeting date." At the same time, emotion analysis technology identifies the user's current emotional state (e.g., high stress). Based on this, the generative AI model predicts the reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[0768] What happens on the device: The device automatically embeds the above reply into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[0769] User action: The user reviews the revised reply, gives a final confirmation, and then sends the email. This process records the sending history.

[0770] Prompt Sentence Examples

[0771] "Can you recommend a relaxing movie?"

[0772] In this way, by using this system, users can receive content that is optimal for their emotional state, thereby improving the quality of communication.

[0773] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0774] Step 1:

[0775] The server periodically checks the user's mail account to retrieve new emails.

[0776] Input: User account information on the mail server

[0777] Output: New emails received

[0778] What happens: The server accesses the user's mailbox and checks if there is new mail. If there is new mail, it adds it to the processing queue.

[0779] Step 2:

[0780] The server analyzes the content of newly arrived emails using natural language processing technology.

[0781] Input: Text data of newly arrived email

[0782] Output: Important information and intent of the email (e.g., meeting schedule confirmation)

[0783] What it does: The server uses a natural language processing library such as TextBlob to extract key keywords and intent from the text of the email.

[0784] Step 3:

[0785] The server uses emotion analysis techniques to identify the user's emotional state.

[0786] Input: Text data such as user input comments and emails

[0787] Output: User's emotional state (e.g. positive, neutral, negative)

[0788] What it does: The server uses TextBlob to calculate the emotional polarity of the input text and identifies the user's emotional state based on that.

[0789] Step 4:

[0790] Generative AI models are used to generate content based on analysis results and emotional data.

[0791] Input: Email analysis results, user emotional state

[0792] Output: Content recommendation or reply

[0793] How it works: The server uses the OpenAI API to send a prompt (e.g., "Tell me a relaxing movie") and have the AI ​​model generate appropriate content.

[0794] Step 5:

[0795] The server automatically embeds the generated content into a template and sends it to the terminal.

[0796] Input: Generated content or reply content

[0797] Output: Content embedded in the template

[0798] Specific operation: The server embeds the generated content into a predefined template and sends the template to the terminal.

[0799] Step 6:

[0800] The terminal displays the template to the user, who then checks and edits it.

[0801] Input: Content embedded in the template

[0802] Output: Reply edited by the user

[0803] Specific behavior: The device displays the template in the user interface, allowing the user to review the content and make any necessary modifications.

[0804] Step 7:

[0805] The reply content that the user has confirmed and edited is sent to the server and then transmitted to the other party.

[0806] Input: Reply content confirmed and edited by the user

[0807] Output: Reply email sent to the recipient

[0808] Specific operation: After the user makes a final confirmation, the reply is sent to the server, which then sends it to the other party and records the sending history.

[0809] These processing steps allow users to receive replies and content that are efficient, emotionally relevant, and relevant.

[0810] 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.

[0811] 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.

[0812] 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.

[0813] [Third embodiment]

[0814] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0815] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0816] 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).

[0817] 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.

[0818] 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.

[0819] 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).

[0820] 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.

[0821] 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.

[0822] 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.

[0823] 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.

[0824] 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.

[0825] 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."

[0826] The present invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[0827] composition

[0828] The system has the following main functions:

[0829] 1. Receiving new emails

[0830] The server periodically checks the user's mail account to retrieve new emails, eliminating the need for the user to manually check their mail.

[0831] 2. Email Content Analysis

[0832] The server analyzes the content of the email using natural language processing technology, which extracts the email's intent and important information.

[0833] 3. Predicting replies

[0834] The server uses AI technology to predict the reply content based on the analyzed email content, using data from past emails and business context.

[0835] 4. Generate a reply template

[0836] The server embeds the predicted reply content into a predefined template and presents the completed reply email to the user.

[0837] 5. User review and editing

[0838] The terminal displays the template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[0839] 6. Sending emails

[0840] The server sends a reply email that the user finally checks and corrects.

[0841] Operation details

[0842] 1. Server Operation

[0843] The server connects to the user's email account and checks for new emails. If there are new emails, they are received and added to a processing queue for analysis. Next, natural language processing techniques are used to analyze the email's content and extract its intent (e.g., confirming a meeting date). Based on the results of this analysis, the server uses an AI model to predict an appropriate reply (e.g., "How about next Tuesday?").

[0844] 2. Device Operation

[0845] The terminal receives the predicted reply content from the server, embeds it in a template, and presents it to the user. An interface is provided for the user to view and edit the reply template.

[0846] 3. User Actions

[0847] The user reviews the proposed reply and edits it if necessary, and the final, reviewed reply is sent to the server, which then sends it to the email recipient.

[0848] Specific examples

[0849] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[0850] Server operation: The server receives the email and analyzes its contents. As a result of the analysis, the intent of "confirming the meeting date" is extracted. Based on this intent, the AI ​​model predicts the reply content, such as "How about next Tuesday?" This is then embedded in a template and presented to the user.

[0851] Terminal behavior: The terminal displays a template ("Dear [Recipient],\n\nHow about next Tuesday?\n\nBest regards,\n[Your Name]") to the user. The user edits this template, for example, to "How about next Tuesday at 3:00 PM?"

[0852] User action: User reviews the modified reply and sends it to the server, which finally sends this email.

[0853] In this way, the system of the present invention makes replying to e-mails more efficient and improves user productivity.

[0854] The processing flow will be explained below.

[0855] Step 1:

[0856] Receive new emails

[0857] The server periodically checks the user's mail account to see if there is any new email. If new mail is found, it is retrieved from the mail server and added to a processing queue on the server.

[0858] Step 2:

[0859] Email content analysis

[0860] The server analyzes the body of newly arrived emails using natural language processing technology. Specifically, the body of the email is fed into an NLP (natural language processing) model to extract the email's intent and important information. The results of this analysis are used as input for subsequent processing.

[0861] Step 3:

[0862] Response prediction

[0863] The server then uses an AI model to predict the appropriate reply based on the analyzed email content. The AI ​​model incorporates data from past emails and business context, resulting in highly accurate replies.

[0864] Step 4:

[0865] Generate a reply template

[0866] The server embeds the predicted reply content into a standard template, which has fixed parts (e.g., greetings and signatures) and variable parts (reply content generated by AI), resulting in consistent reply emails.

[0867] Step 5:

[0868] Show reply content to user

[0869] The terminal displays the generated reply template to the user, and the user interface provides a form that allows the user to review and edit the reply content, allowing the user to easily review and modify the content.

[0870] Step 6:

[0871] User review and editing

[0872] The user checks the email reply template using the terminal. If necessary, the user edits the content of the template and performs a final check. Once editing is complete, the user clicks the send button to finalize the reply email.

[0873] Step 7:

[0874] Sending emails

[0875] The server receives the reply email that has been finally confirmed by the user. It then sends it to the recipient's email address via the mail server. Once the sending is complete, it is recorded in the sending history.

[0876] Through the above series of processing steps, the user can quickly and efficiently reply to e-mails, allowing the user to concentrate on other important tasks.

[0877] Example 1

[0878] 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."

[0879] In conventional email processing systems, users had to manually check new emails, analyze their content, and think up a reply, which required a great deal of time and effort. Particularly in busy work situations where quick and accurate responses are required, these manual tasks significantly reduced user productivity.

[0880] 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.

[0881] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content using AI technology based on the analyzed content, means for automatically embedding the predicted reply content into a template, means for presenting the template to the user and allowing the user to edit and confirm it, and means for sending the reply content edited and confirmed by the user. This enables the user to automate a series of processes from checking newly arrived emails, generating reply content, embedding it into a template, and sending the final email, quickly and efficiently.

[0882] "New email" is email that has just been received and that the user has not yet opened.

[0883] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes text tokenization, keyword extraction, and intent analysis.

[0884] "AI technology" is a technology that uses artificial intelligence to analyze data and generate appropriate answers and predictions.

[0885] "Means for predicting reply content" refers to functions and algorithms for generating appropriate reply content based on the content of received email.

[0886] A "template" is a model of a document created according to a specific format, and is designed to be easy for a user to edit.

[0887] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task (e.g., generating email replies).

[0888] A "prompt" is text data input to a generative AI model that provides instructions and context for the model to generate a reply.

[0889] "Automation" refers to the process by which machines or software automatically perform tasks that humans would otherwise perform manually.

[0890] "User review and editing" refers to the process in which the user reviews the email reply content generated by the system and makes corrections as necessary.

[0891] This invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[0892] Server behavior:

[0893] The server periodically connects to the user's email account and checks for new emails. If new emails are available, the server receives them using the IMAP or POP3 protocol. The server then analyzes the email content using natural language processing techniques (e.g., Python's NLTK or SpaCy). This analysis extracts the email's intent and important keywords. Next, the server uses a generative AI model (e.g., GPT-4) to predict an appropriate reply based on the analyzed content.

[0894] As a concrete example, if an email is received saying "Please confirm the date of the next meeting," the server analyzes the content of the email and extracts the intent of "confirm the date of the meeting." Based on that intent, the server inputs the prompt "Please confirm the date of the next meeting" into the generative AI model and obtains the reply text (for example, "How about next Tuesday?") obtained from the AI ​​model.

[0895] Terminal behavior:

[0896] The server sends the predicted reply content to the terminal to present to the user. The terminal embeds the predicted reply content in a template and displays it to the user. For example, this template has the format "Dear [Recipient],\n\n[How about next Tuesday?]\n\nBest regards,\n[Your Name]". The user checks the template presented to them through the terminal and edits the content as necessary. Once editing is complete, the user checks the content and finally sends it to the server.

[0897] User Action:

[0898] The user creates the final reply based on the template that has been checked and edited on the terminal. If any corrections are necessary, the template "How about next Tuesday?" can be edited to "How about next Tuesday at 3:00 PM?". Once editing is complete, the user sends the reply to the server. The server uses the SMTP protocol to send the final email to the recipient.

[0899] In this way, users can efficiently complete a series of automated processes, from checking new emails to creating, editing, and sending replies. This significantly reduces the time and effort required for users to process emails, improving productivity. This system ensures speed and accuracy in responding to emails, contributing to work efficiency.

[0900] As a concrete example, the following prompt sentences are fed into a generative AI model to generate a reply:

[0901] "Generate an appropriate response to an email asking to confirm the next meeting date."

[0902] Please use this information to create a polite business reply email.

[0903] This allows the generative AI model to automatically generate appropriate reply content.

[0904] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0905] Step 1: Check for new emails

[0906] The server periodically connects to the user's email account and checks for new email. The server connects to the mail server using the IMAP or POP3 protocol and checks for new email. If new email is available, the server accepts it and proceeds to the next step.

[0907] Specific behavior:

[0908] Input: Email account information, IMAP / POP3 protocol

[0909] Data processing: Query the mail server and check for new mail.

[0910] Output: List of new emails (email ID, sender, subject, body)

[0911] Step 2: Analyzing the email content

[0912] The server acquires the content of newly received emails and analyzes them using natural language processing technology. Tools such as NLTK and SpaCy are used for the analysis. This analysis extracts the intent of the email and important keywords.

[0913] Specific behavior:

[0914] Input: Body of new email

[0915] Data processing: Using natural language processing tools to tokenize the email body and extract key keywords and intent

[0916] Output: Analysis results (intention, keywords)

[0917] Step 3: Anticipate replies

[0918] Based on the parsed content, the server inputs prompts into a generative AI model to predict appropriate responses, taking into account past email data and business context in the process.

[0919] Specific behavior:

[0920] Input: Analysis results (intention, keywords), prompt

[0921] Data calculation: Enter a prompt into the generative AI model and generate a reply.

[0922] Output: Predicted reply

[0923] Step 4: Generate a reply template

[0924] The server generates a reply template by embedding the predicted reply content into a predefined template, which is then generated in a format that is easy for the user to edit.

[0925] Specific behavior:

[0926] Input: predicted reply

[0927] Data processing: Embed reply content in a template

[0928] Output: Reply template (e.g., "Dear [Recipient],\n\n[Reply]\n\nBest regards,\n[Your Name]")

[0929] Step 5: User review and editing

[0930] The terminal presents the reply template received from the server to the user, who is then provided with an interface to review the template and edit its contents as necessary.

[0931] Specific behavior:

[0932] Input: Reply template

[0933] Data processing: Displaying a template on the user interface

[0934] Output: User edits

[0935] Step 6: Sending the final email

[0936] The server receives the final reply, edited and reviewed by the user, and sends it to the email recipient using the SMTP protocol.

[0937] Specific behavior:

[0938] Input: Reply confirmed by the user

[0939] Data Processing: Sending email using the SMTP protocol

[0940] Output: Sending result log

[0941] In this way, all steps are automated, allowing users to reply to emails quickly and accurately with minimal effort, improving user productivity and significantly increasing work efficiency.

[0942] (Application example 1)

[0943] 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."

[0944] For companies and organizations, responding to security-related emails is extremely important, but manual response requires a great deal of time and effort. Furthermore, it is not easy to respond appropriately and promptly to security alerts, which require a rapid response. Therefore, there is a need for a system that automates the entire process of receiving emails, analyzing them, generating replies, and sending them, thereby streamlining responses to security-related emails in particular.

[0945] 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.

[0946] In this invention, the server includes a means for acquiring newly arrived emails, a means for analyzing the content of the emails using natural language processing technology, a means for analyzing security-related emails and generating reply content, a means for verifying that the generated reply content is in accordance with the security policy, and a means for providing an interface designed to enable a prompt response to security alerts, thereby enabling prompt and appropriate responses to security-related emails in companies and organizations.

[0947] "Newly arrived email" refers to email with the most recent received date and time.

[0948] "Natural language processing technology" refers to computer technology for analyzing and understanding human language.

[0949] "Reply content prediction" refers to the process of estimating what should be replied to based on the content of the received email.

[0950] "Template embedding" refers to incorporating predicted reply content into a predetermined format or framework.

[0951] "Security-related email" refers to emails that contain content related to security threats such as phishing, malware, and unauthorized access.

[0952] A "security policy" refers to a guideline that describes the security measures and response policies established by an organization.

[0953] "Interface" refers to the screen and operation method used by users to operate the system.

[0954] A "generative AI model" refers to an artificial intelligence model that uses machine learning to generate information for specific tasks.

[0955] A "prompt" is a piece of text that describes instructions or questions to be input into a generative AI model.

[0956] A "server" refers to a computer system that performs a series of processes such as receiving, analyzing, generating a reply, and sending an email.

[0957] To implement this invention, it is necessary to build a system that automates security-related email correspondence. This system has the following main functions:

[0958] First, the server connects to the company or organization's mail server and retrieves new emails. This is done using the IMAP protocol. When new emails are received, their contents are stored in a database. The software used in this step is an email sending and receiving library (e.g., IMAP, SMTP).

[0959] Next, the server analyzes the content of the received email using natural language processing technology. This analysis uses natural language processing libraries such as spaCy and NLTK. As a result of the analysis, information particularly related to security is extracted. For example, keywords such as "phishing," "malware," and "anomaly detection" are identified.

[0960] The server then predicts the reply content based on the analysis results. This is where an AI model (e.g., GPT-4) is used. The AI ​​model references past data and security policies to generate the optimal reply content. This simultaneously verifies that the generated reply content complies with the security policy.

[0961] The generated reply is embedded in a pre-prepared template and presented to the user. The template uses the following prompt:

[0962] "Generate an appropriate response to the following phishing report email: Subject: Phishing Report Content: User reporting phishing activity Example of expected response: "Dear [reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter.""

[0963] The terminal displays this template to the user and provides an interface that allows the user to view and edit the content. This interface is designed to allow the user to select appropriate actions based on security policies.

[0964] After the user has finished editing and checking, the reply is finally sent to the server, which then sends it to the recipient using the SMTP protocol.

[0965] Through the above process, this system will enable companies and organizations to streamline security-related email responses and achieve prompt and appropriate responses.

[0966] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0967] Step 1:

[0968] The server connects to the mail server of the company or organization using the IMAP protocol to receive new emails and stores them in a database.

[0969] Input: New email from the company's email server

[0970] Output: New emails stored in the database

[0971] Specific operation: Connect to the mail server using the IMAP library, retrieve unread emails, and store their contents (body, subject, sender information, etc.) in a database.

[0972] Step 2:

[0973] The server analyzes the content of the stored emails using natural language processing techniques (e.g., spaCy or NLTK), particularly to extract security-related information (e.g., phishing, malware, unauthorized access).

[0974] Input: Contents of new emails stored in the database

[0975] Output: Parsed email content (security-related information)

[0976] Specific operations: Tokenize the email content, extract keywords, and analyze sentence structure to identify security-related information.

[0977] Step 3:

[0978] The server predicts the reply content based on the analysis results. Here, a generative AI model (e.g., GPT-4) is used. The AI ​​model generates the optimal reply content by referring to past data and security policies.

[0979] Input: Parsed email content (security-related information)

[0980] Output: Predicted reply

[0981] Specific operation: The analyzed information is input into the AI ​​model, and prompts are used to generate predicted replies. Example: "Generate an appropriate reply to the following phishing report email: Subject: Phishing Report Content: User's report of phishing activity Example of predicted reply: "Dear [Reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter."

[0982] Step 4:

[0983] The server embeds the generated reply content into a template based on a pre-defined format to create data to present to the user.

[0984] Input: predicted reply

[0985] Output: Reply content embedded in template

[0986] Specific operation: The predicted reply content is automatically embedded in the defined template and converted into a format that can be presented to the user.

[0987] Step 5:

[0988] The terminal displays the reply content embedded in the template to the user, and provides an interface for the user to view and edit the content.

[0989] Input: Reply content embedded in template

[0990] Output: A screen that the user can see and edit

[0991] What it does: Displays a form through the user interface to allow the user to review and edit the reply content embedded in the template, and waits for the user to finish editing.

[0992] Step 6:

[0993] The user then sends the reply content after reviewing and editing it to the server, which receives it and sends the final reply email to the email recipient.

[0994] Input: The reply that the user has reviewed and edited.

[0995] Output: Reply email sent

[0996] Specific behavior: Receives the content that the user has completed editing and sends a final reply email using the SMTP protocol.

[0997] The above steps create a system that automates security-related email responses and allows them to be handled efficiently and quickly.

[0998] 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.

[0999] The present invention relates to a system that automates the entire process from receiving new emails, analyzing them, generating reply content, embedding them in templates, and finally sending them, and combines it with an emotion engine that recognizes the user's emotions and optimizes the reply content.

[1000] composition

[1001] The system has the following main functions:

[1002] 1. Receiving new emails

[1003] The server periodically checks the user's mail account and receives new email, saving the user the trouble of having to manually check their mail.

[1004] 2. Email Content Analysis

[1005] The server analyzes the content of the email received using natural language processing technology to extract the intent of the email and important information.

[1006] 3. Emotional Recognition

[1007] The emotion engine recognizes the user's emotions and analyzes the user's emotional state from the email text and the user's reading.

[1008] 4. Predicting replies

[1009] The server uses AI technology to predict the appropriate reply based on the analysis results and data from the emotion engine. The emotional information from the emotion engine is reflected in the tone and expression of the reply.

[1010] 5. Generate a reply template

[1011] The server embeds the predicted reply content into a standard template and presents the completed reply email to the user, generating a consistent reply document.

[1012] 6. User Review and Editing

[1013] The terminal displays the reply template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[1014] 7. Sending emails

[1015] The server receives the reply email that has been finally confirmed by the user and sends it to the recipient. During this process, the sending history is recorded.

[1016] Operation details

[1017] 1. Server Operation

[1018] The server connects to the user's email account and receives new emails. The received emails are added to a processing queue within the server. Next, natural language processing technology is used to analyze the content of the email and extract its intent and important information. In parallel, an emotion engine recognizes the user's emotions and processes them by combining them with the analysis results.

[1019] 2. Device Operation

[1020] Based on the analysis data and emotion data received by the device from the server, the AI ​​model generates an appropriate reply. The generated reply is automatically embedded in a template and presented to the user. The user can then review the reply through the device interface and edit it as necessary.

[1021] 3. User Actions

[1022] The user reviews the reply displayed on the device, especially the tone and expressions based on emotions, and edits the reply if necessary. Once the final review is complete, the reply email is sent.

[1023] Specific examples

[1024] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[1025] Server operation: The server receives the email, analyzes its contents, and extracts the intent of "confirming the meeting date." At the same time, the emotion engine recognizes the user's current emotional state (e.g., high stress). Based on this, the AI ​​model predicts the appropriate reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[1026] What happens on the device: The device embeds the above reply content into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[1027] User action: The user checks the revised reply and gives a final confirmation before sending it to the server, which then sends the reply to the recipient.

[1028] This system allows users to reply to e-mails quickly and efficiently, and improves the quality of communication by taking emotional information into account when responding.

[1029] The processing flow will be explained below.

[1030] Step 1:

[1031] Receive new emails

[1032] The server periodically connects to the user's mail account to check for new emails. If new emails are found, they are retrieved from the mail server and added to a processing queue on the server.

[1033] Step 2:

[1034] Email content analysis

[1035] The server analyzes the body of the newly received email using natural language processing technology. During the analysis, the intent and important information are extracted from the email content. This data becomes the input information required for subsequent processing.

[1036] Step 3:

[1037] Emotion recognition

[1038] The emotion engine installed on the server collects the user's emotional data, analyzes the current emotional state (e.g., stress, joy, anger, etc.) from the text of the email, the user's past behavioral history, reading, etc., and generates quantified emotional data.

[1039] Step 4:

[1040] Response prediction

[1041] The server uses an AI model to predict an appropriate reply based on the analysis data obtained through natural language processing and the emotional data recognized by the emotion engine. For example, if the user is feeling stressed in response to an email asking, "Please confirm the date of the next meeting," the server will generate a gentler reply such as, "How about next Tuesday? We can also arrange other dates, so please let us know if that's convenient for you."

[1042] Step 5:

[1043] Generate a reply template

[1044] The server embeds predicted replies into standard templates, which have fixed parts (such as a greeting or signature) and variable parts (the predicted replies), resulting in consistent replies.

[1045] Step 6:

[1046] Show reply content to user

[1047] The terminal displays the generated reply template to the user, and the user interface provides a form for reviewing and editing the reply content, allowing the user to easily modify the content.

[1048] Step 7:

[1049] User review and editing

[1050] The user uses the device to check the displayed reply template. The user reviews the content and edits the template as needed. For example, the user can add a specific date, such as "How about 3:00 PM next Tuesday?"

[1051] Step 8:

[1052] Sending emails

[1053] The user finally sends the reply email that has been confirmed and edited to the server. The server receives and confirms this reply email, and sends it to the recipient's email address. Once the sending is complete, the history is recorded.

[1054] This allows users to reply to emails efficiently and appropriately according to their emotions, which not only improves work productivity but also the quality of communication.

[1055] Example 2

[1056] 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."

[1057] Conventional email reply systems require a lot of time and effort for manually checking emails and creating replies, making it difficult to reply quickly. Furthermore, the system generates a uniform reply without considering the user's feelings, which results in a decline in the quality of communication. The objective of this invention is to solve these problems and provide a system that realizes efficient, emotionally appropriate automatic replies.

[1058] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1059] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for recognizing the user's emotions using an emotion recognition engine based on the analyzed content, means for predicting reply content based on the analysis result and emotion data using a generative AI model, means for embedding the predicted reply content in a template, means for presenting the template to the user, and means for sending the reply content edited and confirmed by the user. This enables fast and efficient email replies and realizes high-quality communication that takes the user's emotions into consideration.

[1060] The "means for obtaining new e-mails" is a function for periodically accessing the user's e-mail account, checking whether new e-mails have arrived, and receiving them.

[1061] "Natural language processing technology" is a set of technologies that allow computers to understand, analyze, and generate human language, and is a language analysis technology used to understand the content of emails.

[1062] An "emotion recognition engine" is a system for recognizing a user's emotional state, and is a technology that estimates a user's emotions based on emails and past data.

[1063] A "generative AI model" is a model that uses artificial intelligence techniques to generate appropriate responses or documents based on given data or information.

[1064] "Analysis results" refers to information obtained when analyzing the content of an email using natural language processing technology, and includes the intent and important information.

[1065] "Emotion data" is information obtained as a result of analyzing the user's emotional state using an emotion recognition engine.

[1066] "Means for predicting reply content" is a function that uses a generative AI model to generate appropriate reply content based on the analysis results and emotional data.

[1067] The "means for embedding in a template" is a function for incorporating predicted reply content into a standard format prepared in advance.

[1068] The "means for presenting to the user" is a function for displaying the reply content embedded in the template to the user on the terminal.

[1069] The "means for sending reply content edited and confirmed by the user" is a function for sending reply content that the user has confirmed and edited to the recipient, and is necessary to complete the email sending process.

[1070] This invention relates to a system that automates the entire process from receiving new e-mails, analyzing their contents, generating reply content, embedding them in templates, and finally sending them, and further optimizes the reply content by recognizing the user's emotions. This system operates in cooperation with a server, a terminal, and a user.

[1071] Server Operation

[1072] The server periodically connects to the user's mail account to receive new emails using the IMAP or POP3 protocols, which are then added to a processing queue on the server.

[1073] The server then analyzes the email content using natural language processing (NLP) techniques, such as Google BERT or OpenAI GPT. The server then uses NLP techniques to extract the main intent and important information from the email.

[1074] Furthermore, an emotion recognition engine is activated to recognize the user's emotions. This emotion recognition is performed using tools such as the IBM Watson Tone Analyzer. The analyzed emotion data is stored on the server and used to generate subsequent reply content.

[1075] The server uses a generative AI model (such as OpenAI GPT-3) to predict an appropriate reply based on the analysis results and emotional data. The reply is generated in a tone that reflects the user's emotions.

[1076] The predicted reply content is embedded in a standard template, which has a predefined format based on which a reply document is automatically generated.

[1077] Device behavior

[1078] The terminal displays the templated reply content received from the server to the user. The user can check the reply content on the terminal and edit it as necessary. The interface is user-friendly, allowing for easy editing.

[1079] User Actions

[1080] The user checks the reply displayed on their device, particularly the tone and expressions reflected by the emotion engine, and makes any necessary corrections. Once the corrections are complete, the user makes a final check and sends the reply email to the server. The server then sends this reply email to the recipient and saves the sending history.

[1081] Specific examples

[1082] For example, when a user receives an email message saying "Please confirm the date of the next meeting," the process proceeds as follows:

[1083] The server receives the email and analyzes its contents. At the same time, the emotion engine recognizes the user's emotional state and determines that they are in a "high stress state." Based on this, the AI ​​model generates a reply such as, "How about next Tuesday at 3:00 PM? Please let me know if other dates are also convenient."

[1084] The terminal embeds this reply content in a template and presents it to the user. The user then modifies it to "Can you make it 4:00 PM next Tuesday instead of 3:00 PM?" and presses the confirmation button. The server then sends the modified reply content to the other party and saves it as a transmission history.

[1085] Prompt Sentence Examples

[1086] To generate a reply to the email "Please confirm the date of the next meeting," the prompt would look like this:

[1087] The email read: "Please confirm the date of our next meeting."

[1088] User Emotion: "High Stress"

[1089] Example reply: "How about next Tuesday? Tuesday at 3:00 PM? We can also arrange other dates, so please let us know if that's convenient for you."

[1090] This allows the system to reply to emails quickly and efficiently, and also enables responses that take into account the user's feelings, improving the quality of communication.

[1091] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1092] Program processing flow

[1093] Step 1: Receive new email

[1094] Input: User's email account information

[1095] What happens: The server connects to the user's email account (for example, a mail server using the IMAP or POP3 protocol).

[1096] Data processing / calculation: Periodically check for new emails.

[1097] Output: Receives any new mail and adds it to the processing queue.

[1098] Step 2: Email Content Analysis

[1099] Input: New email content

[1100] Specific behavior: Obtains the contents of emails received by the server.

[1101] Data processing / calculation: Analyze email content using natural language processing (NLP) techniques, using NLP models such as "Google BERT" and "OpenAI GPT."

[1102] Output: The intent and important information of the email is extracted.

[1103] Step 3: Recognize emotions

[1104] Input: Parsed email content

[1105] Specific operation: The server launches an emotion recognition engine (e.g., "IBM Watson Tone Analyzer").

[1106] Data processing / calculation: Recognize the user's emotional state based on the text of the email and past email history.

[1107] Output: User emotion data (e.g. stress, joy, anger, etc.).

[1108] Step 4: Anticipate replies

[1109] Input: Analysis results and emotion data

[1110] Specific operation: The server generates the reply content using a generative AI model (e.g., "OpenAI GPT-3").

[1111] Data processing / calculation: Predicting the appropriate tone and content of a reply based on analysis results and sentiment data.

[1112] Output: The predicted reply.

[1113] Step 5: Generate a reply template

[1114] Input: Predicted reply

[1115] What happens: The server embeds the predicted reply content into a standard template.

[1116] Data processing / calculation: Assembling the reply content into a predetermined format.

[1117] Output: A templated reply email.

[1118] Step 6: User review and editing

[1119] Input: Templated reply email

[1120] Specific behavior: The device displays the templated reply email to the user.

[1121] Data processing / calculation: The user checks the reply content on the terminal and edits it as necessary. Editing is done using the interface.

[1122] Output: The final reply that the user reviewed and edited.

[1123] Step 7: Sending an email

[1124] Input: The last reply the user confirmed

[1125] Specific action: The server receives the final confirmed reply email.

[1126] Data processing / calculation: Sends email to the recipient using the SMTP protocol.

[1127] Output: Sent emails and their sending history are recorded.

[1128] Based on the above steps, the system automatically and efficiently analyzes and replies to emails, allowing users to reply quickly and achieve appropriate communication according to their emotional state.

[1129] (Application example 2)

[1130] 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."

[1131] Currently, there are many systems that automate email replies, but they lack the ability to generate replies and content that take the user's emotional state into account. This makes it difficult to respond appropriately to the user's emotions, and improving the quality of communication is a challenge.

[1132] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1133] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content based on the analyzed content, means for identifying the user's emotional state using emotion analysis technology, means for generating optimal content based on the user's emotional state, means for automatically embedding the content in a template, means for presenting the template to the user, and means for sending reply content edited and confirmed by the user, thereby enabling the generation of appropriate replies and content that take the user's emotional state into consideration.

[1134] "New email" refers to new, unread email that has arrived in a user's email account.

[1135] "Natural language processing technology" refers to technology that allows computers to understand, analyze, and generate human language.

[1136] "Emotion analysis technology" is a technology that identifies and analyzes a user's emotional state from text and voice data.

[1137] The "reply content" is a sentence that indicates an appropriate answer or response to the received email.

[1138] A "template" is a model for automatically filling in content based on a defined format or structure.

[1139] "Content" refers to the information and entertainment content provided to users.

[1140] A "generative AI model" refers to an algorithm or program that uses artificial intelligence technology to generate new information or text from data.

[1141] "User interface" refers to the screens and operating means that allow users to interact with the system.

[1142] "Emotional state" refers to the mental or emotional state a user is in at a particular moment.

[1143] An "information processing system" is a computer system that collects, analyzes, generates, and communicates data.

[1144] This invention provides a specific implementation method for an emotion-aware media recommendation system, which includes processes for receiving new emails, analyzing their contents, predicting and generating reply messages, and analyzing the user's emotional state to provide optimal content.

[1145] Hardware and Software Configuration

[1146] 1. Hardware:

[1147] Smartphone

[1148] Smart Glasses

[1149] head-mounted display

[1150] 2. Software:

[1151] Python

[1152] TextBlob library (for natural language processing techniques)

[1153] OpenAI API (for generative AI models)

[1154] Processing Details

[1155] The server first retrieves new emails. The server periodically checks the user's email account and receives newly arrived unread emails. These new emails are added to a processing queue and their contents are analyzed using natural language processing techniques.

[1156] The analyzed content extracts the email's subject, intent, and key information, and then uses sentiment analysis technology to identify the user's emotional state, which is then classified as positive, neutral, or negative.

[1157] Using a generative AI model, the system generates replies and content appropriate to the user's emotional state based on the analysis results and data from emotion analysis technology. For example, if a user types "I'm a little tired" and TextBlob identifies this as a negative emotion, the system will use the OpenAI API to recommend relaxing movies.

[1158] The generated reply and content are automatically embedded in a template and presented to the user, who can then view the template displayed on their smartphone, smart glasses, or head-mounted display and edit it as needed.

[1159] Finally, the reply or content that the user has confirmed and edited is sent to the other party, allowing the user to provide the most appropriate reply or content that suits their current emotional state.

[1160] Specific examples

[1161] When a user receives an email saying "Please confirm the date of your next meeting," the process goes like this:

[1162] Server operation: The server receives the email, analyzes its content, and extracts the intent of "confirming the meeting date." At the same time, emotion analysis technology identifies the user's current emotional state (e.g., high stress). Based on this, the generative AI model predicts the reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[1163] What happens on the device: The device automatically embeds the above reply into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[1164] User action: The user reviews the revised reply, gives a final confirmation, and then sends the email. This process records the sending history.

[1165] Prompt Sentence Examples

[1166] "Can you recommend a relaxing movie?"

[1167] In this way, by using this system, users can receive content that is optimal for their emotional state, thereby improving the quality of communication.

[1168] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1169] Step 1:

[1170] The server periodically checks the user's mail account to retrieve new emails.

[1171] Input: User account information on the mail server

[1172] Output: New emails received

[1173] What happens: The server accesses the user's mailbox and checks if there is new mail. If there is new mail, it adds it to the processing queue.

[1174] Step 2:

[1175] The server analyzes the content of newly arrived emails using natural language processing technology.

[1176] Input: Text data of newly arrived email

[1177] Output: Important information and intent of the email (e.g., meeting schedule confirmation)

[1178] What it does: The server uses a natural language processing library such as TextBlob to extract key keywords and intent from the text of the email.

[1179] Step 3:

[1180] The server uses emotion analysis techniques to identify the user's emotional state.

[1181] Input: Text data such as user input comments and emails

[1182] Output: User's emotional state (e.g. positive, neutral, negative)

[1183] What it does: The server uses TextBlob to calculate the emotional polarity of the input text and identifies the user's emotional state based on that.

[1184] Step 4:

[1185] Generative AI models are used to generate content based on analysis results and emotional data.

[1186] Input: Email analysis results, user emotional state

[1187] Output: Content recommendation or reply

[1188] How it works: The server uses the OpenAI API to send a prompt (e.g., "Tell me a relaxing movie") and have the AI ​​model generate appropriate content.

[1189] Step 5:

[1190] The server automatically embeds the generated content into a template and sends it to the terminal.

[1191] Input: Generated content or reply content

[1192] Output: Content embedded in the template

[1193] Specific operation: The server embeds the generated content into a predefined template and sends the template to the terminal.

[1194] Step 6:

[1195] The terminal displays the template to the user, who then checks and edits it.

[1196] Input: Content embedded in the template

[1197] Output: Reply edited by the user

[1198] Specific behavior: The device displays the template in the user interface, allowing the user to review the content and make any necessary modifications.

[1199] Step 7:

[1200] The reply content that the user has confirmed and edited is sent to the server and then transmitted to the other party.

[1201] Input: Reply content confirmed and edited by the user

[1202] Output: Reply email sent to the recipient

[1203] Specific operation: After the user makes a final confirmation, the reply is sent to the server, which then sends it to the other party and records the sending history.

[1204] These processing steps allow users to receive replies and content that are efficient, emotionally relevant, and relevant.

[1205] 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.

[1206] 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.

[1207] 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.

[1208] [Fourth embodiment]

[1209] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1210] 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.

[1211] 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).

[1212] 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.

[1213] 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.

[1214] 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).

[1215] 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.

[1216] 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.

[1217] 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.

[1218] 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.

[1219] 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.

[1220] 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.

[1221] 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."

[1222] The present invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[1223] composition

[1224] The system has the following main functions:

[1225] 1. Receiving new emails

[1226] The server periodically checks the user's mail account to retrieve new emails, eliminating the need for the user to manually check their mail.

[1227] 2. Email Content Analysis

[1228] The server analyzes the content of the email using natural language processing technology, which extracts the email's intent and important information.

[1229] 3. Predicting replies

[1230] The server uses AI technology to predict the reply content based on the analyzed email content, using data from past emails and business context.

[1231] 4. Generate a reply template

[1232] The server embeds the predicted reply content into a predefined template and presents the completed reply email to the user.

[1233] 5. User review and editing

[1234] The terminal displays the template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[1235] 6. Sending emails

[1236] The server sends a reply email that the user finally checks and corrects.

[1237] Operation details

[1238] 1. Server Operation

[1239] The server connects to the user's email account and checks for new emails. If there are new emails, they are received and added to a processing queue for analysis. Next, natural language processing techniques are used to analyze the email's content and extract its intent (e.g., confirming a meeting date). Based on the results of this analysis, the server uses an AI model to predict an appropriate reply (e.g., "How about next Tuesday?").

[1240] 2. Device Operation

[1241] The terminal receives the predicted reply content from the server, embeds it in a template, and presents it to the user. An interface is provided for the user to view and edit the reply template.

[1242] 3. User Actions

[1243] The user reviews the proposed reply and edits it if necessary, and the final, reviewed reply is sent to the server, which then sends it to the email recipient.

[1244] Specific examples

[1245] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[1246] Server operation: The server receives the email and analyzes its contents. As a result of the analysis, the intent of "confirming the meeting date" is extracted. Based on this intent, the AI ​​model predicts the reply content, such as "How about next Tuesday?" This is then embedded in a template and presented to the user.

[1247] Terminal behavior: The terminal displays a template ("Dear [Recipient],\n\nHow about next Tuesday?\n\nBest regards,\n[Your Name]") to the user. The user edits this template, for example, to "How about next Tuesday at 3:00 PM?"

[1248] User action: User reviews the modified reply and sends it to the server, which finally sends this email.

[1249] In this way, the system of the present invention makes replying to e-mails more efficient and improves user productivity.

[1250] The processing flow will be explained below.

[1251] Step 1:

[1252] Receive new emails

[1253] The server periodically checks the user's mail account to see if there is any new email. If new mail is found, it is retrieved from the mail server and added to a processing queue on the server.

[1254] Step 2:

[1255] Email content analysis

[1256] The server analyzes the body of newly arrived emails using natural language processing technology. Specifically, the body of the email is fed into an NLP (natural language processing) model to extract the email's intent and important information. The results of this analysis are used as input for subsequent processing.

[1257] Step 3:

[1258] Response prediction

[1259] The server then uses an AI model to predict the appropriate reply based on the analyzed email content. The AI ​​model incorporates data from past emails and business context, resulting in highly accurate replies.

[1260] Step 4:

[1261] Generate a reply template

[1262] The server embeds the predicted reply content into a standard template, which has fixed parts (e.g., greetings and signatures) and variable parts (reply content generated by AI), resulting in consistent reply emails.

[1263] Step 5:

[1264] Show reply content to user

[1265] The terminal displays the generated reply template to the user, and the user interface provides a form that allows the user to review and edit the reply content, allowing the user to easily review and modify the content.

[1266] Step 6:

[1267] User review and editing

[1268] The user checks the email reply template using the terminal. If necessary, the user edits the content of the template and performs a final check. Once editing is complete, the user clicks the send button to finalize the reply email.

[1269] Step 7:

[1270] Sending emails

[1271] The server receives the reply email that has been finally confirmed by the user. It then sends it to the recipient's email address via the mail server. Once the sending is complete, it is recorded in the sending history.

[1272] Through the above series of processing steps, the user can quickly and efficiently reply to e-mails, allowing the user to concentrate on other important tasks.

[1273] Example 1

[1274] 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."

[1275] In conventional email processing systems, users had to manually check new emails, analyze their content, and think up a reply, which required a great deal of time and effort. Particularly in busy work situations where quick and accurate responses are required, these manual tasks significantly reduced user productivity.

[1276] 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.

[1277] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content using AI technology based on the analyzed content, means for automatically embedding the predicted reply content into a template, means for presenting the template to the user and allowing the user to edit and confirm it, and means for sending the reply content edited and confirmed by the user. This enables the user to automate a series of processes from checking newly arrived emails, generating reply content, embedding it into a template, and sending the final email, quickly and efficiently.

[1278] "New email" is email that has just been received and that the user has not yet opened.

[1279] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language, and includes text tokenization, keyword extraction, and intent analysis.

[1280] "AI technology" is a technology that uses artificial intelligence to analyze data and generate appropriate answers and predictions.

[1281] "Means for predicting reply content" refers to functions and algorithms for generating appropriate reply content based on the content of received email.

[1282] A "template" is a model of a document created according to a specific format, and is designed to be easy for a user to edit.

[1283] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task (e.g., generating email replies).

[1284] A "prompt" is text data input to a generative AI model that provides instructions and context for the model to generate a reply.

[1285] "Automation" refers to the process by which machines or software automatically perform tasks that humans would otherwise perform manually.

[1286] "User review and editing" refers to the process in which the user reviews the email reply content generated by the system and makes corrections as necessary.

[1287] This invention relates to a system that automates a series of processes from receiving new e-mails, analyzing them, generating reply content, embedding them in templates, and finally sending them.

[1288] Server behavior:

[1289] The server periodically connects to the user's email account and checks for new emails. If new emails are available, the server receives them using the IMAP or POP3 protocol. The server then analyzes the email content using natural language processing techniques (e.g., Python's NLTK or SpaCy). This analysis extracts the email's intent and important keywords. Next, the server uses a generative AI model (e.g., GPT-4) to predict an appropriate reply based on the analyzed content.

[1290] As a concrete example, if an email is received saying "Please confirm the date of the next meeting," the server analyzes the content of the email and extracts the intent of "confirm the date of the meeting." Based on that intent, the server inputs the prompt "Please confirm the date of the next meeting" into the generative AI model and obtains the reply text (for example, "How about next Tuesday?") obtained from the AI ​​model.

[1291] Terminal behavior:

[1292] The server sends the predicted reply content to the terminal to present to the user. The terminal embeds the predicted reply content in a template and displays it to the user. For example, this template has the format "Dear [Recipient],\n\n[How about next Tuesday?]\n\nBest regards,\n[Your Name]". The user checks the template presented to them through the terminal and edits the content as necessary. Once editing is complete, the user checks the content and finally sends it to the server.

[1293] User Action:

[1294] The user creates the final reply based on the template that has been checked and edited on the terminal. If any corrections are necessary, the template "How about next Tuesday?" can be edited to "How about next Tuesday at 3:00 PM?". Once editing is complete, the user sends the reply to the server. The server uses the SMTP protocol to send the final email to the recipient.

[1295] In this way, users can efficiently complete a series of automated processes, from checking new emails to creating, editing, and sending replies. This significantly reduces the time and effort required for users to process emails, improving productivity. This system ensures speed and accuracy in responding to emails, contributing to work efficiency.

[1296] As a concrete example, the following prompt sentences are fed into a generative AI model to generate a reply:

[1297] "Generate an appropriate response to an email asking to confirm the next meeting date."

[1298] Please use this information to create a polite business reply email.

[1299] This allows the generative AI model to automatically generate appropriate reply content.

[1300] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1301] Step 1: Check for new emails

[1302] The server periodically connects to the user's email account and checks for new email. The server connects to the mail server using the IMAP or POP3 protocol and checks for new email. If new email is available, the server accepts it and proceeds to the next step.

[1303] Specific behavior:

[1304] Input: Email account information, IMAP / POP3 protocol

[1305] Data processing: Query the mail server and check for new mail.

[1306] Output: List of new emails (email ID, sender, subject, body)

[1307] Step 2: Analyzing the email content

[1308] The server acquires the content of newly received emails and analyzes them using natural language processing technology. Tools such as NLTK and SpaCy are used for the analysis. This analysis extracts the intent of the email and important keywords.

[1309] Specific behavior:

[1310] Input: Body of new email

[1311] Data processing: Using natural language processing tools to tokenize the email body and extract key keywords and intent

[1312] Output: Analysis results (intention, keywords)

[1313] Step 3: Anticipate replies

[1314] Based on the parsed content, the server inputs prompts into a generative AI model to predict appropriate responses, taking into account past email data and business context in the process.

[1315] Specific behavior:

[1316] Input: Analysis results (intention, keywords), prompt

[1317] Data calculation: Enter a prompt into the generative AI model and generate a reply.

[1318] Output: Predicted reply

[1319] Step 4: Generate a reply template

[1320] The server generates a reply template by embedding the predicted reply content into a predefined template, which is then generated in a format that is easy for the user to edit.

[1321] Specific behavior:

[1322] Input: predicted reply

[1323] Data processing: Embed reply content in a template

[1324] Output: Reply template (e.g., "Dear [Recipient],\n\n[Reply]\n\nBest regards,\n[Your Name]")

[1325] Step 5: User review and editing

[1326] The terminal presents the reply template received from the server to the user, who is then provided with an interface to review the template and edit its contents as necessary.

[1327] Specific behavior:

[1328] Input: Reply template

[1329] Data processing: Displaying a template on the user interface

[1330] Output: User edits

[1331] Step 6: Sending the final email

[1332] The server receives the final reply, edited and reviewed by the user, and sends it to the email recipient using the SMTP protocol.

[1333] Specific behavior:

[1334] Input: Reply confirmed by the user

[1335] Data Processing: Sending email using the SMTP protocol

[1336] Output: Sending result log

[1337] In this way, all steps are automated, allowing users to reply to emails quickly and accurately with minimal effort, improving user productivity and significantly increasing work efficiency.

[1338] (Application example 1)

[1339] 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."

[1340] For companies and organizations, responding to security-related emails is extremely important, but manual response requires a great deal of time and effort. Furthermore, it is not easy to respond appropriately and promptly to security alerts, which require a rapid response. Therefore, there is a need for a system that automates the entire process of receiving emails, analyzing them, generating replies, and sending them, thereby streamlining responses to security-related emails in particular.

[1341] 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.

[1342] In this invention, the server includes a means for acquiring newly arrived emails, a means for analyzing the content of the emails using natural language processing technology, a means for analyzing security-related emails and generating reply content, a means for verifying that the generated reply content is in accordance with the security policy, and a means for providing an interface designed to enable a prompt response to security alerts, thereby enabling prompt and appropriate responses to security-related emails in companies and organizations.

[1343] "Newly arrived email" refers to email with the most recent received date and time.

[1344] "Natural language processing technology" refers to computer technology for analyzing and understanding human language.

[1345] "Reply content prediction" refers to the process of estimating what should be replied to based on the content of the received email.

[1346] "Template embedding" refers to incorporating predicted reply content into a predetermined format or framework.

[1347] "Security-related email" refers to emails that contain content related to security threats such as phishing, malware, and unauthorized access.

[1348] A "security policy" refers to a guideline that describes the security measures and response policies established by an organization.

[1349] "Interface" refers to the screen and operation method used by users to operate the system.

[1350] A "generative AI model" refers to an artificial intelligence model that uses machine learning to generate information for specific tasks.

[1351] A "prompt" is a piece of text that describes instructions or questions to be input into a generative AI model.

[1352] A "server" refers to a computer system that performs a series of processes such as receiving, analyzing, generating a reply, and sending an email.

[1353] To implement this invention, it is necessary to build a system that automates security-related email correspondence. This system has the following main functions:

[1354] First, the server connects to the company or organization's mail server and retrieves new emails. This is done using the IMAP protocol. When new emails are received, their contents are stored in a database. The software used in this step is an email sending and receiving library (e.g., IMAP, SMTP).

[1355] Next, the server analyzes the content of the received email using natural language processing technology. This analysis uses natural language processing libraries such as spaCy and NLTK. As a result of the analysis, information particularly related to security is extracted. For example, keywords such as "phishing," "malware," and "anomaly detection" are identified.

[1356] The server then predicts the reply content based on the analysis results. This is where an AI model (e.g., GPT-4) is used. The AI ​​model references past data and security policies to generate the optimal reply content. This simultaneously verifies that the generated reply content complies with the security policy.

[1357] The generated reply is embedded in a pre-prepared template and presented to the user. The template uses the following prompt:

[1358] "Generate an appropriate response to the following phishing report email: Subject: Phishing Report Content: User reporting phishing activity Example of expected response: "Dear [reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter.""

[1359] The terminal displays this template to the user and provides an interface that allows the user to view and edit the content. This interface is designed to allow the user to select appropriate actions based on security policies.

[1360] After the user has finished editing and checking, the reply is finally sent to the server, which then sends it to the recipient using the SMTP protocol.

[1361] Through the above process, this system will enable companies and organizations to streamline security-related email responses and achieve prompt and appropriate responses.

[1362] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1363] Step 1:

[1364] The server connects to the mail server of the company or organization using the IMAP protocol to receive new emails and stores them in a database.

[1365] Input: New email from the company's email server

[1366] Output: New emails stored in the database

[1367] Specific operation: Connect to the mail server using the IMAP library, retrieve unread emails, and store their contents (body, subject, sender information, etc.) in a database.

[1368] Step 2:

[1369] The server analyzes the content of the stored emails using natural language processing techniques (e.g., spaCy or NLTK), particularly to extract security-related information (e.g., phishing, malware, unauthorized access).

[1370] Input: Contents of new emails stored in the database

[1371] Output: Parsed email content (security-related information)

[1372] Specific operations: Tokenize the email content, extract keywords, and analyze sentence structure to identify security-related information.

[1373] Step 3:

[1374] The server predicts the reply content based on the analysis results. Here, a generative AI model (e.g., GPT-4) is used. The AI ​​model generates the optimal reply content by referring to past data and security policies.

[1375] Input: Parsed email content (security-related information)

[1376] Output: Predicted reply

[1377] Specific operation: The analyzed information is input into the AI ​​model, and prompts are used to generate predicted replies. Example: "Generate an appropriate reply to the following phishing report email: Subject: Phishing Report Content: User's report of phishing activity Example of predicted reply: "Dear [Reporter's name], Thank you for reporting this phishing email. We are currently investigating the matter."

[1378] Step 4:

[1379] The server embeds the generated reply content into a template based on a pre-defined format to create data to present to the user.

[1380] Input: predicted reply

[1381] Output: Reply content embedded in template

[1382] Specific operation: The predicted reply content is automatically embedded in the defined template and converted into a format that can be presented to the user.

[1383] Step 5:

[1384] The terminal displays the reply content embedded in the template to the user, and provides an interface for the user to view and edit the content.

[1385] Input: Reply content embedded in template

[1386] Output: A screen that the user can see and edit

[1387] What it does: Displays a form through the user interface to allow the user to review and edit the reply content embedded in the template, and waits for the user to finish editing.

[1388] Step 6:

[1389] The user then sends the reply content after reviewing and editing it to the server, which receives it and sends the final reply email to the email recipient.

[1390] Input: The reply that the user has reviewed and edited.

[1391] Output: Reply email sent

[1392] Specific behavior: Receives the content that the user has completed editing and sends a final reply email using the SMTP protocol.

[1393] The above steps create a system that automates security-related email responses and allows them to be handled efficiently and quickly.

[1394] 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.

[1395] The present invention relates to a system that automates the entire process from receiving new emails, analyzing them, generating reply content, embedding them in templates, and finally sending them, and combines it with an emotion engine that recognizes the user's emotions and optimizes the reply content.

[1396] composition

[1397] The system has the following main functions:

[1398] 1. Receiving new emails

[1399] The server periodically checks the user's mail account and receives new email, saving the user the trouble of having to manually check their mail.

[1400] 2. Email Content Analysis

[1401] The server analyzes the content of the email received using natural language processing technology to extract the intent of the email and important information.

[1402] 3. Emotional Recognition

[1403] The emotion engine recognizes the user's emotions and analyzes the user's emotional state from the email text and the user's reading.

[1404] 4. Predicting replies

[1405] The server uses AI technology to predict the appropriate reply based on the analysis results and data from the emotion engine. The emotional information from the emotion engine is reflected in the tone and expression of the reply.

[1406] 5. Generate a reply template

[1407] The server embeds the predicted reply content into a standard template and presents the completed reply email to the user, generating a consistent reply document.

[1408] 6. User Review and Editing

[1409] The terminal displays the reply template to the user, allowing the user to review and edit the content, and the user can modify the reply content as needed.

[1410] 7. Sending emails

[1411] The server receives the reply email that has been finally confirmed by the user and sends it to the recipient. During this process, the sending history is recorded.

[1412] Operation details

[1413] 1. Server Operation

[1414] The server connects to the user's email account and receives new emails. The received emails are added to a processing queue within the server. Next, natural language processing technology is used to analyze the content of the email and extract its intent and important information. In parallel, an emotion engine recognizes the user's emotions and processes them by combining them with the analysis results.

[1415] 2. Device Operation

[1416] Based on the analysis data and emotion data received by the device from the server, the AI ​​model generates an appropriate reply. The generated reply is automatically embedded in a template and presented to the user. The user can then review the reply through the device interface and edit it as necessary.

[1417] 3. User Actions

[1418] The user reviews the reply displayed on the device, especially the tone and expressions based on emotions, and edits the reply if necessary. Once the final review is complete, the reply email is sent.

[1419] Specific examples

[1420] For example, if a user receives an email saying "Please confirm the date of the next meeting," the process would go like this:

[1421] Server operation: The server receives the email, analyzes its contents, and extracts the intent of "confirming the meeting date." At the same time, the emotion engine recognizes the user's current emotional state (e.g., high stress). Based on this, the AI ​​model predicts the appropriate reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[1422] What happens on the device: The device embeds the above reply content into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[1423] User action: The user checks the revised reply and gives a final confirmation before sending it to the server, which then sends the reply to the recipient.

[1424] This system allows users to reply to e-mails quickly and efficiently, and improves the quality of communication by taking emotional information into account when responding.

[1425] The processing flow will be explained below.

[1426] Step 1:

[1427] Receive new emails

[1428] The server periodically connects to the user's mail account to check for new emails. If new emails are found, they are retrieved from the mail server and added to a processing queue on the server.

[1429] Step 2:

[1430] Email content analysis

[1431] The server analyzes the body of the newly received email using natural language processing technology. During the analysis, the intent and important information are extracted from the email content. This data becomes the input information required for subsequent processing.

[1432] Step 3:

[1433] Emotion recognition

[1434] The emotion engine installed on the server collects the user's emotional data, analyzes the current emotional state (e.g., stress, joy, anger, etc.) from the text of the email, the user's past behavioral history, reading, etc., and generates quantified emotional data.

[1435] Step 4:

[1436] Response prediction

[1437] The server uses an AI model to predict an appropriate reply based on the analysis data obtained through natural language processing and the emotional data recognized by the emotion engine. For example, if the user is feeling stressed in response to an email asking, "Please confirm the date of the next meeting," the server will generate a gentler reply such as, "How about next Tuesday? We can also arrange other dates, so please let us know if that's convenient for you."

[1438] Step 5:

[1439] Generate a reply template

[1440] The server embeds predicted replies into standard templates, which have fixed parts (such as a greeting or signature) and variable parts (the predicted replies), resulting in consistent replies.

[1441] Step 6:

[1442] Show reply content to user

[1443] The terminal displays the generated reply template to the user, and the user interface provides a form for reviewing and editing the reply content, allowing the user to easily modify the content.

[1444] Step 7:

[1445] User review and editing

[1446] The user uses the device to check the displayed reply template. The user reviews the content and edits the template as needed. For example, the user can add a specific date, such as "How about 3:00 PM next Tuesday?"

[1447] Step 8:

[1448] Sending emails

[1449] The user finally sends the reply email that has been confirmed and edited to the server. The server receives and confirms this reply email, and sends it to the recipient's email address. Once the sending is complete, the history is recorded.

[1450] This allows users to reply to emails efficiently and appropriately according to their emotions, which not only improves work productivity but also the quality of communication.

[1451] Example 2

[1452] 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."

[1453] Conventional email reply systems require a lot of time and effort for manually checking emails and creating replies, making it difficult to reply quickly. Furthermore, the system generates a uniform reply without considering the user's feelings, which results in a decline in the quality of communication. The objective of this invention is to solve these problems and provide a system that realizes efficient, emotionally appropriate automatic replies.

[1454] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1455] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for recognizing the user's emotions using an emotion recognition engine based on the analyzed content, means for predicting reply content based on the analysis result and emotion data using a generative AI model, means for embedding the predicted reply content in a template, means for presenting the template to the user, and means for sending the reply content edited and confirmed by the user. This enables fast and efficient email replies and realizes high-quality communication that takes the user's emotions into consideration.

[1456] The "means for obtaining new e-mails" is a function for periodically accessing the user's e-mail account, checking whether new e-mails have arrived, and receiving them.

[1457] "Natural language processing technology" is a set of technologies that allow computers to understand, analyze, and generate human language, and is a language analysis technology used to understand the content of emails.

[1458] An "emotion recognition engine" is a system for recognizing a user's emotional state, and is a technology that estimates a user's emotions based on emails and past data.

[1459] A "generative AI model" is a model that uses artificial intelligence techniques to generate appropriate responses or documents based on given data or information.

[1460] "Analysis results" refers to information obtained when analyzing the content of an email using natural language processing technology, and includes the intent and important information.

[1461] "Emotion data" is information obtained as a result of analyzing the user's emotional state using an emotion recognition engine.

[1462] "Means for predicting reply content" is a function that uses a generative AI model to generate appropriate reply content based on the analysis results and emotional data.

[1463] The "means for embedding in a template" is a function for incorporating predicted reply content into a standard format prepared in advance.

[1464] The "means for presenting to the user" is a function for displaying the reply content embedded in the template to the user on the terminal.

[1465] The "means for sending reply content edited and confirmed by the user" is a function for sending reply content that the user has confirmed and edited to the recipient, and is necessary to complete the email sending process.

[1466] This invention relates to a system that automates the entire process from receiving new e-mails, analyzing their contents, generating reply content, embedding them in templates, and finally sending them, and further optimizes the reply content by recognizing the user's emotions. This system operates in cooperation with a server, a terminal, and a user.

[1467] Server Operation

[1468] The server periodically connects to the user's mail account to receive new emails using the IMAP or POP3 protocols, which are then added to a processing queue on the server.

[1469] The server then analyzes the email content using natural language processing (NLP) techniques, such as Google BERT or OpenAI GPT. The server then uses NLP techniques to extract the main intent and important information from the email.

[1470] Furthermore, an emotion recognition engine is activated to recognize the user's emotions. This emotion recognition is performed using tools such as the IBM Watson Tone Analyzer. The analyzed emotion data is stored on the server and used to generate subsequent reply content.

[1471] The server uses a generative AI model (such as OpenAI GPT-3) to predict an appropriate reply based on the analysis results and emotional data. The reply is generated in a tone that reflects the user's emotions.

[1472] The predicted reply content is embedded in a standard template, which has a predefined format based on which a reply document is automatically generated.

[1473] Device behavior

[1474] The terminal displays the templated reply content received from the server to the user. The user can check the reply content on the terminal and edit it as necessary. The interface is user-friendly, allowing for easy editing.

[1475] User Actions

[1476] The user checks the reply displayed on their device, particularly the tone and expressions reflected by the emotion engine, and makes any necessary corrections. Once the corrections are complete, the user makes a final check and sends the reply email to the server. The server then sends this reply email to the recipient and saves the sending history.

[1477] Specific examples

[1478] For example, when a user receives an email message saying "Please confirm the date of the next meeting," the process proceeds as follows:

[1479] The server receives the email and analyzes its contents. At the same time, the emotion engine recognizes the user's emotional state and determines that they are in a "high stress state." Based on this, the AI ​​model generates a reply such as, "How about next Tuesday at 3:00 PM? Please let me know if other dates are also convenient."

[1480] The terminal embeds this reply content in a template and presents it to the user. The user then modifies it to "Can you make it 4:00 PM next Tuesday instead of 3:00 PM?" and presses the confirmation button. The server then sends the modified reply content to the other party and saves it as a transmission history.

[1481] Prompt Sentence Examples

[1482] To generate a reply to the email "Please confirm the date of the next meeting," the prompt would look like this:

[1483] The email read: "Please confirm the date of our next meeting."

[1484] User Emotion: "High Stress"

[1485] Example reply: "How about next Tuesday? Tuesday at 3:00 PM? We can also arrange other dates, so please let us know if that's convenient for you."

[1486] This allows the system to reply to emails quickly and efficiently, and also enables responses that take into account the user's feelings, improving the quality of communication.

[1487] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1488] Program processing flow

[1489] Step 1: Receive new email

[1490] Input: User's email account information

[1491] What happens: The server connects to the user's email account (for example, a mail server using the IMAP or POP3 protocol).

[1492] Data processing / calculation: Periodically check for new emails.

[1493] Output: Receives any new mail and adds it to the processing queue.

[1494] Step 2: Email Content Analysis

[1495] Input: New email content

[1496] Specific behavior: Obtains the contents of emails received by the server.

[1497] Data processing / calculation: Analyze email content using natural language processing (NLP) techniques, using NLP models such as "Google BERT" and "OpenAI GPT."

[1498] Output: The intent and important information of the email is extracted.

[1499] Step 3: Recognize emotions

[1500] Input: Parsed email content

[1501] Specific operation: The server launches an emotion recognition engine (e.g., "IBM Watson Tone Analyzer").

[1502] Data processing / calculation: Recognize the user's emotional state based on the text of the email and past email history.

[1503] Output: User emotion data (e.g. stress, joy, anger, etc.).

[1504] Step 4: Anticipate replies

[1505] Input: Analysis results and emotion data

[1506] Specific operation: The server generates the reply content using a generative AI model (e.g., "OpenAI GPT-3").

[1507] Data processing / calculation: Predicting the appropriate tone and content of a reply based on analysis results and sentiment data.

[1508] Output: The predicted reply.

[1509] Step 5: Generate a reply template

[1510] Input: Predicted reply

[1511] What happens: The server embeds the predicted reply content into a standard template.

[1512] Data processing / calculation: Assembling the reply content into a predetermined format.

[1513] Output: A templated reply email.

[1514] Step 6: User review and editing

[1515] Input: Templated reply email

[1516] Specific behavior: The device displays the templated reply email to the user.

[1517] Data processing / calculation: The user checks the reply content on the terminal and edits it as necessary. Editing is done using the interface.

[1518] Output: The final reply that the user reviewed and edited.

[1519] Step 7: Sending an email

[1520] Input: The last reply the user confirmed

[1521] Specific action: The server receives the final confirmed reply email.

[1522] Data processing / calculation: Sends email to the recipient using the SMTP protocol.

[1523] Output: Sent emails and their sending history are recorded.

[1524] Based on the above steps, the system automatically and efficiently analyzes and replies to emails, allowing users to reply quickly and achieve appropriate communication according to their emotional state.

[1525] (Application example 2)

[1526] 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."

[1527] Currently, there are many systems that automate email replies, but they lack the ability to generate replies and content that take the user's emotional state into account. This makes it difficult to respond appropriately to the user's emotions, and improving the quality of communication is a challenge.

[1528] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1529] In this invention, the server includes means for acquiring newly arrived emails, means for analyzing the content of the emails using natural language processing technology, means for predicting reply content based on the analyzed content, means for identifying the user's emotional state using emotion analysis technology, means for generating optimal content based on the user's emotional state, means for automatically embedding the content in a template, means for presenting the template to the user, and means for sending reply content edited and confirmed by the user, thereby enabling the generation of appropriate replies and content that take the user's emotional state into consideration.

[1530] "New email" refers to new, unread email that has arrived in a user's email account.

[1531] "Natural language processing technology" refers to technology that allows computers to understand, analyze, and generate human language.

[1532] "Emotion analysis technology" is a technology that identifies and analyzes a user's emotional state from text and voice data.

[1533] The "reply content" is a sentence that indicates an appropriate answer or response to the received email.

[1534] A "template" is a model for automatically filling in content based on a defined format or structure.

[1535] "Content" refers to the information and entertainment content provided to users.

[1536] A "generative AI model" refers to an algorithm or program that uses artificial intelligence technology to generate new information or text from data.

[1537] "User interface" refers to the screens and operating means that allow users to interact with the system.

[1538] "Emotional state" refers to the mental or emotional state a user is in at a particular moment.

[1539] An "information processing system" is a computer system that collects, analyzes, generates, and communicates data.

[1540] This invention provides a specific implementation method for an emotion-aware media recommendation system, which includes processes for receiving new emails, analyzing their contents, predicting and generating reply messages, and analyzing the user's emotional state to provide optimal content.

[1541] Hardware and Software Configuration

[1542] 1. Hardware:

[1543] Smartphone

[1544] Smart Glasses

[1545] head-mounted display

[1546] 2. Software:

[1547] Python

[1548] TextBlob library (for natural language processing techniques)

[1549] OpenAI API (for generative AI models)

[1550] Processing Details

[1551] The server first retrieves new emails. The server periodically checks the user's email account and receives newly arrived unread emails. These new emails are added to a processing queue and their contents are analyzed using natural language processing techniques.

[1552] The analyzed content extracts the email's subject, intent, and key information, and then uses sentiment analysis technology to identify the user's emotional state, which is then classified as positive, neutral, or negative.

[1553] Using a generative AI model, the system generates replies and content appropriate to the user's emotional state based on the analysis results and data from emotion analysis technology. For example, if a user types "I'm a little tired" and TextBlob identifies this as a negative emotion, the system will use the OpenAI API to recommend relaxing movies.

[1554] The generated reply and content are automatically embedded in a template and presented to the user, who can then view the template displayed on their smartphone, smart glasses, or head-mounted display and edit it as needed.

[1555] Finally, the reply or content that the user has confirmed and edited is sent to the other party, allowing the user to provide the most appropriate reply or content that suits their current emotional state.

[1556] Specific examples

[1557] When a user receives an email saying "Please confirm the date of your next meeting," the process goes like this:

[1558] Server operation: The server receives the email, analyzes its content, and extracts the intent of "confirming the meeting date." At the same time, emotion analysis technology identifies the user's current emotional state (e.g., high stress). Based on this, the generative AI model predicts the reply, such as "How about next Tuesday? We're also open to other dates, so please let us know if that's convenient for you."

[1559] What happens on the device: The device automatically embeds the above reply into a template and displays it to the user. If the user wants to change it to "How about 3:00 PM next Tuesday?", they can easily do so through the interface.

[1560] User action: The user reviews the revised reply, gives a final confirmation, and then sends the email. This process records the sending history.

[1561] Prompt Sentence Examples

[1562] "Can you recommend a relaxing movie?"

[1563] In this way, by using this system, users can receive content that is optimal for their emotional state, thereby improving the quality of communication.

[1564] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1565] Step 1:

[1566] The server periodically checks the user's mail account to retrieve new emails.

[1567] Input: User account information on the mail server

[1568] Output: New emails received

[1569] What happens: The server accesses the user's mailbox and checks if there is new mail. If there is new mail, it adds it to the processing queue.

[1570] Step 2:

[1571] The server analyzes the content of newly arrived emails using natural language processing technology.

[1572] Input: Text data of newly arrived email

[1573] Output: Important information and intent of the email (e.g., meeting schedule confirmation)

[1574] What it does: The server uses a natural language processing library such as TextBlob to extract key keywords and intent from the text of the email.

[1575] Step 3:

[1576] The server uses emotion analysis techniques to identify the user's emotional state.

[1577] Input: Text data such as user input comments and emails

[1578] Output: User's emotional state (e.g. positive, neutral, negative)

[1579] What it does: The server uses TextBlob to calculate the emotional polarity of the input text and identifies the user's emotional state based on that.

[1580] Step 4:

[1581] Generative AI models are used to generate content based on analysis results and emotional data.

[1582] Input: Email analysis results, user emotional state

[1583] Output: Content recommendation or reply

[1584] How it works: The server uses the OpenAI API to send a prompt (e.g., "Tell me a relaxing movie") and have the AI ​​model generate appropriate content.

[1585] Step 5:

[1586] The server automatically embeds the generated content into a template and sends it to the terminal.

[1587] Input: Generated content or reply content

[1588] Output: Content embedded in the template

[1589] Specific operation: The server embeds the generated content into a predefined template and sends the template to the terminal.

[1590] Step 6:

[1591] The terminal displays the template to the user, who then checks and edits it.

[1592] Input: Content embedded in the template

[1593] Output: Reply edited by the user

[1594] Specific behavior: The device displays the template in the user interface, allowing the user to review the content and make any necessary modifications.

[1595] Step 7:

[1596] The reply content that the user has confirmed and edited is sent to the server and then transmitted to the other party.

[1597] Input: Reply content confirmed and edited by the user

[1598] Output: Reply email sent to the recipient

[1599] Specific operation: After the user makes a final confirmation, the reply is sent to the server, which then sends it to the other party and records the sending history.

[1600] These processing steps allow users to receive replies and content that are efficient, emotionally relevant, and relevant.

[1601] 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.

[1602] 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.

[1603] 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.

[1604] 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.

[1605] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion 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.

[1606] 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.

[1607] 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).

[1608] 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.

[1609] 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."

[1610] 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.

[1611] 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).

[1612] 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.

[1613] 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.

[1614] 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.

[1615] 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.

[1616] 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.

[1617] 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.

[1618] 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.

[1619] 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.

[1620] 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.

[1621] 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.

[1622] The following is further disclosed regarding the above embodiment.

[1623] (Claim 1)

[1624] a means for retrieving new emails;

[1625] A means for analyzing the content of emails using natural language processing technology;

[1626] A means for predicting reply content based on the analyzed content;

[1627] A means of embedding predicted reply content into templates;

[1628] means for presenting the template to a user;

[1629] The system includes a means for sending a reply that has been edited and reviewed by the user.

[1630] (Claim 2)

[1631] A means for generating appropriate reply candidates based on the content of the email analyzed by natural language processing technology,

[1632] 10. The system of claim 1, further comprising means for automatically embedding the reply candidates into a template.

[1633] (Claim 3)

[1634] The system according to claim 1, characterized in that the reply content to be embedded in the template is generated using AI technology.

[1635] "Example 1"

[1636] (Claim 1)

[1637] a means for retrieving new emails;

[1638] A means for analyzing the content of emails using natural language processing technology;

[1639] A means of predicting reply content using AI technology based on the analyzed content;

[1640] A way to automatically fill templates with predicted reply content;

[1641] a means for presenting the template to a user for editing and review by the user;

[1642] means for sending a reply that has been edited and reviewed by the user;

[1643] A system that includes a means to automate the process of checking new emails, analyzing emails, predicting reply content, generating templates, checking edits, and sending the final email.

[1644] (Claim 2)

[1645] a means for using a generative AI model to generate appropriate reply candidates based on the content of the email analyzed by natural language processing technology;

[1646] 2. The system according to claim 1, further comprising means for automatically embedding the reply candidates in a template and presenting the template to the user.

[1647] (Claim 3)

[1648] The system described in claim 1 is characterized in that it inputs a prompt sentence into a generative AI model to predict reply content, and uses the user's past email data and business context when embedding it into a template.

[1649] "Application Example 1"

[1650] (Claim 1)

[1651] a means for retrieving new emails;

[1652] A means for analyzing the content of emails using natural language processing technology;

[1653] A means for predicting reply content based on the analyzed content;

[1654] A means of embedding predicted reply content into templates;

[1655] means for presenting the template to a user;

[1656] In addition to the means to send replies that have been edited and reviewed by the user,

[1657] a means for generating security-related email analysis and reply content;

[1658] a means for verifying that the generated reply is in accordance with the security policy;

[1659] A system that includes a means for providing an interface designed to enable rapid response to security alerts.

[1660] (Claim 2)

[1661] A means for generating appropriate reply candidates based on the content of the email analyzed by natural language processing technology;

[1662] means for verifying whether the reply candidate complies with a security policy;

[1663] 10. The system of claim 1, further comprising means for automatically embedding the reply candidates into a template.

[1664] (Claim 3)

[1665] The system of claim 1, characterized in that the reply content embedded in the template is generated using a generative AI model and uses prompt sentences designed to enable a quick response to security alerts.

[1666] "Example 2: Combining Emotion Engines"

[1667] (Claim 1)

[1668] a means for retrieving new emails;

[1669] A means for analyzing the content of emails using natural language processing technology;

[1670] means for recognizing the user's emotion using an emotion recognition engine based on the analyzed content;

[1671] A means of predicting reply content based on analysis results and emotional data using a generative AI model;

[1672] A means of embedding predicted reply content into templates;

[1673] means for presenting the template to a user;

[1674] The system includes a means for sending a reply that has been edited and reviewed by the user.

[1675] (Claim 2)

[1676] A means for generating appropriate reply candidates based on the content and emotional data of emails analyzed using natural language processing technology,

[1677] 10. The system of claim 1, further comprising means for automatically embedding the reply candidates into a template.

[1678] (Claim 3)

[1679] 2. The system of claim 1, wherein the reply content embedded in the template is generated using a generative AI model, the generation taking into account the user's emotional state.

[1680] "Application example 2 when combining emotion engines"

[1681] (Claim 1)

[1682] a means for retrieving new emails;

[1683] A means for analyzing the content of emails using natural language processing technology;

[1684] A means for predicting reply content based on the analyzed content;

[1685] means for identifying the emotional state of a user using emotion analysis techniques;

[1686] means for generating optimal content based on the emotional state of a user;

[1687] A way to automatically embed it in a template,

[1688] means for presenting the template to a user;

[1689] The system includes a means for sending a reply that has been edited and reviewed by the user.

[1690] (Claim 2)

[1691] A means for automatically generating appropriate content based on the emotional state of a user analyzed by an emotion analysis technique,

[1692] 10. The system of claim 1, further comprising means for automatically populating the generated content into a template.

[1693] (Claim 3)

[1694] The system according to claim 1, characterized in that the reply content to be embedded in the template is generated using a generative AI model. [Explanation of symbols]

[1695] 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 retrieving new emails; A means for analyzing the content of emails using natural language processing technology; A means for predicting reply content based on the analyzed content; A means of embedding predicted reply content into templates; means for presenting the template to a user; and means for sending the reply content that has been edited and reviewed by the user.

2. A means for generating appropriate reply candidates based on the content of the email analyzed by natural language processing technology, 2. The system of claim 1, further comprising means for automatically embedding the reply candidates into a template.

3. The system according to claim 1, wherein the reply content to be embedded in the template is generated using AI technology.

Images