system

Abstract

Provide a system. The system includes: a means for automatically generating a subject and a body of a new email based on user input; A method to automatically generate the body of reply emails based on past email exchanges, A way to detect grammatical and stylistic errors in emails and suggest corrections; The system includes a means to extract key points from emails and generate summaries.

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] Recent advances in information and communication technology have led to increased use of email. However, many users spend a lot of time and effort composing and replying to emails. In particular, in business settings, there is a need to compose emails quickly and accurately, which requires specialized skills and experience. It is also difficult to avoid grammatical and stylistic errors and accurately convey the content. Furthermore, summarization techniques are necessary to efficiently understand long emails. To address these issues, there is a need for a system that allows users to compose, reply, correct, and summarize emails efficiently and with minimal effort. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for automatically generating the subject and body of a new email based on user input, a means for automatically generating the body of a reply email based on past email exchanges, a means for detecting grammatical and stylistic errors in emails and suggesting corrections, and a means for extracting important points from emails and generating summaries. This allows users to efficiently compose emails with minimal effort, avoiding grammatical and stylistic errors, and quickly reply or summarize emails. Furthermore, these means utilize natural language processing technology to accurately analyze the input content and the content of past emails, allowing for appropriate email generation and correction.

[0006] "User" refers to a person who operates the system and creates, replies to, corrects, and summarizes emails.

[0007] "Input" refers to information or data that a user provides to a system.

[0008] A "new email" refers to an email that a user creates, which includes a subject and a body.

[0009] The "subject" refers to the title of the email, and includes summary information that allows the recipient to understand the content.

[0010] "Body" refers to the main message content of the email.

[0011] "Automatic generation" means that the system performs the process automatically without requiring manual operation by the user.

[0012] "Past email exchanges" refers to the content of emails previously exchanged between a specific user and the other party.

[0013] A "reply email" refers to an email created by a user to reply to an email that was previously received.

[0014] "Grammar" refers to the rules and structure of a language, which governs correct language usage.

[0015] "Style" refers to the form of a piece of writing and the choice of words used to ensure that content is conveyed appropriately and effectively.

[0016] "Errors" refer to improper grammar or style.

[0017] "Fix" refers to a proposed improvement to correct a detected error.

[0018] "Important points" refer to information or parts of the email content that are considered particularly important.

[0019] A "summary" is a compact summary of the main information in the original text.

[0020] "Natural language processing" refers to techniques and methods that allow computers to understand, generate, and parse human language.

[0021] "System" refers to a collection of software and hardware that provides these functions in an integrated manner. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0030] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0043] The present invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. The system utilizes natural language processing technology to reduce the burden on users and enable fast and effective email management.

[0044] Compose a new email

[0045] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology to analyze the entered information and automatically generate an appropriate subject and message body. The generated email is then sent to the device, where the user can view and edit it.

[0046] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hi, what are your plans for the weekend?" in the "Message" field, the server will generate the following subject and message, taking into account the appropriate context:

[0047] Subject: "Weekend plans"

[0048] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[0049] Create a reply email

[0050] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of past emails and generates a reply taking into account the conversation history. The generated reply email is sent to the terminal, where the user can view and edit it.

[0051] For example, when a user replies to an email, the server generates the following reply based on the most recent exchange:

[0052] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[0053] Email correction

[0054] When a user requests corrections to an email, the device sends the entered email text to the server, which detects grammatical and stylistic errors and generates suggested corrections. These suggested corrections are then sent to the device, where the user can review and apply them.

[0055] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[0056] Before: "I might be late for the meeting tomorrow."

[0057] Corrected: "I might be a little late for the meeting tomorrow."

[0058] Email Summary

[0059] When a user wants to summarize a long email, the device sends the email content to the server, which analyzes the content, extracts the key points, and generates a summary. The summary is then sent to the device so that the user can view it.

[0060] For example, if a long email begins with "Hello, regarding our recent meeting, the following progress has been made...", the server will extract the key elements and generate a summary like this:

[0061] Summary: "I'd like to report on the progress of our recent meeting..."

[0062] Thus, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails, thereby enabling users to manage emails appropriately and accurately with minimal effort.

[0063] The processing flow will be explained below.

[0064] Compose a new email

[0065] Step 1:

[0066] The user presses the new email creation button, and the device sends a new email creation request to the server.

[0067] Step 2:

[0068] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[0069] Step 3:

[0070] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[0071] Step 4:

[0072] When the user completes the input, the terminal transmits the input information to the server.

[0073] Step 5:

[0074] The server uses natural language processing (NLP) technology to analyze the input and generate an appropriate subject line and body of the message.

[0075] Step 6:

[0076] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[0077] Step 7:

[0078] The user reviews the generated email and edits it if necessary.

[0079] Create a reply email

[0080] Step 1:

[0081] The user presses the reply button, and the device sends a reply email creation request to the server.

[0082] Step 2:

[0083] The server retrieves the target email and the most recent exchanges from the user's past email database.

[0084] Step 3:

[0085] The server analyzes the contents of past emails and generates reply content taking into account the dialogue history.

[0086] Step 4:

[0087] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[0088] Step 5:

[0089] The user reviews the generated reply email and edits it if necessary.

[0090] Email correction

[0091] Step 1:

[0092] The user requests correction of an email. The terminal sends the email body to the server.

[0093] Step 2:

[0094] The server receives the email body and begins parsing it for grammatical and stylistic errors.

[0095] Step 3:

[0096] The server generates suggested fixes based on detected errors and improvements.

[0097] Step 4:

[0098] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[0099] Step 5:

[0100] The user checks the correction results and, if necessary, reflects the suggested corrections.

[0101] Email Summary

[0102] Step 1:

[0103] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[0104] Step 2:

[0105] The server receives the email body and begins parsing it to extract key points.

[0106] Step 3:

[0107] The server generates a summary based on the extracted key points.

[0108] Step 4:

[0109] The generated summary is sent to the terminal, which displays the summary to the user.

[0110] Step 5:

[0111] The user checks the summary results and edits them if necessary.

[0112] The above is a specific operation in each processing step.

[0113] Example 1

[0114] 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."

[0115] There are many challenges in current email management. Users must process a large number of emails every day, and tasks such as composing new emails, replying to emails, correcting grammar, and summarizing emails are extremely time-consuming and labor-intensive. Manually creating text, correcting errors, and summarizing emails are particularly burdensome and contribute to reduced work efficiency. Therefore, the present invention aims to automate these tasks, thereby reducing the burden on users and achieving efficient email management.

[0116] 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.

[0117] In this invention, the server includes means for automatically generating the subject and body of a new electronic communication based on user input, means for automatically generating the body of a reply electronic communication based on past electronic communication exchanges, means for detecting grammatical and formatting errors in the electronic communication and suggesting corrections, and means for extracting important points from the electronic communication and generating a summary, thereby enabling users to manage their electronic communications appropriately and accurately with little effort.

[0118] A "User" is a person who uses the System to create, reply to, correct, or summarize electronic communications.

[0119] "Input" is data or information that a user provides to a system.

[0120] "New electronic communications" refers to newly created emails or messages.

[0121] A "reply electronic communication" is a message sent in response to a previously received email or message.

[0122] "Natural language processing" refers to the techniques and methods that allow computers to understand and process human language.

[0123] "Grammar and formatting errors" refers to grammatical errors and improper formatting.

[0124] "Proposed Amendments" means proposed changes to correct grammatical or formatting errors in an Electronic Communication.

[0125] A "summary" is a concise summary of the key points extracted from a lengthy electronic communication.

[0126] A "generative AI model" is an algorithm or technology that uses artificial intelligence to generate text.

[0127] A "prompt sentence" is a guide sentence that prompts the user for input.

[0128] The present invention relates to a system that allows users to efficiently compose, reply, correct, and summarize electronic communications. The system utilizes natural language processing technology to reduce the burden on users and enable them to quickly and effectively manage electronic communications. The main hardware used includes the user's device (e.g., PC, smartphone) and a server. The software includes a natural language processing model (e.g., GPT-3 (registered trademark)) and a server-side program that processes communications.

[0129] Compose a new email

[0130] When a user composes a new email, the server displays prompts for the recipient, subject, and message on the user's device. For example, the prompts might include:

[0131] To create a new email, enter the recipient, subject, and message body. Example: To: example@example.com / Subject: Weekend plans / Message: Hello. What are your plans for the weekend?

[0132] The user enters the necessary information according to the instructions, and the device sends the information to the server. The server uses a natural language processing model (e.g., GPT-3) to analyze the entered information and automatically generate an appropriate subject and body. This generated email is then sent to the device, where the user can view and edit it.

[0133] Create a reply email

[0134] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the contents of the past emails and generates a reply taking into account the conversation history. For example, a prompt might include something like:

[0135] To reply to a specific email, click the reply button below and we'll generate an appropriate reply based on your previous interactions.

[0136] The generated reply email is sent to the terminal so that the user can view and edit it.

[0137] Email correction

[0138] When a user requests corrections to an email, the device sends the text of the email they typed to the server, which detects grammatical and stylistic errors and generates suggested corrections. For example, prompts might include:

[0139] I would like to correct the following email text, so please enter it into your device. Example: "I might be late for the meeting tomorrow."

[0140] The generated revision suggestions are sent to the terminal so that the user can review and apply them.

[0141] Email Summary

[0142] When a user summarizes a long email, the device sends the email content to the server, which analyzes it, extracts key points, and generates a summary. For example, prompts include:

[0143] If you have a long email and want to summarize it, enter the entire text and we'll extract the key points and summarize them for you.

[0144] The generated summary is sent to the terminal and made available for review by the user.

[0145] Thus, the present invention provides a system for users to efficiently compose, reply to, correct, and summarize electronic communications, thereby enabling users to manage their electronic communications appropriately and accurately with less effort.

[0146] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0147] Processing steps for creating a new email

[0148] Step 1: User accesses the compose email screen

[0149] The user opens their email client or web interface and accesses the screen for creating a new email.

[0150] Step 2: The server prompts you

[0151] The server displays a prompt on the user's terminal that includes input fields for "To", "Subject", and "Body".

[0152] Input: User's email creation request

[0153] Output: Show input fields

[0154] Step 3: User enters required information

[0155] The user enters the recipient, subject, and body of the message in the input fields provided.

[0156] Input: User input data (recipient, subject, body)

[0157] Output: Required information entered

[0158] Step 4: The device sends the information to the server

[0159] The terminal sends the user's input data to the server as an HTTP POST request.

[0160] Input: User input data (recipient, subject, body)

[0161] Output: HTTP POST request

[0162] Step 5: The server analyzes the input information using natural language processing technology

[0163] The server uses a natural language processing model (e.g., GPT-3) to analyze the user's input data.

[0164] Input: HTTP POST request data

[0165] Output: Parsed data

[0166] Step 6: The server auto-generates the subject and body

[0167] Based on the analysis results, the server automatically generates an appropriate subject and body of the message using a natural language processing model.

[0168] Input: Parsed data

[0169] Output: Auto-generated subject and body

[0170] Step 7: The generated email is sent to the device

[0171] The generated subject and body are sent to the terminal as an HTTP response.

[0172] Input: Auto-generated subject and body

[0173] Output: HTTP response

[0174] Step 8: User reviews and edits email

[0175] The user checks the generated email displayed on the terminal and manually edits it if necessary.

[0176] Input: HTTP response data (generated email)

[0177] Output: Final email reviewed and edited

[0178] Steps for creating a reply email

[0179] Step 1: User presses reply button

[0180] The user clicks the reply button for a particular received email.

[0181] Input: Click on reply button

[0182] Output: Reply request

[0183] Step 2: The server retrieves past emails

[0184] The server retrieves relevant past email conversations from a database.

[0185] Input: Reply Request

[0186] Output: Past email data

[0187] Step 3: The server analyzes the conversation history

[0188] The server analyzes the content of past emails using natural language processing technology.

[0189] Input: Past email data

[0190] Output: Analysis results

[0191] Step 4: Server generates reply content

[0192] The server uses the analysis results to create an appropriate reply using a generative AI model.

[0193] Input: Analysis results

[0194] Output: Reply content

[0195] Step 5: The generated reply email is sent to the device.

[0196] The server sends the generated reply content to the terminal as an HTTP response.

[0197] Input: Reply content

[0198] Output: HTTP response

[0199] Step 6: User reviews and edits reply email

[0200] The user checks the reply email displayed on the terminal and manually edits it if necessary.

[0201] Input: HTTP response data (reply email)

[0202] Output: Final reply email, reviewed and edited

[0203] Email correction process steps

[0204] Step 1: User enters email content

[0205] The user enters the email text they want to correct in the input field.

[0206] Input: Email body

[0207] Output: Input complete

[0208] Step 2: The device sends the email body to the server

[0209] The terminal sends the entered email body to the server as an HTTP POST request.

[0210] Input: Entered email body

[0211] Output: HTTP POST request

[0212] Step 3: The server detects grammar and style errors

[0213] The server uses natural language processing technology to detect grammatical and style errors.

[0214] Input: HTTP POST request data

[0215] Output: Error detection results

[0216] Step 4: Server generates proposed fixes

[0217] The server generates a correction suggestion based on the error detection results.

[0218] Input: Error detection result

[0219] Output: Revision proposal

[0220] Step 5: The proposed fix is ​​sent to the device

[0221] The server sends the proposed revision to the device as an HTTP response.

[0222] Input: Correction Suggestion

[0223] Output: HTTP response

[0224] Step 6: User reviews and implements proposed changes

[0225] The user checks the suggested changes displayed on the device and manually applies them if necessary.

[0226] Input: HTTP response data (correction)

[0227] Output: Final email confirmed and reflected

[0228] Mail Digest Processing Steps

[0229] Step 1: The user enters the email body they want to summarize.

[0230] The user enters the long email they want to summarize into the input field.

[0231] Input: Long email body

[0232] Output: Input complete

[0233] Step 2: The device sends the email content to the server

[0234] The terminal sends the entered long email text to the server as an HTTP POST request.

[0235] Input: The long email body you entered

[0236] Output: HTTP POST request

[0237] Step 3: The server analyzes the email content

[0238] The server uses natural language processing technology to analyze the email content and extract important points.

[0239] Input: HTTP POST request data

[0240] Output: Analysis results

[0241] Step 4: Server generates digest

[0242] The server generates a summary based on the analysis results.

[0243] Input: Analysis results

[0244] Output: Summary

[0245] Step 5: The summary is sent to the device

[0246] The server sends the generated summary to the terminal as an HTTP response.

[0247] Input: Summary

[0248] Output: HTTP response

[0249] Step 6: User confirms summary

[0250] The user checks the summary displayed on the terminal and manually corrects it if necessary.

[0251] Input: HTTP response data (summary)

[0252] Output: Checked and corrected summary

[0253] (Application example 1)

[0254] 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."

[0255] Modern electronic payment services require fast and accurate responses to customer inquiries. However, dealing with a large volume of email inquiries requires a great deal of time and effort from customer support staff, resulting in a high likelihood of response delays and misunderstandings. In these circumstances, a system for efficiently and effectively responding to emails is needed.

[0256] 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.

[0257] In this invention, the server includes means for automatically generating a subject and a body of a new electronic message based on a user's input, means for automatically generating a body of a reply electronic message based on past electronic message exchanges, means for detecting grammatical and stylistic errors in the electronic message and suggesting corrections, means for extracting important points of the electronic message and generating a summary, and means for efficiently responding to customer support inquiries in the electronic payment service using natural language processing technology, thereby enabling customer support personnel to manage electronic messages quickly and accurately with little effort and improving the quality of customer support.

[0258] "User" means any individual or end-user who uses the System to compose, reply to, correct, or summarize electronic messages.

[0259] "Input" refers to information that a user inputs into a terminal when creating or replying to an electronic message, specifically, the addressee, subject, body, etc.

[0260] "New electronic message" refers to an electronic message that a user has newly created and intends to send.

[0261] "Subject" means the title or headline portion of an electronic message that indicates the subject matter or purpose of the message.

[0262] "Body" means the detailed content or main message portion contained within an Electronic Message.

[0263] "Automatic generation" refers to the process by which a system automatically creates the subject line and body of an electronic message based on user input and historical data using natural language processing techniques.

[0264] "Past electronic message exchanges" refers to the history of past electronic messages exchanged between a user and other people or systems.

[0265] A "reply electronic message" refers to an electronic message created by a user in response to a received electronic message.

[0266] "Grammar and style errors" refers to grammatical errors and stylistic inadequacies in an electronic message.

[0267] "Suggested corrections" refer to corrected sentences suggested by the system based on grammatical and stylistic errors.

[0268] "Key Points" refers to information or points of particular importance within an electronic message.

[0269] "Summary" refers to a short summary of the important points extracted from a long electronic message.

[0270] "Natural language processing technology" refers to technology in the field of computer science for analyzing, understanding, and generating human language.

[0271] "Electronic payment service" refers to electronic payment procedures carried out when purchasing goods or using services via the Internet.

[0272] "Customer support" refers to the function of providing technical support and related information provided by a service provider to its customers.

[0273] "Inquiry" means a question or request submitted by a User to Customer Support.

[0274] "Responding efficiently" means that the system reduces the burden on users and allows them to quickly and effectively manage and respond to electronic messages.

[0275] This invention is a system for improving the efficiency of customer support inquiries in electronic payment services. The system automatically generates the subject and body of new electronic messages based on user input, automatically generates reply electronic messages based on past electronic message exchanges, detects and corrects grammatical and stylistic errors, and generates summaries that extract important points.

[0276] System Configuration

[0277] The main components of this system are a terminal, a server, and natural language processing technology including a generative AI model. The terminal is a hardware device used by a customer support representative, such as a smartphone, smart glasses, or a head-mounted display. The server has a generative AI model installed, which performs natural language processing using prompt sentences.

[0278] Explanation of program processing

[0279] The server performs the following steps and processes.

[0280] 1. Automatic generation of new electronic messages

[0281] The server uses a generative AI model to automatically generate an appropriate subject line and message body based on the recipient, subject, and message body entered by the user, for example, using the following prompt:

[0282] Subject: Weekend plans

[0283] Body: Hello. What are your plans for the weekend?

[0284] Generate a suitable subject and body:

[0285] The resulting output is an electronic message like this:

[0286] Subject: "Weekend plans"

[0287] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[0288] 2. Automatic generation of reply electronic messages

[0289] The server retrieves the user's past conversation history and uses a generative AI model to create a reply based on that. An example of a prompt is as follows:

[0290] Generate a reply email based on the following email exchange:

[0291] Hello, regarding the matter from the other day...

[0292] The resulting reply will look like this:

[0293] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[0294] 3. Grammar and style error detection and correction

[0295] The server analyzes the body of the electronic message entered by the user, detects grammatical and stylistic errors, and suggests corrections, for example using the following prompt:

[0296] Please correct the following email:

[0297] I might be late for the meeting tomorrow

[0298] The resulting output is the following proposed fix:

[0299] Before: "I might be late for the meeting tomorrow."

[0300] Corrected: "I might be a little late for the meeting tomorrow."

[0301] 4. Summary Generation

[0302] The server parses long electronic messages and generates summaries of key points. An example prompt is:

[0303] Please summarize the following lengthy email:

[0304] Hello, regarding the meeting we had the other day, we have made the following progress...

[0305] The resulting summary provides:

[0306] Summary: "I'd like to report on the progress of our recent meeting..."

[0307] This allows customer support agents to efficiently manage electronic messages and improve the quality of customer interactions using a server, terminal, and generative AI model.

[0308] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0309] Step 1:

[0310] A user inputs the recipient, subject, and body of a new electronic message into a terminal, and the terminal transmits the input information to a server.

[0311] Step 2:

[0312] The server invokes the generative AI model based on the received input information to automatically generate the subject and body of a new electronic message. The server inputs the following prompt text to the generative AI model:

[0313] Subject: [Subject]

[0314] Body: [Main text]

[0315] Generate a suitable subject and body:

[0316] The generative AI model analyzes the data, generates an appropriate subject line and body text, and sends it back to the server. The generated result is the appropriate subject line and body text.

[0317] Step 3:

[0318] The server sends the subject and body of the generated new electronic message to the terminal, which displays an interface for the user to review and edit. The user can use this interface to review and edit the generated content.

[0319] Step 4:

[0320] When a user wants to reply to a past electronic message, the user presses the reply button on the terminal. The terminal notifies the server of this operation and sends the reply email and past exchanges to the server. The input information includes the content of the past email.

[0321] Step 5:

[0322] The server calls the generative AI model based on the content of past emails it has received, and automatically generates a reply email message. The server inputs the following prompt text into the generative AI model:

[0323] Generate a reply email based on the following email exchange:

[0324] [Past email content]

[0325] The generative AI model performs analysis, generates an appropriate reply, and sends it back to the server. The generated result is the appropriate reply.

[0326] Step 6:

[0327] The server sends the generated reply electronic message to the terminal, and the terminal displays an interface for the user to view and edit the generated reply content. The user can use this interface to view and edit the generated reply content.

[0328] Step 7:

[0329] When a user requests grammar and style correction for an electronic message, the user inputs the target message body into the terminal, which then sends the input information, including the raw message body, to the server.

[0330] Step 8:

[0331] The server invokes a generative AI model based on the received message body, detects grammatical and stylistic errors, and suggests corrections. The server inputs the following prompt to the generative AI model:

[0332] Please correct the following email:

[0333] [Message body]

[0334] The generative AI model performs the analysis, generates suggested revisions, and sends them back to the server. The resulting revised text is obtained.

[0335] Step 9:

[0336] The server sends the generated revision proposal to the terminal, which displays the content of the revision to the user and provides an interface for review and editing. The user can use this interface to review and edit the revision proposal.

[0337] Step 10:

[0338] When a user wants to summarize a long electronic message, the user inputs the long message into the terminal, and the terminal transmits the input information to the server. The input information includes the content of the long message.

[0339] Step 11:

[0340] The server calls the generative AI model based on the received long message, extracts the key points, and generates a summary. The server inputs the following prompt to the generative AI model:

[0341] Please summarize the following lengthy email:

[0342] [Long message]

[0343] The generative AI model performs the analysis, generates a summary, and sends it back to the server. The resulting summary includes key points.

[0344] Step 12:

[0345] The server sends the generated summary to the terminal, which displays the summary to the user and provides an interface for confirmation, allowing the user to check the summary.

[0346] 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.

[0347] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize e-mails, and in particular, by combining an emotion engine, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion recognition technology to reduce the burden on the user and achieve fast and effective e-mail management.

[0348] Compose a new email

[0349] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and an emotion engine to analyze the entered information and automatically generate an appropriate subject and message body, taking into account the user's emotions. The generated email is then sent to the device, where the user can view and edit it.

[0350] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[0351] Subject: "Great weekend plans"

[0352] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[0353] Create a reply email

[0354] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[0355] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[0356] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[0357] Email correction

[0358] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[0359] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[0360] Before: "I might be late for the meeting tomorrow."

[0361] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[0362] Email Summary

[0363] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[0364] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[0365] Summary: "There was some positive progress in the meeting the other day."

[0366] In this way, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails. In addition, by combining it with an emotion engine, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[0367] The processing flow will be explained below.

[0368] Compose a new email

[0369] Step 1:

[0370] The user presses the button to create a new email. The device sends a request to the server to create a new email.

[0371] Step 2:

[0372] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[0373] Step 3:

[0374] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[0375] Step 4:

[0376] When the user completes the input, the terminal transmits the input information to the server.

[0377] Step 5:

[0378] The server uses an emotion engine to analyze emotions from the user's input information.

[0379] Step 6:

[0380] The server uses natural language processing (NLP) technology to analyze the input content and sentiment information and generate an appropriate subject line and body of the message.

[0381] Step 7:

[0382] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[0383] Step 8:

[0384] The user reviews the generated email and edits it if necessary.

[0385] Create a reply email

[0386] Step 1:

[0387] The user presses the reply button, and the device sends a reply email creation request to the server.

[0388] Step 2:

[0389] The server retrieves the target email and the most recent exchanges from the user's past email database.

[0390] Step 3:

[0391] The server analyzes the content of past emails and the corresponding emotional information.

[0392] Step 4:

[0393] The server uses an emotion engine to generate a reply that takes emotion information into consideration.

[0394] Step 5:

[0395] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[0396] Step 6:

[0397] The user reviews the generated reply email and edits it if necessary.

[0398] Email correction

[0399] Step 1:

[0400] The user requests correction of an email. The terminal sends the email body to the server.

[0401] Step 2:

[0402] The server receives the email body and analyzes the user's emotions using an emotion engine.

[0403] Step 3:

[0404] The server detects grammatical and stylistic errors and generates suggested revisions that take emotional information into account.

[0405] Step 4:

[0406] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[0407] Step 5:

[0408] The user checks the correction results and, if necessary, reflects the suggested corrections.

[0409] Email Summary

[0410] Step 1:

[0411] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[0412] Step 2:

[0413] The server receives the email body and analyzes the user's emotions using an emotion engine.

[0414] Step 3:

[0415] The server analyzes the email content and extracts important points, taking into account emotional information.

[0416] Step 4:

[0417] A summary is generated based on the extracted points and sent to the terminal.

[0418] Step 5:

[0419] The terminal displays the summary to the user, who then checks the summary and edits it as necessary.

[0420] The above is the specific operation of each processing step in the email management system combined with the emotion engine.

[0421] Example 2

[0422] 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."

[0423] In conventional email systems, users often have to spend a lot of time and effort composing, replying, correcting, and summarizing emails. Furthermore, it is difficult to express emotions in email content, which hinders smooth communication.

[0424] 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. In this invention, the server includes means for automatically generating a subject and a body of a new email based on a user's input, means for automatically generating a body of a reply message based on past message exchanges, means for detecting grammatical and stylistic errors in the message and suggesting corrections, means for extracting important points of the message and generating a summary, and means for generating message content taking the user's emotions into consideration using emotion analysis technology. This enables the user to efficiently compose, reply to, correct, and summarize emails, thereby realizing communication that takes the user's emotions into consideration.

[0425] The "means for automatically generating the subject and body of a new email" is a means for automatically creating an appropriate subject and body based on the content entered by the user using natural language processing technology.

[0426] The "means for automatically generating the body of a reply message" is a means for analyzing past message exchanges and automatically generating the body of a reply based on the analysis.

[0427] The "means for detecting grammatical and stylistic errors and suggesting corrections" is a means for detecting grammatical and stylistic errors in the email body and suggesting corrections to the user.

[0428] The "means for extracting important points and generating an outline" is a means for automatically extracting important points from the content of a long email and generating a summary that concisely summarizes the content.

[0429] "Means for generating message content taking into account the user's emotions using emotion analysis technology" refers to means for automatically generating message content that takes into account the user's emotions by analyzing the content entered by the user and using technology to infer the emotions behind it.

[0430] This invention relates to a system that allows users to efficiently compose, reply, correct, and summarize e-mails, and in particular, by combining emotion analysis technology, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective e-mail management.

[0431] First, we generate a program for this system. This program has the following functions:

[0432] 1. When a user composes a new email, the server automatically generates an appropriate subject and body based on the user's input.

[0433] 2. When creating a reply email, the server automatically generates the body of the reply email based on past message exchanges.

[0434] 3. When editing emails, the server detects grammatical and stylistic errors and suggests corrections.

[0435] 4. When summarizing an email, the server extracts the important points and generates a summary.

[0436] 5. At each processing step, emotion analysis technology is used to consider the user's emotions.

[0437] Compose a new email

[0438] When a user creates a new email, the server prompts the user for the recipient, subject, and body of the email. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and a sentiment analysis engine (e.g., Google® Cloud Natural Language API or IBM Watson® Tone Analyzer) to analyze the entered information and automatically generate an appropriate subject and body of the email, taking into account the user's emotions. The generated email is sent to the device, where the user can view and edit it.

[0439] Specific examples

[0440] Consider the case where the user enters the following:

[0441] To: example@example.com

[0442] Subject: Weekend plans

[0443] Body: Hello. What are your plans for the weekend?

[0444] The server receives this and automatically generates it as follows:

[0445] Subject: Great weekend plans

[0446] Body: Hello, I'd like to reminisce about our fun conversation the other day and confirm your plans for the weekend. I'm looking forward to it.

[0447] Create a reply email

[0448] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server uses the content of past emails and an emotion analysis engine to generate a reply, taking into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[0449] Specific examples

[0450] When a user replies to an email, the server generates the following reply based on recent interactions and emotional information:

[0451] Response: Hello, I'm a little surprised by your message the other day. Could you please confirm the following points again?

[0452] Email correction

[0453] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses a sentiment analysis engine to generate suggested corrections that take the user's feelings into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[0454] Specific examples

[0455] If the user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[0456] Before: I might be late for the meeting tomorrow.

[0457] Corrected: I might be a little late for tomorrow's meeting, but I hope you understand the reason.

[0458] Email Summary

[0459] When a user summarizes a long email, the device sends the email content to the server. The server analyzes the content and extracts key points using a sentiment analysis engine. A summary is generated based on the extracted points and sent to the device for the user to review.

[0460] Specific examples

[0461] If a long email says "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[0462] Summary: There was some positive progress in the recent meeting.

[0463] In this way, the present invention provides a system that allows users to efficiently compose, reply, correct, and summarize emails. In addition, by combining it with emotion analysis technology, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[0464] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0465] Compose a new email

[0466] Step 1:

[0467] The user opens a screen to compose a new email. The user clicks the "Create new email" button on the device. This causes the device to send a request to the server to compose a new email.

[0468] Input: User clicks "Create New Email" button

[0469] Output: A request to the server to create a new email

[0470] Step 2:

[0471] The server generates input prompts for "recipient," "subject," and "body" and sends them to the terminal. The terminal displays an input form to the user.

[0472] Input: Triggering prompt generation on the server side

[0473] Output: Input prompt sent to terminal

[0474] Step 3:

[0475] The user enters the "recipient," "subject," and "body" and presses the send button. The device sends the input data to the server.

[0476] Input: "To", "Subject", and "Body" entered by the user

[0477] Output: Input data sent to the server

[0478] Step 4:

[0479] The server analyzes the received data using NLP and sentiment analysis technologies. Specifically, the NLP model analyzes the text and understands its context and structure, while the sentiment engine determines the user's sentiment.

[0480] Input: Data entered by the user ("To", "Subject", "Body")

[0481] Data processing: Text analysis using NLP models, sentiment analysis using emotion engines

[0482] Output: Analysis results

[0483] Step 5:

[0484] Based on the analysis results, the server generates an appropriate subject line and body text. Specifically, it combines an NLP model and an emotion engine to automatically generate email content that reflects the user's intentions and emotions.

[0485] Input: Analysis results

[0486] Data processing: Email content generation using NLP models and emotion engines

[0487] Output: Auto-generated subject and body

[0488] Step 6:

[0489] The server sends the generated email to the terminal, and the terminal displays the contents of the generated email to the user so that the user can check and edit them.

[0490] Input: Auto-generated subject and body

[0491] Output: Email content sent to the terminal

[0492] Create a reply email

[0493] Step 1:

[0494] When the user presses the reply button for the email they want to reply to, the device sends a request to the server to create a reply email.

[0495] Input: The act of the user pressing the reply button

[0496] Output: Reply mail creation request to the server

[0497] Step 2:

[0498] The server retrieves the user's past email correspondence, retrieves the email to be replied to and the past conversation history, and prepares for analysis.

[0499] Input: The request sent to the server

[0500] Output: Retrieved email history

[0501] Step 3:

[0502] The server uses NLP and emotion analysis technologies to analyze the content and emotional information of past emails. Specifically, it uses NLP to analyze the structure of the text and an emotion engine to determine the importance of the emotion.

[0503] Input: Retrieved past email history

[0504] Data processing: Text and sentiment analysis using NLP models and sentiment engines

[0505] Output: Analysis results

[0506] Step 4:

[0507] The server generates an appropriate reply based on the analysis results, automatically generating a reply email that reflects the user's past emotional state.

[0508] Input: Analysis results

[0509] Data processing: Generate reply content using NLP models and emotion engines

[0510] Output: Auto-generated reply

[0511] Step 5:

[0512] The server sends the generated reply email to the terminal, and the terminal displays the generated reply content to the user so that the user can check and edit it.

[0513] Input: Auto-generated reply

[0514] Output: Reply email sent to the device

[0515] Email correction

[0516] Step 1:

[0517] The user opens an input screen to request corrections to the email content. The user enters the email body they want to correct and presses the send button. The device then sends this to the server.

[0518] Input: The email body entered by the user

[0519] Output: Correction request to the server

[0520] Step 2:

[0521] The server analyzes the email body received to detect grammatical and stylistic errors, using NLP models and a sentiment analysis engine.

[0522] Input: The email body entered by the user

[0523] Data processing: Error detection using NLP models and emotion engines

[0524] Output: Error detection results

[0525] Step 3:

[0526] The server generates suggestions for correcting grammar and style errors, providing sentiment-sensitive suggestions based on the analysis results.

[0527] Input: Error detection result

[0528] Data processing: Generating revision suggestions using NLP models and emotion engines

[0529] Output: Revision proposal

[0530] Step 4:

[0531] The server sends the proposed revisions to the terminal, which displays them to the user so that the user can confirm and apply them.

[0532] Input: Proposed fix

[0533] Output: Suggested fixes sent to the device

[0534] Email Summary

[0535] Step 1:

[0536] The user opens an input screen to request a summary of a long email. The user enters the email text they want to summarize and presses the send button. The terminal then sends this to the server.

[0537] Input: A long email body entered by the user

[0538] Output: Summary request to server

[0539] Step 2:

[0540] The server analyzes the email text to be summarized and extracts key points using NLP technology and a sentiment analysis engine.

[0541] Input: A long email body entered by the user

[0542] Data processing: Text analysis and key point extraction using NLP models and sentiment engines

[0543] Output: Analysis results

[0544] Step 3:

[0545] The server generates a summary based on the analysis results, taking into account emotional information and creating a concise summary.

[0546] Input: Analysis results

[0547] Data processing: Generating summaries using NLP models and emotion engines

[0548] Output: Auto-generated summary

[0549] Step 4:

[0550] The server sends the generated summary to the terminal, which displays the summary to the user for the user to review.

[0551] Input: Auto-generated summary

[0552] Output: Summary sent to terminal

[0553] (Application example 2)

[0554] 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."

[0555] Conventional email management systems provide support for users to efficiently compose, reply, correct, and summarize emails, but they do not support generating emails that take user emotions into account or real-time customer support. As a result, when effective communication with customers is required, especially in physical stores, it is difficult for store staff to respond quickly and appropriately. To solve this problem, a system is needed that includes email management that takes user emotions into account and real-time customer support functions.

[0556] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically generating a subject and body of a new email based on user input, means for automatically generating the body of a reply email based on past email exchanges, means for detecting grammatical and stylistic errors in the email and suggesting corrections, means for extracting important points from the email and generating a summary, means for automatically generating replies to customer inquiries using emotion analysis technology, and means for providing real-time customer service support based on customer emotions via a smart device. This enables users to efficiently manage their emails while providing quick and accurate responses to customers in physical stores.

[0557] A "user" is a person or entity who uses the system to compose, reply to, correct, or summarize emails.

[0558] A "new email" is an email that a user creates and sends for the first time.

[0559] The "subject" is the title part that briefly indicates the content of a new email or a reply email.

[0560] The "body" is the part where the main content of the email is written.

[0561] A "server" is a computer system that processes information entered by users and performs various automatic generation and analysis.

[0562] "Natural language processing" is a technology that analyzes and understands human language, and is primarily used for automatic text generation and sentiment analysis.

[0563] "Emotion analysis technology" is a technology that analyzes a user's emotional state from text or voice and extracts that information.

[0564] "Past email exchanges" refers to the content of emails exchanged between the user and other people in the past.

[0565] A "reply email" is an email sent in response to a received email.

[0566] A "grammatical error" is an error in the language rules contained in the body of an email.

[0567] A "style error" is an error in format or expression contained within the body of an email.

[0568] "Suggested fixes" are suggestions for correcting grammatical or stylistic errors.

[0569] A "summary" is a simplified version of a long email message that extracts the important points.

[0570] A "customer" is a person who visits a physical store to make an inquiry or purchase.

[0571] A "smart device" is an electronic device that has Internet connectivity and can run various applications.

[0572] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. In particular, by combining an emotion engine, it enables email management that takes user emotions into consideration and customer service support in physical stores. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective email and customer service.

[0573] Compose a new email

[0574] When a user composes a new email, they follow the prompts displayed on their device to enter the recipient, subject, and message body. The device then sends this information to the server. The server then uses natural language processing technology and an emotion engine to analyze the entered information and generate an appropriate subject and message body, taking the user's emotions into account. The generated email is then sent to the device, where it can be viewed and edited by the user.

[0575] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[0576] Subject: "Great weekend plans"

[0577] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[0578] Create a reply email

[0579] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[0580] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[0581] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[0582] Email correction

[0583] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[0584] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[0585] Before: "I might be late for the meeting tomorrow."

[0586] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[0587] Email Summary

[0588] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[0589] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[0590] Summary: "There was some positive progress in the meeting the other day."

[0591] Physical store applications

[0592] In physical stores, store clerks use smart devices (e.g., smart glasses) to assist customers. When a customer sends an inquiry email to a store clerk, the store clerk checks the email via the smart device, and the server performs sentiment analysis to generate an appropriate reply.

[0593] For example, if a customer asks, "Please tell me about new products arriving," the server will generate the following reply if it detects a positive emotion:

[0594] Response: "Thank you for your question about the new product arrivals. We expect new products to arrive this weekend."

[0595] This system allows users to manage emails more efficiently and also enables them to respond to customers more effectively in physical stores. In addition, the emotion engine enables communication that accurately reflects the emotions of users and customers.

[0596] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0597] Step 1:

[0598] The user enters the "recipient," "subject," and "body" into the terminal. The terminal sends this to the server. Input data: the recipient, subject, and body entered by the user. Output data: the input data sent to the server.

[0599] Step 2:

[0600] The server analyzes the received input data using natural language processing and sentiment analysis technologies. Input data: User's address, subject, and text. Output data: Analyzed sentiment information and natural language processing results.

[0601] Step 3:

[0602] Based on the analysis results, the server uses a generative AI model to automatically generate the subject and body of a new email. Input data: Analyzed emotional information and natural language processing results. Output data: Automatically generated subject and body.

[0603] Step 4:

[0604] The generated email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated email sent from the server. Output data: Final email edited by the user.

[0605] Step 5:

[0606] When a user presses the reply button for a specific email, the device sends that information to the server. Input data: Email information from which the reply was made. Output data: Reply information sent to the server.

[0607] Step 6:

[0608] The server acquires past email exchanges and analyzes them using emotion analysis technology. Input data: Past email exchanges. Output data: Analyzed dialogue history and emotion information.

[0609] Step 7:

[0610] Based on the analysis results, the server uses a generative AI model to automatically generate the reply email body. Input data: Analyzed dialogue history and emotional information. Output data: Automatically generated reply body.

[0611] Step 8:

[0612] The generated reply email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated reply email. Output data: Final reply email edited by the user.

[0613] Step 9:

[0614] When a user requests corrections to an email, the terminal sends the email body to the server. Input data: The email body entered by the user. Output data: The email body sent to the server.

[0615] Step 10:

[0616] The server detects grammatical and style errors and generates correction suggestions using sentiment analysis technology. Input data: User's email body. Output data: Detected error information and correction suggestions.

[0617] Step 11:

[0618] The generated correction proposal is sent to the terminal, where the user can check it and apply it as necessary. Input data: The generated correction proposal. Output data: The correction email applied by the user.

[0619] Step 12:

[0620] When a user requests a summary of an email, the terminal sends the email content to the server. Input data: Content of the long email. Output data: Content of the email sent to the server.

[0621] Step 13:

[0622] The server analyzes the email content, extracts key points using sentiment analysis technology, and generates a summary. Input data: Long email content. Output data: Extracted key points and generated summary.

[0623] Step 14:

[0624] The generated summary is sent to the terminal and confirmed by the user. Input data: Generated summary. Output data: Summary confirmed by the user.

[0625] Step 15:

[0626] When a user uses a smart device to deal with a customer in a physical store, the device receives the customer's inquiry email and sends it to the server. Input data: Customer's inquiry email. Output data: Inquiry email sent to the server.

[0627] Step 16:

[0628] The server uses emotion analysis technology to analyze customer emotions and generate appropriate replies. Input data: Customer inquiry email and emotion analysis results. Output data: Automatically generated replies.

[0629] Step 17:

[0630] The generated reply is sent to the smart device, and the store clerk responds appropriately to the customer. Input data: Automatically generated reply. Output data: Response provided by the store clerk to the customer.

[0631] 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.

[0632] 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.

[0633] 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.

[0634] [Second embodiment]

[0635] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0636] 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.

[0637] 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).

[0638] 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.

[0639] 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.

[0640] 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).

[0641] 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.

[0642] 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.

[0643] 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.

[0644] 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.

[0645] 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.

[0646] 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."

[0647] The present invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. The system utilizes natural language processing technology to reduce the burden on users and enable fast and effective email management.

[0648] Compose a new email

[0649] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology to analyze the entered information and automatically generate an appropriate subject and message body. The generated email is then sent to the device, where the user can view and edit it.

[0650] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hi, what are your plans for the weekend?" in the "Message" field, the server will generate the following subject and message, taking into account the appropriate context:

[0651] Subject: "Weekend plans"

[0652] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[0653] Create a reply email

[0654] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of past emails and generates a reply taking into account the conversation history. The generated reply email is sent to the terminal, where the user can view and edit it.

[0655] For example, when a user replies to an email, the server generates the following reply based on the most recent exchange:

[0656] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[0657] Email correction

[0658] When a user requests corrections to an email, the device sends the entered email text to the server, which detects grammatical and stylistic errors and generates suggested corrections. These suggested corrections are then sent to the device, where the user can review and apply them.

[0659] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[0660] Before: "I might be late for the meeting tomorrow."

[0661] Corrected: "I might be a little late for the meeting tomorrow."

[0662] Email Summary

[0663] When a user wants to summarize a long email, the device sends the email content to the server, which analyzes the content, extracts the key points, and generates a summary. The summary is then sent to the device so that the user can view it.

[0664] For example, if a long email begins with "Hello, regarding our recent meeting, the following progress has been made...", the server will extract the key elements and generate a summary like this:

[0665] Summary: "I'd like to report on the progress of our recent meeting..."

[0666] Thus, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails, thereby enabling users to manage emails appropriately and accurately with minimal effort.

[0667] The processing flow will be explained below.

[0668] Compose a new email

[0669] Step 1:

[0670] The user presses the new email creation button, and the device sends a new email creation request to the server.

[0671] Step 2:

[0672] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[0673] Step 3:

[0674] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[0675] Step 4:

[0676] When the user completes the input, the terminal transmits the input information to the server.

[0677] Step 5:

[0678] The server uses natural language processing (NLP) technology to analyze the input and generate an appropriate subject line and body of the message.

[0679] Step 6:

[0680] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[0681] Step 7:

[0682] The user reviews the generated email and edits it if necessary.

[0683] Create a reply email

[0684] Step 1:

[0685] The user presses the reply button, and the device sends a reply email creation request to the server.

[0686] Step 2:

[0687] The server retrieves the target email and the most recent exchanges from the user's past email database.

[0688] Step 3:

[0689] The server analyzes the contents of past emails and generates reply content taking into account the dialogue history.

[0690] Step 4:

[0691] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[0692] Step 5:

[0693] The user reviews the generated reply email and edits it if necessary.

[0694] Email correction

[0695] Step 1:

[0696] The user requests correction of an email. The terminal sends the email body to the server.

[0697] Step 2:

[0698] The server receives the email body and begins parsing it for grammatical and stylistic errors.

[0699] Step 3:

[0700] The server generates suggested fixes based on detected errors and improvements.

[0701] Step 4:

[0702] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[0703] Step 5:

[0704] The user checks the correction results and, if necessary, reflects the suggested corrections.

[0705] Email Summary

[0706] Step 1:

[0707] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[0708] Step 2:

[0709] The server receives the email body and begins parsing it to extract key points.

[0710] Step 3:

[0711] The server generates a summary based on the extracted key points.

[0712] Step 4:

[0713] The generated summary is sent to the terminal, which displays the summary to the user.

[0714] Step 5:

[0715] The user checks the summary results and edits them if necessary.

[0716] The above is a specific operation in each processing step.

[0717] Example 1

[0718] 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."

[0719] There are many challenges in current email management. Users must process a large number of emails every day, and tasks such as composing new emails, replying to emails, correcting grammar, and summarizing emails are extremely time-consuming and labor-intensive. Manually creating text, correcting errors, and summarizing emails are particularly burdensome and contribute to reduced work efficiency. Therefore, the present invention aims to automate these tasks, thereby reducing the burden on users and achieving efficient email management.

[0720] 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.

[0721] In this invention, the server includes means for automatically generating the subject and body of a new electronic communication based on user input, means for automatically generating the body of a reply electronic communication based on past electronic communication exchanges, means for detecting grammatical and formatting errors in the electronic communication and suggesting corrections, and means for extracting important points from the electronic communication and generating a summary, thereby enabling users to manage their electronic communications appropriately and accurately with little effort.

[0722] A "User" is a person who uses the System to create, reply to, correct, or summarize electronic communications.

[0723] "Input" is data or information that a user provides to a system.

[0724] "New electronic communications" refers to newly created emails or messages.

[0725] A "reply electronic communication" is a message sent in response to a previously received email or message.

[0726] "Natural language processing" refers to the techniques and methods that allow computers to understand and process human language.

[0727] "Grammar and formatting errors" refers to grammatical errors and improper formatting.

[0728] "Proposed Amendments" means proposed changes to correct grammatical or formatting errors in an Electronic Communication.

[0729] A "summary" is a concise summary of the key points extracted from a lengthy electronic communication.

[0730] A "generative AI model" is an algorithm or technology that uses artificial intelligence to generate text.

[0731] A "prompt sentence" is a guide sentence that prompts the user for input.

[0732] The present invention relates to a system that allows users to efficiently compose, reply, correct, and summarize electronic communications. The system utilizes natural language processing technology to reduce the burden on users and enable them to quickly and effectively manage their electronic communications. The main hardware used includes the user's device (e.g., PC, smartphone) and a server. The software includes a natural language processing model (e.g., GPT-3) and a server-side program that processes communications.

[0733] Compose a new email

[0734] When a user composes a new email, the server displays prompts for the recipient, subject, and message on the user's device. For example, the prompts might include:

[0735] To create a new email, enter the recipient, subject, and message body. Example: To: example@example.com / Subject: Weekend plans / Message: Hello. What are your plans for the weekend?

[0736] The user enters the necessary information according to the instructions, and the device sends the information to the server. The server uses a natural language processing model (e.g., GPT-3) to analyze the entered information and automatically generate an appropriate subject and body. This generated email is then sent to the device, where the user can view and edit it.

[0737] Create a reply email

[0738] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the contents of the past emails and generates a reply taking into account the conversation history. For example, a prompt might include something like:

[0739] To reply to a specific email, click the reply button below and we'll generate an appropriate reply based on your previous interactions.

[0740] The generated reply email is sent to the terminal so that the user can view and edit it.

[0741] Email correction

[0742] When a user requests corrections to an email, the device sends the text of the email they typed to the server, which detects grammatical and stylistic errors and generates suggested corrections. For example, prompts might include:

[0743] I would like to correct the following email text, so please enter it into your device. Example: "I might be late for the meeting tomorrow."

[0744] The generated revision suggestions are sent to the terminal so that the user can review and apply them.

[0745] Email Summary

[0746] When a user summarizes a long email, the device sends the email content to the server, which analyzes it, extracts key points, and generates a summary. For example, prompts include:

[0747] If you have a long email and want to summarize it, enter the entire text and we'll extract the key points and summarize them for you.

[0748] The generated summary is sent to the terminal and made available for review by the user.

[0749] Thus, the present invention provides a system for users to efficiently compose, reply to, correct, and summarize electronic communications, thereby enabling users to manage their electronic communications appropriately and accurately with less effort.

[0750] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0751] Processing steps for creating a new email

[0752] Step 1: User accesses the compose email screen

[0753] The user opens their email client or web interface and accesses the screen for creating a new email.

[0754] Step 2: The server prompts you

[0755] The server displays a prompt on the user's terminal that includes input fields for "To", "Subject", and "Body".

[0756] Input: User's email creation request

[0757] Output: Show input fields

[0758] Step 3: User enters required information

[0759] The user enters the recipient, subject, and body of the message in the input fields provided.

[0760] Input: User input data (recipient, subject, body)

[0761] Output: Required information entered

[0762] Step 4: The device sends the information to the server

[0763] The terminal sends the user's input data to the server as an HTTP POST request.

[0764] Input: User input data (recipient, subject, body)

[0765] Output: HTTP POST request

[0766] Step 5: The server analyzes the input information using natural language processing technology

[0767] The server uses a natural language processing model (e.g., GPT-3) to analyze the user's input data.

[0768] Input: HTTP POST request data

[0769] Output: Parsed data

[0770] Step 6: The server auto-generates the subject and body

[0771] Based on the analysis results, the server automatically generates an appropriate subject and body of the message using a natural language processing model.

[0772] Input: Parsed data

[0773] Output: Auto-generated subject and body

[0774] Step 7: The generated email is sent to the device

[0775] The generated subject and body are sent to the terminal as an HTTP response.

[0776] Input: Auto-generated subject and body

[0777] Output: HTTP response

[0778] Step 8: User reviews and edits email

[0779] The user checks the generated email displayed on the terminal and manually edits it if necessary.

[0780] Input: HTTP response data (generated email)

[0781] Output: Final email reviewed and edited

[0782] Steps for creating a reply email

[0783] Step 1: User presses reply button

[0784] The user clicks the reply button for a particular received email.

[0785] Input: Click on reply button

[0786] Output: Reply request

[0787] Step 2: The server retrieves past emails

[0788] The server retrieves relevant past email conversations from a database.

[0789] Input: Reply Request

[0790] Output: Past email data

[0791] Step 3: The server analyzes the conversation history

[0792] The server analyzes the content of past emails using natural language processing technology.

[0793] Input: Past email data

[0794] Output: Analysis results

[0795] Step 4: Server generates reply content

[0796] The server uses the analysis results to create an appropriate reply using a generative AI model.

[0797] Input: Analysis results

[0798] Output: Reply content

[0799] Step 5: The generated reply email is sent to the device.

[0800] The server sends the generated reply content to the terminal as an HTTP response.

[0801] Input: Reply content

[0802] Output: HTTP response

[0803] Step 6: User reviews and edits reply email

[0804] The user checks the reply email displayed on the terminal and manually edits it if necessary.

[0805] Input: HTTP response data (reply email)

[0806] Output: Final reply email, reviewed and edited

[0807] Email correction process steps

[0808] Step 1: User enters email content

[0809] The user enters the email text they want to correct in the input field.

[0810] Input: Email body

[0811] Output: Input complete

[0812] Step 2: The device sends the email body to the server

[0813] The terminal sends the entered email body to the server as an HTTP POST request.

[0814] Input: Entered email body

[0815] Output: HTTP POST request

[0816] Step 3: The server detects grammar and style errors

[0817] The server uses natural language processing technology to detect grammatical and style errors.

[0818] Input: HTTP POST request data

[0819] Output: Error detection results

[0820] Step 4: Server generates proposed fixes

[0821] The server generates a correction suggestion based on the error detection results.

[0822] Input: Error detection result

[0823] Output: Revision proposal

[0824] Step 5: The proposed fix is ​​sent to the device

[0825] The server sends the proposed revision to the device as an HTTP response.

[0826] Input: Correction Suggestion

[0827] Output: HTTP response

[0828] Step 6: User reviews and implements proposed changes

[0829] The user checks the suggested changes displayed on the device and manually applies them if necessary.

[0830] Input: HTTP response data (correction)

[0831] Output: Final email confirmed and reflected

[0832] Mail Digest Processing Steps

[0833] Step 1: The user enters the email body they want to summarize.

[0834] The user enters the long email they want to summarize into the input field.

[0835] Input: Long email body

[0836] Output: Input complete

[0837] Step 2: The device sends the email content to the server

[0838] The terminal sends the entered long email text to the server as an HTTP POST request.

[0839] Input: The long email body you entered

[0840] Output: HTTP POST request

[0841] Step 3: The server analyzes the email content

[0842] The server uses natural language processing technology to analyze the email content and extract important points.

[0843] Input: HTTP POST request data

[0844] Output: Analysis results

[0845] Step 4: Server generates digest

[0846] The server generates a summary based on the analysis results.

[0847] Input: Analysis results

[0848] Output: Summary

[0849] Step 5: The summary is sent to the device

[0850] The server sends the generated summary to the terminal as an HTTP response.

[0851] Input: Summary

[0852] Output: HTTP response

[0853] Step 6: User confirms summary

[0854] The user checks the summary displayed on the terminal and manually corrects it if necessary.

[0855] Input: HTTP response data (summary)

[0856] Output: Checked and corrected summary

[0857] (Application example 1)

[0858] 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."

[0859] Modern electronic payment services require fast and accurate responses to customer inquiries. However, dealing with a large volume of email inquiries requires a great deal of time and effort from customer support staff, resulting in a high likelihood of response delays and misunderstandings. In these circumstances, a system for efficiently and effectively responding to emails is needed.

[0860] 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.

[0861] In this invention, the server includes means for automatically generating a subject and a body of a new electronic message based on a user's input, means for automatically generating a body of a reply electronic message based on past electronic message exchanges, means for detecting grammatical and stylistic errors in the electronic message and suggesting corrections, means for extracting important points of the electronic message and generating a summary, and means for efficiently responding to customer support inquiries in the electronic payment service using natural language processing technology, thereby enabling customer support personnel to manage electronic messages quickly and accurately with little effort and improving the quality of customer support.

[0862] "User" means any individual or end-user who uses the System to compose, reply to, correct, or summarize electronic messages.

[0863] "Input" refers to information that a user inputs into a terminal when creating or replying to an electronic message, specifically, the addressee, subject, body, etc.

[0864] "New electronic message" refers to an electronic message that a user has newly created and intends to send.

[0865] "Subject" means the title or headline portion of an electronic message that indicates the subject matter or purpose of the message.

[0866] "Body" means the detailed content or main message portion contained within an Electronic Message.

[0867] "Automatic generation" refers to the process by which a system automatically creates the subject line and body of an electronic message based on user input and historical data using natural language processing techniques.

[0868] "Past electronic message exchanges" refers to the history of past electronic messages exchanged between a user and other people or systems.

[0869] A "reply electronic message" refers to an electronic message created by a user in response to a received electronic message.

[0870] "Grammar and style errors" refers to grammatical errors and stylistic inadequacies in an electronic message.

[0871] "Suggested corrections" refer to corrected sentences suggested by the system based on grammatical and stylistic errors.

[0872] "Key Points" refers to information or points of particular importance within an electronic message.

[0873] "Summary" refers to a short summary of the important points extracted from a long electronic message.

[0874] "Natural language processing technology" refers to technology in the field of computer science for analyzing, understanding, and generating human language.

[0875] "Electronic payment service" refers to electronic payment procedures carried out when purchasing goods or using services via the Internet.

[0876] "Customer support" refers to the function of providing technical support and related information provided by a service provider to its customers.

[0877] "Inquiry" means a question or request submitted by a User to Customer Support.

[0878] "Responding efficiently" means that the system reduces the burden on users and allows them to quickly and effectively manage and respond to electronic messages.

[0879] This invention is a system for improving the efficiency of customer support inquiries in electronic payment services. The system automatically generates the subject and body of new electronic messages based on user input, automatically generates reply electronic messages based on past electronic message exchanges, detects and corrects grammatical and stylistic errors, and generates summaries that extract important points.

[0880] System Configuration

[0881] The main components of this system are a terminal, a server, and natural language processing technology including a generative AI model. The terminal is a hardware device used by a customer support representative, such as a smartphone, smart glasses, or a head-mounted display. The server has a generative AI model installed, which performs natural language processing using prompt sentences.

[0882] Explanation of program processing

[0883] The server performs the following steps and processes.

[0884] 1. Automatic generation of new electronic messages

[0885] The server uses a generative AI model to automatically generate an appropriate subject line and message body based on the recipient, subject, and message body entered by the user, for example, using the following prompt:

[0886] Subject: Weekend plans

[0887] Body: Hello. What are your plans for the weekend?

[0888] Generate a suitable subject and body:

[0889] The resulting output is an electronic message like this:

[0890] Subject: "Weekend plans"

[0891] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[0892] 2. Automatic generation of reply electronic messages

[0893] The server retrieves the user's past conversation history and uses a generative AI model to create a reply based on that. An example of a prompt is as follows:

[0894] Generate a reply email based on the following email exchange:

[0895] Hello, regarding the matter from the other day...

[0896] The resulting reply will look like this:

[0897] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[0898] 3. Grammar and style error detection and correction

[0899] The server analyzes the body of the electronic message entered by the user, detects grammatical and stylistic errors, and suggests corrections, for example using the following prompt:

[0900] Please correct the following email:

[0901] I might be late for the meeting tomorrow

[0902] The resulting output is the following proposed fix:

[0903] Before: "I might be late for the meeting tomorrow."

[0904] Corrected: "I might be a little late for the meeting tomorrow."

[0905] 4. Summary Generation

[0906] The server parses long electronic messages and generates summaries of key points. An example prompt is:

[0907] Please summarize the following lengthy email:

[0908] Hello, regarding the meeting we had the other day, we have made the following progress...

[0909] The resulting summary provides:

[0910] Summary: "I'd like to report on the progress of our recent meeting..."

[0911] This allows customer support agents to efficiently manage electronic messages and improve the quality of customer interactions using a server, terminal, and generative AI model.

[0912] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0913] Step 1:

[0914] A user inputs the recipient, subject, and body of a new electronic message into a terminal, and the terminal transmits the input information to a server.

[0915] Step 2:

[0916] The server invokes the generative AI model based on the received input information to automatically generate the subject and body of a new electronic message. The server inputs the following prompt text to the generative AI model:

[0917] Subject: [Subject]

[0918] Body: [Main text]

[0919] Generate a suitable subject and body:

[0920] The generative AI model analyzes the data, generates an appropriate subject line and body text, and sends it back to the server. The generated result is the appropriate subject line and body text.

[0921] Step 3:

[0922] The server sends the subject and body of the generated new electronic message to the terminal, which displays an interface for the user to review and edit. The user can use this interface to review and edit the generated content.

[0923] Step 4:

[0924] When a user wants to reply to a past electronic message, the user presses the reply button on the terminal. The terminal notifies the server of this operation and sends the reply email and past exchanges to the server. The input information includes the content of the past email.

[0925] Step 5:

[0926] The server calls the generative AI model based on the content of past emails it has received, and automatically generates a reply email message. The server inputs the following prompt text into the generative AI model:

[0927] Generate a reply email based on the following email exchange:

[0928] [Past email content]

[0929] The generative AI model performs analysis, generates an appropriate reply, and sends it back to the server. The generated result is the appropriate reply.

[0930] Step 6:

[0931] The server sends the generated reply electronic message to the terminal, and the terminal displays an interface for the user to view and edit the generated reply content. The user can use this interface to view and edit the generated reply content.

[0932] Step 7:

[0933] When a user requests grammar and style correction for an electronic message, the user inputs the target message body into the terminal, which then sends the input information, including the raw message body, to the server.

[0934] Step 8:

[0935] The server invokes a generative AI model based on the received message body, detects grammatical and stylistic errors, and suggests corrections. The server inputs the following prompt to the generative AI model:

[0936] Please correct the following email:

[0937] [Message body]

[0938] The generative AI model performs the analysis, generates suggested revisions, and sends them back to the server. The resulting revised text is obtained.

[0939] Step 9:

[0940] The server sends the generated revision proposal to the terminal, which displays the content of the revision to the user and provides an interface for review and editing. The user can use this interface to review and edit the revision proposal.

[0941] Step 10:

[0942] When a user wants to summarize a long electronic message, the user inputs the long message into the terminal, and the terminal transmits the input information to the server. The input information includes the content of the long message.

[0943] Step 11:

[0944] The server calls the generative AI model based on the received long message, extracts the key points, and generates a summary. The server inputs the following prompt to the generative AI model:

[0945] Please summarize the following lengthy email:

[0946] [Long message]

[0947] The generative AI model performs the analysis, generates a summary, and sends it back to the server. The resulting summary includes key points.

[0948] Step 12:

[0949] The server sends the generated summary to the terminal, which displays the summary to the user and provides an interface for confirmation, allowing the user to check the summary.

[0950] 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.

[0951] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize e-mails, and in particular, by combining an emotion engine, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion recognition technology to reduce the burden on the user and achieve fast and effective e-mail management.

[0952] Compose a new email

[0953] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and an emotion engine to analyze the entered information and automatically generate an appropriate subject and message body, taking into account the user's emotions. The generated email is then sent to the device, where the user can view and edit it.

[0954] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[0955] Subject: "Great weekend plans"

[0956] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[0957] Create a reply email

[0958] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[0959] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[0960] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[0961] Email correction

[0962] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[0963] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[0964] Before: "I might be late for the meeting tomorrow."

[0965] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[0966] Email Summary

[0967] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[0968] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[0969] Summary: "There was some positive progress in the meeting the other day."

[0970] In this way, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails. In addition, by combining it with an emotion engine, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[0971] The processing flow will be explained below.

[0972] Compose a new email

[0973] Step 1:

[0974] The user presses the button to create a new email. The device sends a request to the server to create a new email.

[0975] Step 2:

[0976] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[0977] Step 3:

[0978] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[0979] Step 4:

[0980] When the user completes the input, the terminal transmits the input information to the server.

[0981] Step 5:

[0982] The server uses an emotion engine to analyze emotions from the user's input information.

[0983] Step 6:

[0984] The server uses natural language processing (NLP) technology to analyze the input content and sentiment information and generate an appropriate subject line and body of the message.

[0985] Step 7:

[0986] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[0987] Step 8:

[0988] The user reviews the generated email and edits it if necessary.

[0989] Create a reply email

[0990] Step 1:

[0991] The user presses the reply button, and the device sends a reply email creation request to the server.

[0992] Step 2:

[0993] The server retrieves the target email and the most recent exchanges from the user's past email database.

[0994] Step 3:

[0995] The server analyzes the content of past emails and the corresponding emotional information.

[0996] Step 4:

[0997] The server uses an emotion engine to generate a reply that takes emotion information into consideration.

[0998] Step 5:

[0999] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[1000] Step 6:

[1001] The user reviews the generated reply email and edits it if necessary.

[1002] Email correction

[1003] Step 1:

[1004] The user requests correction of an email. The terminal sends the email body to the server.

[1005] Step 2:

[1006] The server receives the email body and analyzes the user's emotions using an emotion engine.

[1007] Step 3:

[1008] The server detects grammatical and stylistic errors and generates suggested revisions that take emotional information into account.

[1009] Step 4:

[1010] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[1011] Step 5:

[1012] The user checks the correction results and, if necessary, reflects the suggested corrections.

[1013] Email Summary

[1014] Step 1:

[1015] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[1016] Step 2:

[1017] The server receives the email body and analyzes the user's emotions using an emotion engine.

[1018] Step 3:

[1019] The server analyzes the email content and extracts important points, taking into account emotional information.

[1020] Step 4:

[1021] A summary is generated based on the extracted points and sent to the terminal.

[1022] Step 5:

[1023] The terminal displays the summary to the user, who then checks the summary and edits it as necessary.

[1024] The above is the specific operation of each processing step in the email management system combined with the emotion engine.

[1025] Example 2

[1026] 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."

[1027] In conventional email systems, users often have to spend a lot of time and effort composing, replying, correcting, and summarizing emails. Furthermore, it is difficult to express emotions in email content, which hinders smooth communication.

[1028] 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. In this invention, the server includes means for automatically generating a subject and a body of a new email based on a user's input, means for automatically generating a body of a reply message based on past message exchanges, means for detecting grammatical and stylistic errors in the message and suggesting corrections, means for extracting important points of the message and generating a summary, and means for generating message content taking the user's emotions into consideration using emotion analysis technology. This enables the user to efficiently compose, reply to, correct, and summarize emails, thereby realizing communication that takes the user's emotions into consideration.

[1029] The "means for automatically generating the subject and body of a new email" is a means for automatically creating an appropriate subject and body based on the content entered by the user using natural language processing technology.

[1030] The "means for automatically generating the body of a reply message" is a means for analyzing past message exchanges and automatically generating the body of a reply based on the analysis.

[1031] The "means for detecting grammatical and stylistic errors and suggesting corrections" is a means for detecting grammatical and stylistic errors in the email body and suggesting corrections to the user.

[1032] The "means for extracting important points and generating an outline" is a means for automatically extracting important points from the content of a long email and generating a summary that concisely summarizes the content.

[1033] "Means for generating message content taking into account the user's emotions using emotion analysis technology" refers to means for automatically generating message content that takes into account the user's emotions by analyzing the content entered by the user and using technology to infer the emotions behind it.

[1034] This invention relates to a system that allows users to efficiently compose, reply, correct, and summarize e-mails, and in particular, by combining emotion analysis technology, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective e-mail management.

[1035] First, we generate a program for this system. This program has the following functions:

[1036] 1. When a user composes a new email, the server automatically generates an appropriate subject and body based on the user's input.

[1037] 2. When creating a reply email, the server automatically generates the body of the reply email based on past message exchanges.

[1038] 3. When editing emails, the server detects grammatical and stylistic errors and suggests corrections.

[1039] 4. When summarizing an email, the server extracts the important points and generates a summary.

[1040] 5. At each processing step, emotion analysis technology is used to consider the user's emotions.

[1041] Compose a new email

[1042] When a user creates a new email, the server prompts the user for the recipient, subject, and body of the email. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and a sentiment analysis engine (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to analyze the entered information and automatically generate an appropriate subject and body of the email, taking into account the user's sentiment. The generated email is then sent to the device, where the user can view and edit it.

[1043] Specific examples

[1044] Consider the case where the user enters the following:

[1045] To: example@example.com

[1046] Subject: Weekend plans

[1047] Body: Hello. What are your plans for the weekend?

[1048] The server receives this and automatically generates it as follows:

[1049] Subject: Great weekend plans

[1050] Body: Hello, I'd like to reminisce about our fun conversation the other day and confirm your plans for the weekend. I'm looking forward to it.

[1051] Create a reply email

[1052] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server uses the content of past emails and an emotion analysis engine to generate a reply, taking into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1053] Specific examples

[1054] When a user replies to an email, the server generates the following reply based on recent interactions and emotional information:

[1055] Response: Hello, I'm a little surprised by your message the other day. Could you please confirm the following points again?

[1056] Email correction

[1057] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses a sentiment analysis engine to generate suggested corrections that take the user's feelings into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[1058] Specific examples

[1059] If the user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1060] Before: I might be late for the meeting tomorrow.

[1061] Corrected: I might be a little late for tomorrow's meeting, but I hope you understand the reason.

[1062] Email Summary

[1063] When a user summarizes a long email, the device sends the email content to the server. The server analyzes the content and extracts key points using a sentiment analysis engine. A summary is generated based on the extracted points and sent to the device for the user to review.

[1064] Specific examples

[1065] If a long email says "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[1066] Summary: There was some positive progress in the recent meeting.

[1067] In this way, the present invention provides a system that allows users to efficiently compose, reply, correct, and summarize emails. In addition, by combining it with emotion analysis technology, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[1068] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1069] Compose a new email

[1070] Step 1:

[1071] The user opens a screen to compose a new email. The user clicks the "Create new email" button on the device. This causes the device to send a request to the server to compose a new email.

[1072] Input: User clicks "Create New Email" button

[1073] Output: A request to the server to create a new email

[1074] Step 2:

[1075] The server generates input prompts for "recipient," "subject," and "body" and sends them to the terminal. The terminal displays an input form to the user.

[1076] Input: Triggering prompt generation on the server side

[1077] Output: Input prompt sent to terminal

[1078] Step 3:

[1079] The user enters the "recipient," "subject," and "body" and presses the send button. The device sends the input data to the server.

[1080] Input: "To", "Subject", and "Body" entered by the user

[1081] Output: Input data sent to the server

[1082] Step 4:

[1083] The server analyzes the received data using NLP and sentiment analysis technologies. Specifically, the NLP model analyzes the text and understands its context and structure, while the sentiment engine determines the user's sentiment.

[1084] Input: Data entered by the user ("To", "Subject", "Body")

[1085] Data processing: Text analysis using NLP models, sentiment analysis using emotion engines

[1086] Output: Analysis results

[1087] Step 5:

[1088] Based on the analysis results, the server generates an appropriate subject line and body text. Specifically, it combines an NLP model and an emotion engine to automatically generate email content that reflects the user's intentions and emotions.

[1089] Input: Analysis results

[1090] Data processing: Email content generation using NLP models and emotion engines

[1091] Output: Auto-generated subject and body

[1092] Step 6:

[1093] The server sends the generated email to the terminal, and the terminal displays the contents of the generated email to the user so that the user can check and edit them.

[1094] Input: Auto-generated subject and body

[1095] Output: Email content sent to the terminal

[1096] Create a reply email

[1097] Step 1:

[1098] When the user presses the reply button for the email they want to reply to, the device sends a request to the server to create a reply email.

[1099] Input: The act of the user pressing the reply button

[1100] Output: Reply mail creation request to the server

[1101] Step 2:

[1102] The server retrieves the user's past email correspondence, retrieves the email to be replied to and the past conversation history, and prepares for analysis.

[1103] Input: The request sent to the server

[1104] Output: Retrieved email history

[1105] Step 3:

[1106] The server uses NLP and emotion analysis technologies to analyze the content and emotional information of past emails. Specifically, it uses NLP to analyze the structure of the text and an emotion engine to determine the importance of the emotion.

[1107] Input: Retrieved past email history

[1108] Data processing: Text and sentiment analysis using NLP models and sentiment engines

[1109] Output: Analysis results

[1110] Step 4:

[1111] The server generates an appropriate reply based on the analysis results, automatically generating a reply email that reflects the user's past emotional state.

[1112] Input: Analysis results

[1113] Data processing: Generate reply content using NLP models and emotion engines

[1114] Output: Auto-generated reply

[1115] Step 5:

[1116] The server sends the generated reply email to the terminal, and the terminal displays the generated reply content to the user so that the user can check and edit it.

[1117] Input: Auto-generated reply

[1118] Output: Reply email sent to the device

[1119] Email correction

[1120] Step 1:

[1121] The user opens an input screen to request corrections to the email content. The user enters the email body they want to correct and presses the send button. The device then sends this to the server.

[1122] Input: The email body entered by the user

[1123] Output: Correction request to the server

[1124] Step 2:

[1125] The server analyzes the email body received to detect grammatical and stylistic errors, using NLP models and a sentiment analysis engine.

[1126] Input: The email body entered by the user

[1127] Data processing: Error detection using NLP models and emotion engines

[1128] Output: Error detection results

[1129] Step 3:

[1130] The server generates suggestions for correcting grammar and style errors, providing sentiment-sensitive suggestions based on the analysis results.

[1131] Input: Error detection result

[1132] Data processing: Generating revision suggestions using NLP models and emotion engines

[1133] Output: Revision proposal

[1134] Step 4:

[1135] The server sends the proposed revisions to the terminal, which displays them to the user so that the user can confirm and apply them.

[1136] Input: Proposed fix

[1137] Output: Suggested fixes sent to the device

[1138] Email Summary

[1139] Step 1:

[1140] The user opens an input screen to request a summary of a long email. The user enters the email text they want to summarize and presses the send button. The terminal then sends this to the server.

[1141] Input: A long email body entered by the user

[1142] Output: Summary request to server

[1143] Step 2:

[1144] The server analyzes the email text to be summarized and extracts key points using NLP technology and a sentiment analysis engine.

[1145] Input: A long email body entered by the user

[1146] Data processing: Text analysis and key point extraction using NLP models and sentiment engines

[1147] Output: Analysis results

[1148] Step 3:

[1149] The server generates a summary based on the analysis results, taking into account emotional information and creating a concise summary.

[1150] Input: Analysis results

[1151] Data processing: Generating summaries using NLP models and emotion engines

[1152] Output: Auto-generated summary

[1153] Step 4:

[1154] The server sends the generated summary to the terminal, which displays the summary to the user for the user to review.

[1155] Input: Auto-generated summary

[1156] Output: Summary sent to terminal

[1157] (Application example 2)

[1158] 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."

[1159] Conventional email management systems provide support for users to efficiently compose, reply, correct, and summarize emails, but they do not support generating emails that take user emotions into account or real-time customer support. As a result, when effective communication with customers is required, especially in physical stores, it is difficult for store staff to respond quickly and appropriately. To solve this problem, a system is needed that includes email management that takes user emotions into account and real-time customer support functions.

[1160] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically generating a subject and body of a new email based on user input, means for automatically generating the body of a reply email based on past email exchanges, means for detecting grammatical and stylistic errors in the email and suggesting corrections, means for extracting important points from the email and generating a summary, means for automatically generating replies to customer inquiries using emotion analysis technology, and means for providing real-time customer service support based on customer emotions via a smart device. This enables users to efficiently manage their emails while providing quick and accurate responses to customers in physical stores.

[1161] A "user" is a person or entity who uses the system to compose, reply to, correct, or summarize emails.

[1162] A "new email" is an email that a user creates and sends for the first time.

[1163] The "subject" is the title part that briefly indicates the content of a new email or a reply email.

[1164] The "body" is the part where the main content of the email is written.

[1165] A "server" is a computer system that processes information entered by users and performs various automatic generation and analysis.

[1166] "Natural language processing" is a technology that analyzes and understands human language, and is primarily used for automatic text generation and sentiment analysis.

[1167] "Emotion analysis technology" is a technology that analyzes a user's emotional state from text or voice and extracts that information.

[1168] "Past email exchanges" refers to the content of emails exchanged between the user and other people in the past.

[1169] A "reply email" is an email sent in response to a received email.

[1170] A "grammatical error" is an error in the language rules contained in the body of an email.

[1171] A "style error" is an error in format or expression contained within the body of an email.

[1172] "Suggested fixes" are suggestions for correcting grammatical or stylistic errors.

[1173] A "summary" is a simplified version of a long email message that extracts the important points.

[1174] A "customer" is a person who visits a physical store to make an inquiry or purchase.

[1175] A "smart device" is an electronic device that has Internet connectivity and can run various applications.

[1176] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. In particular, by combining an emotion engine, it enables email management that takes user emotions into consideration and customer service support in physical stores. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective email and customer service.

[1177] Compose a new email

[1178] When a user composes a new email, they follow the prompts displayed on their device to enter the recipient, subject, and message body. The device then sends this information to the server. The server then uses natural language processing technology and an emotion engine to analyze the entered information and generate an appropriate subject and message body, taking the user's emotions into account. The generated email is then sent to the device, where it can be viewed and edited by the user.

[1179] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[1180] Subject: "Great weekend plans"

[1181] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[1182] Create a reply email

[1183] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1184] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[1185] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[1186] Email correction

[1187] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[1188] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1189] Before: "I might be late for the meeting tomorrow."

[1190] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[1191] Email Summary

[1192] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[1193] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[1194] Summary: "There was some positive progress in the meeting the other day."

[1195] Physical store applications

[1196] In physical stores, store clerks use smart devices (e.g., smart glasses) to assist customers. When a customer sends an inquiry email to a store clerk, the store clerk checks the email via the smart device, and the server performs sentiment analysis to generate an appropriate reply.

[1197] For example, if a customer asks, "Please tell me about new products arriving," the server will generate the following reply if it detects a positive emotion:

[1198] Response: "Thank you for your question about the new product arrivals. We expect new products to arrive this weekend."

[1199] This system allows users to manage emails more efficiently and also enables them to respond to customers more effectively in physical stores. In addition, the emotion engine enables communication that accurately reflects the emotions of users and customers.

[1200] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1201] Step 1:

[1202] The user enters the "recipient," "subject," and "body" into the terminal. The terminal sends this to the server. Input data: the recipient, subject, and body entered by the user. Output data: the input data sent to the server.

[1203] Step 2:

[1204] The server analyzes the received input data using natural language processing and sentiment analysis technologies. Input data: User's address, subject, and text. Output data: Analyzed sentiment information and natural language processing results.

[1205] Step 3:

[1206] Based on the analysis results, the server uses a generative AI model to automatically generate the subject and body of a new email. Input data: Analyzed emotional information and natural language processing results. Output data: Automatically generated subject and body.

[1207] Step 4:

[1208] The generated email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated email sent from the server. Output data: Final email edited by the user.

[1209] Step 5:

[1210] When a user presses the reply button for a specific email, the device sends that information to the server. Input data: Email information from which the reply was made. Output data: Reply information sent to the server.

[1211] Step 6:

[1212] The server acquires past email exchanges and analyzes them using emotion analysis technology. Input data: Past email exchanges. Output data: Analyzed dialogue history and emotion information.

[1213] Step 7:

[1214] Based on the analysis results, the server uses a generative AI model to automatically generate the reply email body. Input data: Analyzed dialogue history and emotional information. Output data: Automatically generated reply body.

[1215] Step 8:

[1216] The generated reply email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated reply email. Output data: Final reply email edited by the user.

[1217] Step 9:

[1218] When a user requests corrections to an email, the terminal sends the email body to the server. Input data: The email body entered by the user. Output data: The email body sent to the server.

[1219] Step 10:

[1220] The server detects grammatical and style errors and generates correction suggestions using sentiment analysis technology. Input data: User's email body. Output data: Detected error information and correction suggestions.

[1221] Step 11:

[1222] The generated correction proposal is sent to the terminal, where the user can check it and apply it as necessary. Input data: The generated correction proposal. Output data: The correction email applied by the user.

[1223] Step 12:

[1224] When a user requests a summary of an email, the terminal sends the email content to the server. Input data: Content of the long email. Output data: Content of the email sent to the server.

[1225] Step 13:

[1226] The server analyzes the email content, extracts key points using sentiment analysis technology, and generates a summary. Input data: Long email content. Output data: Extracted key points and generated summary.

[1227] Step 14:

[1228] The generated summary is sent to the terminal and confirmed by the user. Input data: Generated summary. Output data: Summary confirmed by the user.

[1229] Step 15:

[1230] When a user uses a smart device to deal with a customer in a physical store, the device receives the customer's inquiry email and sends it to the server. Input data: Customer's inquiry email. Output data: Inquiry email sent to the server.

[1231] Step 16:

[1232] The server uses emotion analysis technology to analyze customer emotions and generate appropriate replies. Input data: Customer inquiry email and emotion analysis results. Output data: Automatically generated replies.

[1233] Step 17:

[1234] The generated reply is sent to the smart device, and the store clerk responds appropriately to the customer. Input data: Automatically generated reply. Output data: Response provided by the store clerk to the customer.

[1235] 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.

[1236] 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.

[1237] 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.

[1238] [Third embodiment]

[1239] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1240] 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.

[1241] 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).

[1242] 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.

[1243] 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.

[1244] 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).

[1245] 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.

[1246] 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.

[1247] 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.

[1248] 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.

[1249] 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.

[1250] 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."

[1251] The present invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. The system utilizes natural language processing technology to reduce the burden on users and enable fast and effective email management.

[1252] Compose a new email

[1253] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology to analyze the entered information and automatically generate an appropriate subject and message body. The generated email is then sent to the device, where the user can view and edit it.

[1254] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hi, what are your plans for the weekend?" in the "Message" field, the server will generate the following subject and message, taking into account the appropriate context:

[1255] Subject: "Weekend plans"

[1256] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[1257] Create a reply email

[1258] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of past emails and generates a reply taking into account the conversation history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1259] For example, when a user replies to an email, the server generates the following reply based on the most recent exchange:

[1260] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[1261] Email correction

[1262] When a user requests corrections to an email, the device sends the entered email text to the server, which detects grammatical and stylistic errors and generates suggested corrections. These suggested corrections are then sent to the device, where the user can review and apply them.

[1263] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1264] Before: "I might be late for the meeting tomorrow."

[1265] Corrected: "I might be a little late for the meeting tomorrow."

[1266] Email Summary

[1267] When a user wants to summarize a long email, the device sends the email content to the server, which analyzes the content, extracts the key points, and generates a summary. The summary is then sent to the device so that the user can view it.

[1268] For example, if a long email begins with "Hello, regarding our recent meeting, the following progress has been made...", the server will extract the key elements and generate a summary like this:

[1269] Summary: "I'd like to report on the progress of our recent meeting..."

[1270] Thus, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails, thereby enabling users to manage emails appropriately and accurately with minimal effort.

[1271] The processing flow will be explained below.

[1272] Compose a new email

[1273] Step 1:

[1274] The user presses the new email creation button, and the device sends a new email creation request to the server.

[1275] Step 2:

[1276] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[1277] Step 3:

[1278] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[1279] Step 4:

[1280] When the user completes the input, the terminal transmits the input information to the server.

[1281] Step 5:

[1282] The server uses natural language processing (NLP) technology to analyze the input and generate an appropriate subject line and body of the message.

[1283] Step 6:

[1284] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[1285] Step 7:

[1286] The user reviews the generated email and edits it if necessary.

[1287] Create a reply email

[1288] Step 1:

[1289] The user presses the reply button, and the device sends a reply email creation request to the server.

[1290] Step 2:

[1291] The server retrieves the target email and the most recent exchanges from the user's past email database.

[1292] Step 3:

[1293] The server analyzes the contents of past emails and generates reply content taking into account the dialogue history.

[1294] Step 4:

[1295] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[1296] Step 5:

[1297] The user reviews the generated reply email and edits it if necessary.

[1298] Email correction

[1299] Step 1:

[1300] The user requests correction of an email. The terminal sends the email body to the server.

[1301] Step 2:

[1302] The server receives the email body and begins parsing it for grammatical and stylistic errors.

[1303] Step 3:

[1304] The server generates suggested fixes based on detected errors and improvements.

[1305] Step 4:

[1306] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[1307] Step 5:

[1308] The user checks the correction results and, if necessary, reflects the suggested corrections.

[1309] Email Summary

[1310] Step 1:

[1311] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[1312] Step 2:

[1313] The server receives the email body and begins parsing it to extract key points.

[1314] Step 3:

[1315] The server generates a summary based on the extracted key points.

[1316] Step 4:

[1317] The generated summary is sent to the terminal, which displays the summary to the user.

[1318] Step 5:

[1319] The user checks the summary results and edits them if necessary.

[1320] The above is a specific operation in each processing step.

[1321] Example 1

[1322] 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."

[1323] There are many challenges in current email management. Users must process a large number of emails every day, and tasks such as composing new emails, replying to emails, correcting grammar, and summarizing emails are extremely time-consuming and labor-intensive. Manually creating text, correcting errors, and summarizing emails are particularly burdensome and contribute to reduced work efficiency. Therefore, the present invention aims to automate these tasks, thereby reducing the burden on users and achieving efficient email management.

[1324] 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.

[1325] In this invention, the server includes means for automatically generating the subject and body of a new electronic communication based on user input, means for automatically generating the body of a reply electronic communication based on past electronic communication exchanges, means for detecting grammatical and formatting errors in the electronic communication and suggesting corrections, and means for extracting important points from the electronic communication and generating a summary, thereby enabling users to manage their electronic communications appropriately and accurately with little effort.

[1326] A "User" is a person who uses the System to create, reply to, correct, or summarize electronic communications.

[1327] "Input" is data or information that a user provides to a system.

[1328] "New electronic communications" refers to newly created emails or messages.

[1329] A "reply electronic communication" is a message sent in response to a previously received email or message.

[1330] "Natural language processing" refers to the techniques and methods that allow computers to understand and process human language.

[1331] "Grammar and formatting errors" refers to grammatical errors and improper formatting.

[1332] "Proposed Amendments" means proposed changes to correct grammatical or formatting errors in an Electronic Communication.

[1333] A "summary" is a concise summary of the key points extracted from a lengthy electronic communication.

[1334] A "generative AI model" is an algorithm or technology that uses artificial intelligence to generate text.

[1335] A "prompt sentence" is a guide sentence that prompts the user for input.

[1336] The present invention relates to a system that allows users to efficiently compose, reply, correct, and summarize electronic communications. The system utilizes natural language processing technology to reduce the burden on users and enable them to quickly and effectively manage their electronic communications. The main hardware used includes the user's device (e.g., PC, smartphone) and a server. The software includes a natural language processing model (e.g., GPT-3) and a server-side program that processes communications.

[1337] Compose a new email

[1338] When a user composes a new email, the server displays prompts for the recipient, subject, and message on the user's device. For example, the prompts might include:

[1339] To create a new email, enter the recipient, subject, and message body. Example: To: example@example.com / Subject: Weekend plans / Message: Hello. What are your plans for the weekend?

[1340] The user enters the necessary information according to the instructions, and the device sends the information to the server. The server uses a natural language processing model (e.g., GPT-3) to analyze the entered information and automatically generate an appropriate subject and body. This generated email is then sent to the device, where the user can view and edit it.

[1341] Create a reply email

[1342] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the contents of the past emails and generates a reply taking into account the conversation history. For example, a prompt might include something like:

[1343] To reply to a specific email, click the reply button below and we'll generate an appropriate reply based on your previous interactions.

[1344] The generated reply email is sent to the terminal so that the user can view and edit it.

[1345] Email correction

[1346] When a user requests corrections to an email, the device sends the text of the email they typed to the server, which detects grammatical and stylistic errors and generates suggested corrections. For example, prompts might include:

[1347] I would like to correct the following email text, so please enter it into your device. Example: "I might be late for the meeting tomorrow."

[1348] The generated revision suggestions are sent to the terminal so that the user can review and apply them.

[1349] Email Summary

[1350] When a user summarizes a long email, the device sends the email content to the server, which analyzes it, extracts key points, and generates a summary. For example, prompts include:

[1351] If you have a long email and want to summarize it, enter the entire text and we'll extract the key points and summarize them for you.

[1352] The generated summary is sent to the terminal and made available for review by the user.

[1353] Thus, the present invention provides a system for users to efficiently compose, reply to, correct, and summarize electronic communications, thereby enabling users to manage their electronic communications appropriately and accurately with less effort.

[1354] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1355] Processing steps for creating a new email

[1356] Step 1: User accesses the compose email screen

[1357] The user opens their email client or web interface and accesses the screen for creating a new email.

[1358] Step 2: The server prompts you

[1359] The server displays a prompt on the user's terminal that includes input fields for "To", "Subject", and "Body".

[1360] Input: User's email creation request

[1361] Output: Show input fields

[1362] Step 3: User enters required information

[1363] The user enters the recipient, subject, and body of the message in the input fields provided.

[1364] Input: User input data (recipient, subject, body)

[1365] Output: Required information entered

[1366] Step 4: The device sends the information to the server

[1367] The terminal sends the user's input data to the server as an HTTP POST request.

[1368] Input: User input data (recipient, subject, body)

[1369] Output: HTTP POST request

[1370] Step 5: The server analyzes the input information using natural language processing technology

[1371] The server uses a natural language processing model (e.g., GPT-3) to analyze the user's input data.

[1372] Input: HTTP POST request data

[1373] Output: Parsed data

[1374] Step 6: The server auto-generates the subject and body

[1375] Based on the analysis results, the server automatically generates an appropriate subject and body of the message using a natural language processing model.

[1376] Input: Parsed data

[1377] Output: Auto-generated subject and body

[1378] Step 7: The generated email is sent to the device

[1379] The generated subject and body are sent to the terminal as an HTTP response.

[1380] Input: Auto-generated subject and body

[1381] Output: HTTP response

[1382] Step 8: User reviews and edits email

[1383] The user checks the generated email displayed on the terminal and manually edits it if necessary.

[1384] Input: HTTP response data (generated email)

[1385] Output: Final email reviewed and edited

[1386] Steps for creating a reply email

[1387] Step 1: User presses reply button

[1388] The user clicks the reply button for a particular received email.

[1389] Input: Click on reply button

[1390] Output: Reply request

[1391] Step 2: The server retrieves past emails

[1392] The server retrieves relevant past email conversations from a database.

[1393] Input: Reply Request

[1394] Output: Past email data

[1395] Step 3: The server analyzes the conversation history

[1396] The server analyzes the content of past emails using natural language processing technology.

[1397] Input: Past email data

[1398] Output: Analysis results

[1399] Step 4: Server generates reply content

[1400] The server uses the analysis results to create an appropriate reply using a generative AI model.

[1401] Input: Analysis results

[1402] Output: Reply content

[1403] Step 5: The generated reply email is sent to the device.

[1404] The server sends the generated reply content to the terminal as an HTTP response.

[1405] Input: Reply content

[1406] Output: HTTP response

[1407] Step 6: User reviews and edits reply email

[1408] The user checks the reply email displayed on the terminal and manually edits it if necessary.

[1409] Input: HTTP response data (reply email)

[1410] Output: Final reply email, reviewed and edited

[1411] Email correction process steps

[1412] Step 1: User enters email content

[1413] The user enters the email text they want to correct in the input field.

[1414] Input: Email body

[1415] Output: Input complete

[1416] Step 2: The device sends the email body to the server

[1417] The terminal sends the entered email body to the server as an HTTP POST request.

[1418] Input: Entered email body

[1419] Output: HTTP POST request

[1420] Step 3: The server detects grammar and style errors

[1421] The server uses natural language processing technology to detect grammatical and style errors.

[1422] Input: HTTP POST request data

[1423] Output: Error detection results

[1424] Step 4: Server generates proposed fixes

[1425] The server generates a correction suggestion based on the error detection results.

[1426] Input: Error detection result

[1427] Output: Revision proposal

[1428] Step 5: The proposed fix is ​​sent to the device

[1429] The server sends the proposed revision to the device as an HTTP response.

[1430] Input: Correction Suggestion

[1431] Output: HTTP response

[1432] Step 6: User reviews and implements proposed changes

[1433] The user checks the suggested changes displayed on the device and manually applies them if necessary.

[1434] Input: HTTP response data (correction)

[1435] Output: Final email confirmed and reflected

[1436] Mail Digest Processing Steps

[1437] Step 1: The user enters the email body they want to summarize.

[1438] The user enters the long email they want to summarize into the input field.

[1439] Input: Long email body

[1440] Output: Input complete

[1441] Step 2: The device sends the email content to the server

[1442] The terminal sends the entered long email text to the server as an HTTP POST request.

[1443] Input: The long email body you entered

[1444] Output: HTTP POST request

[1445] Step 3: The server analyzes the email content

[1446] The server uses natural language processing technology to analyze the email content and extract important points.

[1447] Input: HTTP POST request data

[1448] Output: Analysis results

[1449] Step 4: Server generates digest

[1450] The server generates a summary based on the analysis results.

[1451] Input: Analysis results

[1452] Output: Summary

[1453] Step 5: The summary is sent to the device

[1454] The server sends the generated summary to the terminal as an HTTP response.

[1455] Input: Summary

[1456] Output: HTTP response

[1457] Step 6: User confirms summary

[1458] The user checks the summary displayed on the terminal and manually corrects it if necessary.

[1459] Input: HTTP response data (summary)

[1460] Output: Checked and corrected summary

[1461] (Application example 1)

[1462] 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."

[1463] Modern electronic payment services require fast and accurate responses to customer inquiries. However, dealing with a large volume of email inquiries requires a great deal of time and effort from customer support staff, resulting in a high likelihood of response delays and misunderstandings. In these circumstances, a system for efficiently and effectively responding to emails is needed.

[1464] 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.

[1465] In this invention, the server includes means for automatically generating a subject and a body of a new electronic message based on a user's input, means for automatically generating a body of a reply electronic message based on past electronic message exchanges, means for detecting grammatical and stylistic errors in the electronic message and suggesting corrections, means for extracting important points of the electronic message and generating a summary, and means for efficiently responding to customer support inquiries in the electronic payment service using natural language processing technology, thereby enabling customer support personnel to manage electronic messages quickly and accurately with little effort and improving the quality of customer support.

[1466] "User" means any individual or end-user who uses the System to compose, reply to, correct, or summarize electronic messages.

[1467] "Input" refers to information that a user inputs into a terminal when creating or replying to an electronic message, specifically, the addressee, subject, body, etc.

[1468] "New electronic message" refers to an electronic message that a user has newly created and intends to send.

[1469] "Subject" means the title or headline portion of an electronic message that indicates the subject matter or purpose of the message.

[1470] "Body" means the detailed content or main message portion contained within an Electronic Message.

[1471] "Automatic generation" refers to the process by which a system automatically creates the subject line and body of an electronic message based on user input and historical data using natural language processing techniques.

[1472] "Past electronic message exchanges" refers to the history of past electronic messages exchanged between a user and other people or systems.

[1473] A "reply electronic message" refers to an electronic message created by a user in response to a received electronic message.

[1474] "Grammar and style errors" refers to grammatical errors and stylistic inadequacies in an electronic message.

[1475] "Suggested corrections" refer to corrected sentences suggested by the system based on grammatical and stylistic errors.

[1476] "Key Points" refers to information or points of particular importance within an electronic message.

[1477] "Summary" refers to a short summary of the important points extracted from a long electronic message.

[1478] "Natural language processing technology" refers to technology in the field of computer science for analyzing, understanding, and generating human language.

[1479] "Electronic payment service" refers to electronic payment procedures carried out when purchasing goods or using services via the Internet.

[1480] "Customer support" refers to the function of providing technical support and related information provided by a service provider to its customers.

[1481] "Inquiry" means a question or request submitted by a User to Customer Support.

[1482] "Responding efficiently" means that the system reduces the burden on users and allows them to quickly and effectively manage and respond to electronic messages.

[1483] This invention is a system for improving the efficiency of customer support inquiries in electronic payment services. The system automatically generates the subject and body of new electronic messages based on user input, automatically generates reply electronic messages based on past electronic message exchanges, detects and corrects grammatical and stylistic errors, and generates summaries that extract important points.

[1484] System Configuration

[1485] The main components of this system are a terminal, a server, and natural language processing technology including a generative AI model. The terminal is a hardware device used by a customer support representative, such as a smartphone, smart glasses, or a head-mounted display. The server has a generative AI model installed, which performs natural language processing using prompt sentences.

[1486] Explanation of program processing

[1487] The server performs the following steps and processes.

[1488] 1. Automatic generation of new electronic messages

[1489] The server uses a generative AI model to automatically generate an appropriate subject line and message body based on the recipient, subject, and message body entered by the user, for example, using the following prompt:

[1490] Subject: Weekend plans

[1491] Body: Hello. What are your plans for the weekend?

[1492] Generate a suitable subject and body:

[1493] The resulting output is an electronic message like this:

[1494] Subject: "Weekend plans"

[1495] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[1496] 2. Automatic generation of reply electronic messages

[1497] The server retrieves the user's past conversation history and uses a generative AI model to create a reply based on that. An example of a prompt is as follows:

[1498] Generate a reply email based on the following email exchange:

[1499] Hello, regarding the matter from the other day...

[1500] The resulting reply will look like this:

[1501] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[1502] 3. Grammar and style error detection and correction

[1503] The server analyzes the body of the electronic message entered by the user, detects grammatical and stylistic errors, and suggests corrections, for example using the following prompt:

[1504] Please correct the following email:

[1505] I might be late for the meeting tomorrow

[1506] The resulting output is the following proposed fix:

[1507] Before: "I might be late for the meeting tomorrow."

[1508] Corrected: "I might be a little late for the meeting tomorrow."

[1509] 4. Summary Generation

[1510] The server parses long electronic messages and generates summaries of key points. An example prompt is:

[1511] Please summarize the following lengthy email:

[1512] Hello, regarding the meeting we had the other day, we have made the following progress...

[1513] The resulting summary provides:

[1514] Summary: "I'd like to report on the progress of our recent meeting..."

[1515] This allows customer support agents to efficiently manage electronic messages and improve the quality of customer interactions using a server, terminal, and generative AI model.

[1516] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1517] Step 1:

[1518] A user inputs the recipient, subject, and body of a new electronic message into a terminal, and the terminal transmits the input information to a server.

[1519] Step 2:

[1520] The server invokes the generative AI model based on the received input information to automatically generate the subject and body of a new electronic message. The server inputs the following prompt text to the generative AI model:

[1521] Subject: [Subject]

[1522] Body: [Main text]

[1523] Generate a suitable subject and body:

[1524] The generative AI model analyzes the data, generates an appropriate subject line and body text, and sends it back to the server. The generated result is the appropriate subject line and body text.

[1525] Step 3:

[1526] The server sends the subject and body of the generated new electronic message to the terminal, which displays an interface for the user to review and edit. The user can use this interface to review and edit the generated content.

[1527] Step 4:

[1528] When a user wants to reply to a past electronic message, the user presses the reply button on the terminal. The terminal notifies the server of this operation and sends the reply email and past exchanges to the server. The input information includes the content of the past email.

[1529] Step 5:

[1530] The server calls the generative AI model based on the content of past emails it has received, and automatically generates a reply email message. The server inputs the following prompt text into the generative AI model:

[1531] Generate a reply email based on the following email exchange:

[1532] [Past email content]

[1533] The generative AI model performs analysis, generates an appropriate reply, and sends it back to the server. The generated result is the appropriate reply.

[1534] Step 6:

[1535] The server sends the generated reply electronic message to the terminal, and the terminal displays an interface for the user to view and edit the generated reply content. The user can use this interface to view and edit the generated reply content.

[1536] Step 7:

[1537] When a user requests grammar and style correction for an electronic message, the user inputs the target message body into the terminal, which then sends the input information, including the raw message body, to the server.

[1538] Step 8:

[1539] The server invokes a generative AI model based on the received message body, detects grammatical and stylistic errors, and suggests corrections. The server inputs the following prompt to the generative AI model:

[1540] Please correct the following email:

[1541] [Message body]

[1542] The generative AI model performs the analysis, generates suggested revisions, and sends them back to the server. The resulting revised text is obtained.

[1543] Step 9:

[1544] The server sends the generated revision proposal to the terminal, which displays the content of the revision to the user and provides an interface for review and editing. The user can use this interface to review and edit the revision proposal.

[1545] Step 10:

[1546] When a user wants to summarize a long electronic message, the user inputs the long message into the terminal, and the terminal transmits the input information to the server. The input information includes the content of the long message.

[1547] Step 11:

[1548] The server calls the generative AI model based on the received long message, extracts the key points, and generates a summary. The server inputs the following prompt to the generative AI model:

[1549] Please summarize the following lengthy email:

[1550] [Long message]

[1551] The generative AI model performs the analysis, generates a summary, and sends it back to the server. The resulting summary includes key points.

[1552] Step 12:

[1553] The server sends the generated summary to the terminal, which displays the summary to the user and provides an interface for confirmation, allowing the user to check the summary.

[1554] 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.

[1555] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize e-mails, and in particular, by combining an emotion engine, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion recognition technology to reduce the burden on the user and achieve fast and effective e-mail management.

[1556] Compose a new email

[1557] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and an emotion engine to analyze the entered information and automatically generate an appropriate subject and message body, taking into account the user's emotions. The generated email is then sent to the device, where the user can view and edit it.

[1558] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[1559] Subject: "Great weekend plans"

[1560] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[1561] Create a reply email

[1562] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1563] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[1564] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[1565] Email correction

[1566] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[1567] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1568] Before: "I might be late for the meeting tomorrow."

[1569] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[1570] Email Summary

[1571] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[1572] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[1573] Summary: "There was some positive progress in the meeting the other day."

[1574] In this way, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails. In addition, by combining it with an emotion engine, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[1575] The processing flow will be explained below.

[1576] Compose a new email

[1577] Step 1:

[1578] The user presses the button to create a new email. The device sends a request to the server to create a new email.

[1579] Step 2:

[1580] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[1581] Step 3:

[1582] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[1583] Step 4:

[1584] When the user completes the input, the terminal transmits the input information to the server.

[1585] Step 5:

[1586] The server uses an emotion engine to analyze emotions from the user's input information.

[1587] Step 6:

[1588] The server uses natural language processing (NLP) technology to analyze the input content and sentiment information and generate an appropriate subject line and body of the message.

[1589] Step 7:

[1590] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[1591] Step 8:

[1592] The user reviews the generated email and edits it if necessary.

[1593] Create a reply email

[1594] Step 1:

[1595] The user presses the reply button, and the device sends a reply email creation request to the server.

[1596] Step 2:

[1597] The server retrieves the target email and the most recent exchanges from the user's past email database.

[1598] Step 3:

[1599] The server analyzes the content of past emails and the corresponding emotional information.

[1600] Step 4:

[1601] The server uses an emotion engine to generate a reply that takes emotion information into consideration.

[1602] Step 5:

[1603] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[1604] Step 6:

[1605] The user reviews the generated reply email and edits it if necessary.

[1606] Email correction

[1607] Step 1:

[1608] The user requests correction of an email. The terminal sends the email body to the server.

[1609] Step 2:

[1610] The server receives the email body and analyzes the user's emotions using an emotion engine.

[1611] Step 3:

[1612] The server detects grammatical and stylistic errors and generates suggested revisions that take emotional information into account.

[1613] Step 4:

[1614] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[1615] Step 5:

[1616] The user checks the correction results and, if necessary, reflects the suggested corrections.

[1617] Email Summary

[1618] Step 1:

[1619] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[1620] Step 2:

[1621] The server receives the email body and analyzes the user's emotions using an emotion engine.

[1622] Step 3:

[1623] The server analyzes the email content and extracts important points, taking into account emotional information.

[1624] Step 4:

[1625] A summary is generated based on the extracted points and sent to the terminal.

[1626] Step 5:

[1627] The terminal displays the summary to the user, who then checks the summary and edits it as necessary.

[1628] The above is the specific operation of each processing step in the email management system combined with the emotion engine.

[1629] Example 2

[1630] 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."

[1631] In conventional email systems, users often have to spend a lot of time and effort composing, replying, correcting, and summarizing emails. Furthermore, it is difficult to express emotions in email content, which hinders smooth communication.

[1632] 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. In this invention, the server includes means for automatically generating a subject and a body of a new email based on a user's input, means for automatically generating a body of a reply message based on past message exchanges, means for detecting grammatical and stylistic errors in the message and suggesting corrections, means for extracting important points of the message and generating a summary, and means for generating message content taking the user's emotions into consideration using emotion analysis technology. This enables the user to efficiently compose, reply to, correct, and summarize emails, thereby realizing communication that takes the user's emotions into consideration.

[1633] The "means for automatically generating the subject and body of a new email" is a means for automatically creating an appropriate subject and body based on the content entered by the user using natural language processing technology.

[1634] The "means for automatically generating the body of a reply message" is a means for analyzing past message exchanges and automatically generating the body of a reply based on the analysis.

[1635] The "means for detecting grammatical and stylistic errors and suggesting corrections" is a means for detecting grammatical and stylistic errors in the email body and suggesting corrections to the user.

[1636] The "means for extracting important points and generating an outline" is a means for automatically extracting important points from the content of a long email and generating a summary that concisely summarizes the content.

[1637] "Means for generating message content taking into account the user's emotions using emotion analysis technology" refers to means for automatically generating message content that takes into account the user's emotions by analyzing the content entered by the user and using technology to infer the emotions behind it.

[1638] This invention relates to a system that allows users to efficiently compose, reply, correct, and summarize e-mails, and in particular, by combining emotion analysis technology, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective e-mail management.

[1639] First, we generate a program for this system. This program has the following functions:

[1640] 1. When a user composes a new email, the server automatically generates an appropriate subject and body based on the user's input.

[1641] 2. When creating a reply email, the server automatically generates the body of the reply email based on past message exchanges.

[1642] 3. When editing emails, the server detects grammatical and stylistic errors and suggests corrections.

[1643] 4. When summarizing an email, the server extracts the important points and generates a summary.

[1644] 5. At each processing step, emotion analysis technology is used to consider the user's emotions.

[1645] Compose a new email

[1646] When a user creates a new email, the server prompts the user for the recipient, subject, and body of the email. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and a sentiment analysis engine (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to analyze the entered information and automatically generate an appropriate subject and body of the email, taking into account the user's sentiment. The generated email is then sent to the device, where the user can view and edit it.

[1647] Specific examples

[1648] Consider the case where the user enters the following:

[1649] To: example@example.com

[1650] Subject: Weekend plans

[1651] Body: Hello. What are your plans for the weekend?

[1652] The server receives this and automatically generates it as follows:

[1653] Subject: Great weekend plans

[1654] Body: Hello, I'd like to reminisce about our fun conversation the other day and confirm your plans for the weekend. I'm looking forward to it.

[1655] Create a reply email

[1656] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server uses the content of past emails and an emotion analysis engine to generate a reply, taking into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1657] Specific examples

[1658] When a user replies to an email, the server generates the following reply based on recent interactions and emotional information:

[1659] Response: Hello, I'm a little surprised by your message the other day. Could you please confirm the following points again?

[1660] Email correction

[1661] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses a sentiment analysis engine to generate suggested corrections that take the user's feelings into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[1662] Specific examples

[1663] If the user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1664] Before: I might be late for the meeting tomorrow.

[1665] Corrected: I might be a little late for tomorrow's meeting, but I hope you understand the reason.

[1666] Email Summary

[1667] When a user summarizes a long email, the device sends the email content to the server. The server analyzes the content and extracts key points using a sentiment analysis engine. A summary is generated based on the extracted points and sent to the device for the user to review.

[1668] Specific examples

[1669] If a long email says "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[1670] Summary: There was some positive progress in the recent meeting.

[1671] In this way, the present invention provides a system that allows users to efficiently compose, reply, correct, and summarize emails. In addition, by combining it with emotion analysis technology, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[1672] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1673] Compose a new email

[1674] Step 1:

[1675] The user opens a screen to compose a new email. The user clicks the "Create new email" button on the device. This causes the device to send a request to the server to compose a new email.

[1676] Input: User clicks "Create New Email" button

[1677] Output: A request to the server to create a new email

[1678] Step 2:

[1679] The server generates input prompts for "recipient," "subject," and "body" and sends them to the terminal. The terminal displays an input form to the user.

[1680] Input: Triggering prompt generation on the server side

[1681] Output: Input prompt sent to terminal

[1682] Step 3:

[1683] The user enters the "recipient," "subject," and "body" and presses the send button. The device sends the input data to the server.

[1684] Input: "To", "Subject", and "Body" entered by the user

[1685] Output: Input data sent to the server

[1686] Step 4:

[1687] The server analyzes the received data using NLP and sentiment analysis technologies. Specifically, the NLP model analyzes the text and understands its context and structure, while the sentiment engine determines the user's sentiment.

[1688] Input: Data entered by the user ("To", "Subject", "Body")

[1689] Data processing: Text analysis using NLP models, sentiment analysis using emotion engines

[1690] Output: Analysis results

[1691] Step 5:

[1692] Based on the analysis results, the server generates an appropriate subject line and body text. Specifically, it combines an NLP model and an emotion engine to automatically generate email content that reflects the user's intentions and emotions.

[1693] Input: Analysis results

[1694] Data processing: Email content generation using NLP models and emotion engines

[1695] Output: Auto-generated subject and body

[1696] Step 6:

[1697] The server sends the generated email to the terminal, and the terminal displays the contents of the generated email to the user so that the user can check and edit them.

[1698] Input: Auto-generated subject and body

[1699] Output: Email content sent to the terminal

[1700] Create a reply email

[1701] Step 1:

[1702] When the user presses the reply button for the email they want to reply to, the device sends a request to the server to create a reply email.

[1703] Input: The act of the user pressing the reply button

[1704] Output: Reply mail creation request to the server

[1705] Step 2:

[1706] The server retrieves the user's past email correspondence, retrieves the email to be replied to and the past conversation history, and prepares for analysis.

[1707] Input: The request sent to the server

[1708] Output: Retrieved email history

[1709] Step 3:

[1710] The server uses NLP and emotion analysis technologies to analyze the content and emotional information of past emails. Specifically, it uses NLP to analyze the structure of the text and an emotion engine to determine the importance of the emotion.

[1711] Input: Retrieved past email history

[1712] Data processing: Text and sentiment analysis using NLP models and sentiment engines

[1713] Output: Analysis results

[1714] Step 4:

[1715] The server generates an appropriate reply based on the analysis results, automatically generating a reply email that reflects the user's past emotional state.

[1716] Input: Analysis results

[1717] Data processing: Generate reply content using NLP models and emotion engines

[1718] Output: Auto-generated reply

[1719] Step 5:

[1720] The server sends the generated reply email to the terminal, and the terminal displays the generated reply content to the user so that the user can check and edit it.

[1721] Input: Auto-generated reply

[1722] Output: Reply email sent to the device

[1723] Email correction

[1724] Step 1:

[1725] The user opens an input screen to request corrections to the email content. The user enters the email body they want to correct and presses the send button. The device then sends this to the server.

[1726] Input: The email body entered by the user

[1727] Output: Correction request to the server

[1728] Step 2:

[1729] The server analyzes the email body received to detect grammatical and stylistic errors, using NLP models and a sentiment analysis engine.

[1730] Input: The email body entered by the user

[1731] Data processing: Error detection using NLP models and emotion engines

[1732] Output: Error detection results

[1733] Step 3:

[1734] The server generates suggestions for correcting grammar and style errors, providing sentiment-sensitive suggestions based on the analysis results.

[1735] Input: Error detection result

[1736] Data processing: Generating revision suggestions using NLP models and emotion engines

[1737] Output: Revision proposal

[1738] Step 4:

[1739] The server sends the proposed revisions to the terminal, which displays them to the user so that the user can confirm and apply them.

[1740] Input: Proposed fix

[1741] Output: Suggested fixes sent to the device

[1742] Email Summary

[1743] Step 1:

[1744] The user opens an input screen to request a summary of a long email. The user enters the email text they want to summarize and presses the send button. The terminal then sends this to the server.

[1745] Input: A long email body entered by the user

[1746] Output: Summary request to server

[1747] Step 2:

[1748] The server analyzes the email text to be summarized and extracts key points using NLP technology and a sentiment analysis engine.

[1749] Input: A long email body entered by the user

[1750] Data processing: Text analysis and key point extraction using NLP models and sentiment engines

[1751] Output: Analysis results

[1752] Step 3:

[1753] The server generates a summary based on the analysis results, taking into account emotional information and creating a concise summary.

[1754] Input: Analysis results

[1755] Data processing: Generating summaries using NLP models and emotion engines

[1756] Output: Auto-generated summary

[1757] Step 4:

[1758] The server sends the generated summary to the terminal, which displays the summary to the user for the user to review.

[1759] Input: Auto-generated summary

[1760] Output: Summary sent to terminal

[1761] (Application example 2)

[1762] 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."

[1763] Conventional email management systems provide support for users to efficiently compose, reply, correct, and summarize emails, but they do not support generating emails that take user emotions into account or real-time customer support. As a result, when effective communication with customers is required, especially in physical stores, it is difficult for store staff to respond quickly and appropriately. To solve this problem, a system is needed that includes email management that takes user emotions into account and real-time customer support functions.

[1764] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically generating a subject and body of a new email based on user input, means for automatically generating the body of a reply email based on past email exchanges, means for detecting grammatical and stylistic errors in the email and suggesting corrections, means for extracting important points from the email and generating a summary, means for automatically generating replies to customer inquiries using emotion analysis technology, and means for providing real-time customer service support based on customer emotions via a smart device. This enables users to efficiently manage their emails while providing quick and accurate responses to customers in physical stores.

[1765] A "user" is a person or entity who uses the system to compose, reply to, correct, or summarize emails.

[1766] A "new email" is an email that a user creates and sends for the first time.

[1767] The "subject" is the title part that briefly indicates the content of a new email or a reply email.

[1768] The "body" is the part where the main content of the email is written.

[1769] A "server" is a computer system that processes information entered by users and performs various automatic generation and analysis.

[1770] "Natural language processing" is a technology that analyzes and understands human language, and is primarily used for automatic text generation and sentiment analysis.

[1771] "Emotion analysis technology" is a technology that analyzes a user's emotional state from text or voice and extracts that information.

[1772] "Past email exchanges" refers to the content of emails exchanged between the user and other people in the past.

[1773] A "reply email" is an email sent in response to a received email.

[1774] A "grammatical error" is an error in the language rules contained in the body of an email.

[1775] A "style error" is an error in format or expression contained within the body of an email.

[1776] "Suggested fixes" are suggestions for correcting grammatical or stylistic errors.

[1777] A "summary" is a simplified version of a long email message that extracts the important points.

[1778] A "customer" is a person who visits a physical store to make an inquiry or purchase.

[1779] A "smart device" is an electronic device that has Internet connectivity and can run various applications.

[1780] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. In particular, by combining an emotion engine, it enables email management that takes user emotions into consideration and customer service support in physical stores. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective email and customer service.

[1781] Compose a new email

[1782] When a user composes a new email, they follow the prompts displayed on their device to enter the recipient, subject, and message body. The device then sends this information to the server. The server then uses natural language processing technology and an emotion engine to analyze the entered information and generate an appropriate subject and message body, taking the user's emotions into account. The generated email is then sent to the device, where it can be viewed and edited by the user.

[1783] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[1784] Subject: "Great weekend plans"

[1785] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[1786] Create a reply email

[1787] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1788] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[1789] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[1790] Email correction

[1791] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[1792] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1793] Before: "I might be late for the meeting tomorrow."

[1794] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[1795] Email Summary

[1796] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[1797] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[1798] Summary: "There was some positive progress in the meeting the other day."

[1799] Physical store applications

[1800] In physical stores, store clerks use smart devices (e.g., smart glasses) to assist customers. When a customer sends an inquiry email to a store clerk, the store clerk checks the email via the smart device, and the server performs sentiment analysis to generate an appropriate reply.

[1801] For example, if a customer asks, "Please tell me about new products arriving," the server will generate the following reply if it detects a positive emotion:

[1802] Response: "Thank you for your question about the new product arrivals. We expect new products to arrive this weekend."

[1803] This system allows users to manage emails more efficiently and also enables them to respond to customers more effectively in physical stores. In addition, the emotion engine enables communication that accurately reflects the emotions of users and customers.

[1804] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1805] Step 1:

[1806] The user enters the "recipient," "subject," and "body" into the terminal. The terminal sends this to the server. Input data: the recipient, subject, and body entered by the user. Output data: the input data sent to the server.

[1807] Step 2:

[1808] The server analyzes the received input data using natural language processing and sentiment analysis technologies. Input data: User's address, subject, and text. Output data: Analyzed sentiment information and natural language processing results.

[1809] Step 3:

[1810] Based on the analysis results, the server uses a generative AI model to automatically generate the subject and body of a new email. Input data: Analyzed emotional information and natural language processing results. Output data: Automatically generated subject and body.

[1811] Step 4:

[1812] The generated email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated email sent from the server. Output data: Final email edited by the user.

[1813] Step 5:

[1814] When a user presses the reply button for a specific email, the device sends that information to the server. Input data: Email information from which the reply was made. Output data: Reply information sent to the server.

[1815] Step 6:

[1816] The server acquires past email exchanges and analyzes them using emotion analysis technology. Input data: Past email exchanges. Output data: Analyzed dialogue history and emotion information.

[1817] Step 7:

[1818] Based on the analysis results, the server uses a generative AI model to automatically generate the reply email body. Input data: Analyzed dialogue history and emotional information. Output data: Automatically generated reply body.

[1819] Step 8:

[1820] The generated reply email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated reply email. Output data: Final reply email edited by the user.

[1821] Step 9:

[1822] When a user requests corrections to an email, the terminal sends the email body to the server. Input data: The email body entered by the user. Output data: The email body sent to the server.

[1823] Step 10:

[1824] The server detects grammatical and style errors and generates correction suggestions using sentiment analysis technology. Input data: User's email body. Output data: Detected error information and correction suggestions.

[1825] Step 11:

[1826] The generated correction proposal is sent to the terminal, where the user can check it and apply it as necessary. Input data: The generated correction proposal. Output data: The correction email applied by the user.

[1827] Step 12:

[1828] When a user requests a summary of an email, the terminal sends the email content to the server. Input data: Content of the long email. Output data: Content of the email sent to the server.

[1829] Step 13:

[1830] The server analyzes the email content, extracts key points using sentiment analysis technology, and generates a summary. Input data: Long email content. Output data: Extracted key points and generated summary.

[1831] Step 14:

[1832] The generated summary is sent to the terminal and confirmed by the user. Input data: Generated summary. Output data: Summary confirmed by the user.

[1833] Step 15:

[1834] When a user uses a smart device to deal with a customer in a physical store, the device receives the customer's inquiry email and sends it to the server. Input data: Customer's inquiry email. Output data: Inquiry email sent to the server.

[1835] Step 16:

[1836] The server uses emotion analysis technology to analyze customer emotions and generate appropriate replies. Input data: Customer inquiry email and emotion analysis results. Output data: Automatically generated replies.

[1837] Step 17:

[1838] The generated reply is sent to the smart device, and the store clerk responds appropriately to the customer. Input data: Automatically generated reply. Output data: Response provided by the store clerk to the customer.

[1839] 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.

[1840] 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.

[1841] 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.

[1842] [Fourth embodiment]

[1843] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1844] 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.

[1845] 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).

[1846] 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.

[1847] 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.

[1848] 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).

[1849] 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.

[1850] 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.

[1851] 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.

[1852] 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.

[1853] 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.

[1854] 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.

[1855] 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."

[1856] The present invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. The system utilizes natural language processing technology to reduce the burden on users and enable fast and effective email management.

[1857] Compose a new email

[1858] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology to analyze the entered information and automatically generate an appropriate subject and message body. The generated email is then sent to the device, where the user can view and edit it.

[1859] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hi, what are your plans for the weekend?" in the "Message" field, the server will generate the following subject and message, taking into account the appropriate context:

[1860] Subject: "Weekend plans"

[1861] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[1862] Create a reply email

[1863] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of past emails and generates a reply taking into account the conversation history. The generated reply email is sent to the terminal, where the user can view and edit it.

[1864] For example, when a user replies to an email, the server generates the following reply based on the most recent exchange:

[1865] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[1866] Email correction

[1867] When a user requests corrections to an email, the device sends the entered email text to the server, which detects grammatical and stylistic errors and generates suggested corrections. These suggested corrections are then sent to the device, where the user can review and apply them.

[1868] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[1869] Before: "I might be late for the meeting tomorrow."

[1870] Corrected: "I might be a little late for the meeting tomorrow."

[1871] Email Summary

[1872] When a user wants to summarize a long email, the device sends the email content to the server, which analyzes the content, extracts the key points, and generates a summary. The summary is then sent to the device so that the user can view it.

[1873] For example, if a long email begins with "Hello, regarding our recent meeting, the following progress has been made...", the server will extract the key elements and generate a summary like this:

[1874] Summary: "I'd like to report on the progress of our recent meeting..."

[1875] Thus, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails, thereby enabling users to manage emails appropriately and accurately with minimal effort.

[1876] The processing flow will be explained below.

[1877] Compose a new email

[1878] Step 1:

[1879] The user presses the new email creation button, and the device sends a new email creation request to the server.

[1880] Step 2:

[1881] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[1882] Step 3:

[1883] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[1884] Step 4:

[1885] When the user completes the input, the terminal transmits the input information to the server.

[1886] Step 5:

[1887] The server uses natural language processing (NLP) technology to analyze the input and generate an appropriate subject line and body of the message.

[1888] Step 6:

[1889] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[1890] Step 7:

[1891] The user reviews the generated email and edits it if necessary.

[1892] Create a reply email

[1893] Step 1:

[1894] The user presses the reply button, and the device sends a reply email creation request to the server.

[1895] Step 2:

[1896] The server retrieves the target email and the most recent exchanges from the user's past email database.

[1897] Step 3:

[1898] The server analyzes the contents of past emails and generates reply content taking into account the dialogue history.

[1899] Step 4:

[1900] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[1901] Step 5:

[1902] The user reviews the generated reply email and edits it if necessary.

[1903] Email correction

[1904] Step 1:

[1905] The user requests correction of an email. The terminal sends the email body to the server.

[1906] Step 2:

[1907] The server receives the email body and begins parsing it for grammatical and stylistic errors.

[1908] Step 3:

[1909] The server generates suggested fixes based on detected errors and improvements.

[1910] Step 4:

[1911] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[1912] Step 5:

[1913] The user checks the correction results and, if necessary, reflects the suggested corrections.

[1914] Email Summary

[1915] Step 1:

[1916] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[1917] Step 2:

[1918] The server receives the email body and begins parsing it to extract key points.

[1919] Step 3:

[1920] The server generates a summary based on the extracted key points.

[1921] Step 4:

[1922] The generated summary is sent to the terminal, which displays the summary to the user.

[1923] Step 5:

[1924] The user checks the summary results and edits them if necessary.

[1925] The above is a specific operation in each processing step.

[1926] Example 1

[1927] 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."

[1928] There are many challenges in current email management. Users must process a large number of emails every day, and tasks such as composing new emails, replying to emails, correcting grammar, and summarizing emails are extremely time-consuming and labor-intensive. Manually creating text, correcting errors, and summarizing emails are particularly burdensome and contribute to reduced work efficiency. Therefore, the present invention aims to automate these tasks, thereby reducing the burden on users and achieving efficient email management.

[1929] 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.

[1930] In this invention, the server includes means for automatically generating the subject and body of a new electronic communication based on user input, means for automatically generating the body of a reply electronic communication based on past electronic communication exchanges, means for detecting grammatical and formatting errors in the electronic communication and suggesting corrections, and means for extracting important points from the electronic communication and generating a summary, thereby enabling users to manage their electronic communications appropriately and accurately with little effort.

[1931] A "User" is a person who uses the System to create, reply to, correct, or summarize electronic communications.

[1932] "Input" is data or information that a user provides to a system.

[1933] "New electronic communications" refers to newly created emails or messages.

[1934] A "reply electronic communication" is a message sent in response to a previously received email or message.

[1935] "Natural language processing" refers to the techniques and methods that allow computers to understand and process human language.

[1936] "Grammar and formatting errors" refers to grammatical errors and improper formatting.

[1937] "Proposed Amendments" means proposed changes to correct grammatical or formatting errors in an Electronic Communication.

[1938] A "summary" is a concise summary of the key points extracted from a lengthy electronic communication.

[1939] A "generative AI model" is an algorithm or technology that uses artificial intelligence to generate text.

[1940] A "prompt sentence" is a guide sentence that prompts the user for input.

[1941] The present invention relates to a system that allows users to efficiently compose, reply, correct, and summarize electronic communications. The system utilizes natural language processing technology to reduce the burden on users and enable them to quickly and effectively manage their electronic communications. The main hardware used includes the user's device (e.g., PC, smartphone) and a server. The software includes a natural language processing model (e.g., GPT-3) and a server-side program that processes communications.

[1942] Compose a new email

[1943] When a user composes a new email, the server displays prompts for the recipient, subject, and message on the user's device. For example, the prompts might include:

[1944] To create a new email, enter the recipient, subject, and message body. Example: To: example@example.com / Subject: Weekend plans / Message: Hello. What are your plans for the weekend?

[1945] The user enters the necessary information according to the instructions, and the device sends the information to the server. The server uses a natural language processing model (e.g., GPT-3) to analyze the entered information and automatically generate an appropriate subject and body. This generated email is then sent to the device, where the user can view and edit it.

[1946] Create a reply email

[1947] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the contents of the past emails and generates a reply taking into account the conversation history. For example, a prompt might include something like:

[1948] To reply to a specific email, click the reply button below and we'll generate an appropriate reply based on your previous interactions.

[1949] The generated reply email is sent to the terminal so that the user can view and edit it.

[1950] Email correction

[1951] When a user requests corrections to an email, the device sends the text of the email they typed to the server, which detects grammatical and stylistic errors and generates suggested corrections. For example, prompts might include:

[1952] I would like to correct the following email text, so please enter it into your device. Example: "I might be late for the meeting tomorrow."

[1953] The generated revision suggestions are sent to the terminal so that the user can review and apply them.

[1954] Email Summary

[1955] When a user summarizes a long email, the device sends the email content to the server, which analyzes it, extracts key points, and generates a summary. For example, prompts include:

[1956] If you have a long email and want to summarize it, enter the entire text and we'll extract the key points and summarize them for you.

[1957] The generated summary is sent to the terminal and made available for review by the user.

[1958] Thus, the present invention provides a system for users to efficiently compose, reply to, correct, and summarize electronic communications, thereby enabling users to manage their electronic communications appropriately and accurately with less effort.

[1959] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1960] Processing steps for creating a new email

[1961] Step 1: User accesses the compose email screen

[1962] The user opens their email client or web interface and accesses the screen for creating a new email.

[1963] Step 2: The server prompts you

[1964] The server displays a prompt on the user's terminal that includes input fields for "To", "Subject", and "Body".

[1965] Input: User's email creation request

[1966] Output: Show input fields

[1967] Step 3: User enters required information

[1968] The user enters the recipient, subject, and body of the message in the input fields provided.

[1969] Input: User input data (recipient, subject, body)

[1970] Output: Required information entered

[1971] Step 4: The device sends the information to the server

[1972] The terminal sends the user's input data to the server as an HTTP POST request.

[1973] Input: User input data (recipient, subject, body)

[1974] Output: HTTP POST request

[1975] Step 5: The server analyzes the input information using natural language processing technology

[1976] The server uses a natural language processing model (e.g., GPT-3) to analyze the user's input data.

[1977] Input: HTTP POST request data

[1978] Output: Parsed data

[1979] Step 6: The server auto-generates the subject and body

[1980] Based on the analysis results, the server automatically generates an appropriate subject and body of the message using a natural language processing model.

[1981] Input: Parsed data

[1982] Output: Auto-generated subject and body

[1983] Step 7: The generated email is sent to the device

[1984] The generated subject and body are sent to the terminal as an HTTP response.

[1985] Input: Auto-generated subject and body

[1986] Output: HTTP response

[1987] Step 8: User reviews and edits email

[1988] The user checks the generated email displayed on the terminal and manually edits it if necessary.

[1989] Input: HTTP response data (generated email)

[1990] Output: Final email reviewed and edited

[1991] Steps for creating a reply email

[1992] Step 1: User presses reply button

[1993] The user clicks the reply button for a particular received email.

[1994] Input: Click on reply button

[1995] Output: Reply request

[1996] Step 2: The server retrieves past emails

[1997] The server retrieves relevant past email conversations from a database.

[1998] Input: Reply Request

[1999] Output: Past email data

[2000] Step 3: The server analyzes the conversation history

[2001] The server analyzes the content of past emails using natural language processing technology.

[2002] Input: Past email data

[2003] Output: Analysis results

[2004] Step 4: Server generates reply content

[2005] The server uses the analysis results to create an appropriate reply using a generative AI model.

[2006] Input: Analysis results

[2007] Output: Reply content

[2008] Step 5: The generated reply email is sent to the device.

[2009] The server sends the generated reply content to the terminal as an HTTP response.

[2010] Input: Reply content

[2011] Output: HTTP response

[2012] Step 6: User reviews and edits reply email

[2013] The user checks the reply email displayed on the terminal and manually edits it if necessary.

[2014] Input: HTTP response data (reply email)

[2015] Output: Final reply email, reviewed and edited

[2016] Email correction process steps

[2017] Step 1: User enters email content

[2018] The user enters the email text they want to correct in the input field.

[2019] Input: Email body

[2020] Output: Input complete

[2021] Step 2: The device sends the email body to the server

[2022] The terminal sends the entered email body to the server as an HTTP POST request.

[2023] Input: Entered email body

[2024] Output: HTTP POST request

[2025] Step 3: The server detects grammar and style errors

[2026] The server uses natural language processing technology to detect grammatical and style errors.

[2027] Input: HTTP POST request data

[2028] Output: Error detection results

[2029] Step 4: Server generates proposed fixes

[2030] The server generates a correction suggestion based on the error detection results.

[2031] Input: Error detection result

[2032] Output: Revision proposal

[2033] Step 5: The proposed fix is ​​sent to the device

[2034] The server sends the proposed revision to the device as an HTTP response.

[2035] Input: Correction Suggestion

[2036] Output: HTTP response

[2037] Step 6: User reviews and implements proposed changes

[2038] The user checks the suggested changes displayed on the device and manually applies them if necessary.

[2039] Input: HTTP response data (correction)

[2040] Output: Final email confirmed and reflected

[2041] Mail Digest Processing Steps

[2042] Step 1: The user enters the email body they want to summarize.

[2043] The user enters the long email they want to summarize into the input field.

[2044] Input: Long email body

[2045] Output: Input complete

[2046] Step 2: The device sends the email content to the server

[2047] The terminal sends the entered long email text to the server as an HTTP POST request.

[2048] Input: The long email body you entered

[2049] Output: HTTP POST request

[2050] Step 3: The server analyzes the email content

[2051] The server uses natural language processing technology to analyze the email content and extract important points.

[2052] Input: HTTP POST request data

[2053] Output: Analysis results

[2054] Step 4: Server generates digest

[2055] The server generates a summary based on the analysis results.

[2056] Input: Analysis results

[2057] Output: Summary

[2058] Step 5: The summary is sent to the device

[2059] The server sends the generated summary to the terminal as an HTTP response.

[2060] Input: Summary

[2061] Output: HTTP response

[2062] Step 6: User confirms summary

[2063] The user checks the summary displayed on the terminal and manually corrects it if necessary.

[2064] Input: HTTP response data (summary)

[2065] Output: Checked and corrected summary

[2066] (Application example 1)

[2067] 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."

[2068] Modern electronic payment services require fast and accurate responses to customer inquiries. However, dealing with a large volume of email inquiries requires a great deal of time and effort from customer support staff, resulting in a high likelihood of response delays and misunderstandings. In these circumstances, a system for efficiently and effectively responding to emails is needed.

[2069] 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.

[2070] In this invention, the server includes means for automatically generating a subject and a body of a new electronic message based on a user's input, means for automatically generating a body of a reply electronic message based on past electronic message exchanges, means for detecting grammatical and stylistic errors in the electronic message and suggesting corrections, means for extracting important points of the electronic message and generating a summary, and means for efficiently responding to customer support inquiries in the electronic payment service using natural language processing technology, thereby enabling customer support personnel to manage electronic messages quickly and accurately with little effort and improving the quality of customer support.

[2071] "User" means any individual or end-user who uses the System to compose, reply to, correct, or summarize electronic messages.

[2072] "Input" refers to information that a user inputs into a terminal when creating or replying to an electronic message, specifically, the addressee, subject, body, etc.

[2073] "New electronic message" refers to an electronic message that a user has newly created and intends to send.

[2074] "Subject" means the title or headline portion of an electronic message that indicates the subject matter or purpose of the message.

[2075] "Body" means the detailed content or main message portion contained within an Electronic Message.

[2076] "Automatic generation" refers to the process by which a system automatically creates the subject line and body of an electronic message based on user input and historical data using natural language processing techniques.

[2077] "Past electronic message exchanges" refers to the history of past electronic messages exchanged between a user and other people or systems.

[2078] A "reply electronic message" refers to an electronic message created by a user in response to a received electronic message.

[2079] "Grammar and style errors" refers to grammatical errors and stylistic inadequacies in an electronic message.

[2080] "Suggested corrections" refer to corrected sentences suggested by the system based on grammatical and stylistic errors.

[2081] "Key Points" refers to information or points of particular importance within an electronic message.

[2082] "Summary" refers to a short summary of the important points extracted from a long electronic message.

[2083] "Natural language processing technology" refers to technology in the field of computer science for analyzing, understanding, and generating human language.

[2084] "Electronic payment service" refers to electronic payment procedures carried out when purchasing goods or using services via the Internet.

[2085] "Customer support" refers to the function of providing technical support and related information provided by a service provider to its customers.

[2086] "Inquiry" means a question or request submitted by a User to Customer Support.

[2087] "Responding efficiently" means that the system reduces the burden on users and allows them to quickly and effectively manage and respond to electronic messages.

[2088] This invention is a system for improving the efficiency of customer support inquiries in electronic payment services. The system automatically generates the subject and body of new electronic messages based on user input, automatically generates reply electronic messages based on past electronic message exchanges, detects and corrects grammatical and stylistic errors, and generates summaries that extract important points.

[2089] System Configuration

[2090] The main components of this system are a terminal, a server, and natural language processing technology including a generative AI model. The terminal is a hardware device used by a customer support representative, such as a smartphone, smart glasses, or a head-mounted display. The server has a generative AI model installed, which performs natural language processing using prompt sentences.

[2091] Explanation of program processing

[2092] The server performs the following steps and processes.

[2093] 1. Automatic generation of new electronic messages

[2094] The server uses a generative AI model to automatically generate an appropriate subject line and message body based on the recipient, subject, and message body entered by the user, for example, using the following prompt:

[2095] Subject: Weekend plans

[2096] Body: Hello. What are your plans for the weekend?

[2097] Generate a suitable subject and body:

[2098] The resulting output is an electronic message like this:

[2099] Subject: "Weekend plans"

[2100] Body of text: "Hello, based on our conversation the other day, I'd like to confirm your plans for the weekend. Thanks in advance."

[2101] 2. Automatic generation of reply electronic messages

[2102] The server retrieves the user's past conversation history and uses a generative AI model to create a reply based on that. An example of a prompt is as follows:

[2103] Generate a reply email based on the following email exchange:

[2104] Hello, regarding the matter from the other day...

[2105] The resulting reply will look like this:

[2106] Response: "Hello, regarding your email from the other day, I would like to confirm the following points..."

[2107] 3. Grammar and style error detection and correction

[2108] The server analyzes the body of the electronic message entered by the user, detects grammatical and stylistic errors, and suggests corrections, for example using the following prompt:

[2109] Please correct the following email:

[2110] I might be late for the meeting tomorrow

[2111] The resulting output is the following proposed fix:

[2112] Before: "I might be late for the meeting tomorrow."

[2113] Corrected: "I might be a little late for the meeting tomorrow."

[2114] 4. Summary Generation

[2115] The server parses long electronic messages and generates summaries of key points. An example prompt is:

[2116] Please summarize the following lengthy email:

[2117] Hello, regarding the meeting we had the other day, we have made the following progress...

[2118] The resulting summary provides:

[2119] Summary: "I'd like to report on the progress of our recent meeting..."

[2120] This allows customer support agents to efficiently manage electronic messages and improve the quality of customer interactions using a server, terminal, and generative AI model.

[2121] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2122] Step 1:

[2123] A user inputs the recipient, subject, and body of a new electronic message into a terminal, and the terminal transmits the input information to a server.

[2124] Step 2:

[2125] The server invokes the generative AI model based on the received input information to automatically generate the subject and body of a new electronic message. The server inputs the following prompt text to the generative AI model:

[2126] Subject: [Subject]

[2127] Body: [Main text]

[2128] Generate a suitable subject and body:

[2129] The generative AI model analyzes the data, generates an appropriate subject line and body text, and sends it back to the server. The generated result is the appropriate subject line and body text.

[2130] Step 3:

[2131] The server sends the subject and body of the generated new electronic message to the terminal, which displays an interface for the user to review and edit. The user can use this interface to review and edit the generated content.

[2132] Step 4:

[2133] When a user wants to reply to a past electronic message, the user presses the reply button on the terminal. The terminal notifies the server of this operation and sends the reply email and past exchanges to the server. The input information includes the content of the past email.

[2134] Step 5:

[2135] The server calls the generative AI model based on the content of past emails it has received, and automatically generates a reply email message. The server inputs the following prompt text into the generative AI model:

[2136] Generate a reply email based on the following email exchange:

[2137] [Past email content]

[2138] The generative AI model performs analysis, generates an appropriate reply, and sends it back to the server. The generated result is the appropriate reply.

[2139] Step 6:

[2140] The server sends the generated reply electronic message to the terminal, and the terminal displays an interface for the user to view and edit the generated reply content. The user can use this interface to view and edit the generated reply content.

[2141] Step 7:

[2142] When a user requests grammar and style correction for an electronic message, the user inputs the target message body into the terminal, which then sends the input information, including the raw message body, to the server.

[2143] Step 8:

[2144] The server invokes a generative AI model based on the received message body, detects grammatical and stylistic errors, and suggests corrections. The server inputs the following prompt to the generative AI model:

[2145] Please correct the following email:

[2146] [Message body]

[2147] The generative AI model performs the analysis, generates suggested revisions, and sends them back to the server. The resulting revised text is obtained.

[2148] Step 9:

[2149] The server sends the generated revision proposal to the terminal, which displays the content of the revision to the user and provides an interface for review and editing. The user can use this interface to review and edit the revision proposal.

[2150] Step 10:

[2151] When a user wants to summarize a long electronic message, the user inputs the long message into the terminal, and the terminal transmits the input information to the server. The input information includes the content of the long message.

[2152] Step 11:

[2153] The server calls the generative AI model based on the received long message, extracts the key points, and generates a summary. The server inputs the following prompt to the generative AI model:

[2154] Please summarize the following lengthy email:

[2155] [Long message]

[2156] The generative AI model performs the analysis, generates a summary, and sends it back to the server. The resulting summary includes key points.

[2157] Step 12:

[2158] The server sends the generated summary to the terminal, which displays the summary to the user and provides an interface for confirmation, allowing the user to check the summary.

[2159] 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.

[2160] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize e-mails, and in particular, by combining an emotion engine, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion recognition technology to reduce the burden on the user and achieve fast and effective e-mail management.

[2161] Compose a new email

[2162] When a user creates a new email, the server prompts the user for the recipient, subject, and message body. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and an emotion engine to analyze the entered information and automatically generate an appropriate subject and message body, taking into account the user's emotions. The generated email is then sent to the device, where the user can view and edit it.

[2163] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[2164] Subject: "Great weekend plans"

[2165] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[2166] Create a reply email

[2167] When a user clicks the reply button to create a reply email, the server retrieves the user's past email correspondence. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[2168] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[2169] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[2170] Email correction

[2171] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[2172] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[2173] Before: "I might be late for the meeting tomorrow."

[2174] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[2175] Email Summary

[2176] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[2177] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[2178] Summary: "There was some positive progress in the meeting the other day."

[2179] In this way, the present invention provides a system that allows users to efficiently compose, reply to, correct, and summarize emails. In addition, by combining it with an emotion engine, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[2180] The processing flow will be explained below.

[2181] Compose a new email

[2182] Step 1:

[2183] The user presses the button to create a new email. The device sends a request to the server to create a new email.

[2184] Step 2:

[2185] The server generates an email creation screen including prompts for "recipient," "subject," and "body," and sends it to the terminal.

[2186] Step 3:

[2187] The device displays an email creation screen, and the user enters information in the "To," "Subject," and "Body" fields.

[2188] Step 4:

[2189] When the user completes the input, the terminal transmits the input information to the server.

[2190] Step 5:

[2191] The server uses an emotion engine to analyze emotions from the user's input information.

[2192] Step 6:

[2193] The server uses natural language processing (NLP) technology to analyze the input content and sentiment information and generate an appropriate subject line and body of the message.

[2194] Step 7:

[2195] An email containing the generated subject and body is sent to the terminal, and the terminal displays the email.

[2196] Step 8:

[2197] The user reviews the generated email and edits it if necessary.

[2198] Create a reply email

[2199] Step 1:

[2200] The user presses the reply button, and the device sends a reply email creation request to the server.

[2201] Step 2:

[2202] The server retrieves the target email and the most recent exchanges from the user's past email database.

[2203] Step 3:

[2204] The server analyzes the content of past emails and the corresponding emotional information.

[2205] Step 4:

[2206] The server uses an emotion engine to generate a reply that takes emotion information into consideration.

[2207] Step 5:

[2208] An email containing the generated reply content is sent to the terminal, and the terminal displays the reply email to the user.

[2209] Step 6:

[2210] The user reviews the generated reply email and edits it if necessary.

[2211] Email correction

[2212] Step 1:

[2213] The user requests correction of an email. The terminal sends the email body to the server.

[2214] Step 2:

[2215] The server receives the email body and analyzes the user's emotions using an emotion engine.

[2216] Step 3:

[2217] The server detects grammatical and stylistic errors and generates suggested revisions that take emotional information into account.

[2218] Step 4:

[2219] The proposed corrections and the corrected email are sent to the terminal, which displays the correction results to the user.

[2220] Step 5:

[2221] The user checks the correction results and, if necessary, reflects the suggested corrections.

[2222] Email Summary

[2223] Step 1:

[2224] The user sends a request to summarize a long email. The terminal sends the email body to the server.

[2225] Step 2:

[2226] The server receives the email body and analyzes the user's emotions using an emotion engine.

[2227] Step 3:

[2228] The server analyzes the email content and extracts important points, taking into account emotional information.

[2229] Step 4:

[2230] A summary is generated based on the extracted points and sent to the terminal.

[2231] Step 5:

[2232] The terminal displays the summary to the user, who then checks the summary and edits it as necessary.

[2233] The above is the specific operation of each processing step in the email management system combined with the emotion engine.

[2234] Example 2

[2235] 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."

[2236] In conventional email systems, users often have to spend a lot of time and effort composing, replying, correcting, and summarizing emails. Furthermore, it is difficult to express emotions in email content, which hinders smooth communication.

[2237] 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. In this invention, the server includes means for automatically generating a subject and a body of a new email based on a user's input, means for automatically generating a body of a reply message based on past message exchanges, means for detecting grammatical and stylistic errors in the message and suggesting corrections, means for extracting important points of the message and generating a summary, and means for generating message content taking the user's emotions into consideration using emotion analysis technology. This enables the user to efficiently compose, reply to, correct, and summarize emails, thereby realizing communication that takes the user's emotions into consideration.

[2238] The "means for automatically generating the subject and body of a new email" is a means for automatically creating an appropriate subject and body based on the content entered by the user using natural language processing technology.

[2239] The "means for automatically generating the body of a reply message" is a means for analyzing past message exchanges and automatically generating the body of a reply based on the analysis.

[2240] The "means for detecting grammatical and stylistic errors and suggesting corrections" is a means for detecting grammatical and stylistic errors in the email body and suggesting corrections to the user.

[2241] The "means for extracting important points and generating an outline" is a means for automatically extracting important points from the content of a long email and generating a summary that concisely summarizes the content.

[2242] "Means for generating message content taking into account the user's emotions using emotion analysis technology" refers to means for automatically generating message content that takes into account the user's emotions by analyzing the content entered by the user and using technology to infer the emotions behind it.

[2243] This invention relates to a system that allows users to efficiently compose, reply, correct, and summarize e-mails, and in particular, by combining emotion analysis technology, it enables e-mail management that takes into account the user's emotions. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective e-mail management.

[2244] First, we generate a program for this system. This program has the following functions:

[2245] 1. When a user composes a new email, the server automatically generates an appropriate subject and body based on the user's input.

[2246] 2. When creating a reply email, the server automatically generates the body of the reply email based on past message exchanges.

[2247] 3. When editing emails, the server detects grammatical and stylistic errors and suggests corrections.

[2248] 4. When summarizing an email, the server extracts the important points and generates a summary.

[2249] 5. At each processing step, emotion analysis technology is used to consider the user's emotions.

[2250] Compose a new email

[2251] When a user creates a new email, the server prompts the user for the recipient, subject, and body of the email. The user enters the required information according to the prompts, and the device sends the information to the server. The server uses natural language processing (NLP) technology and a sentiment analysis engine (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to analyze the entered information and automatically generate an appropriate subject and body of the email, taking into account the user's sentiment. The generated email is then sent to the device, where the user can view and edit it.

[2252] Specific examples

[2253] Consider the case where the user enters the following:

[2254] To: example@example.com

[2255] Subject: Weekend plans

[2256] Body: Hello. What are your plans for the weekend?

[2257] The server receives this and automatically generates it as follows:

[2258] Subject: Great weekend plans

[2259] Body: Hello, I'd like to reminisce about our fun conversation the other day and confirm your plans for the weekend. I'm looking forward to it.

[2260] Create a reply email

[2261] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server uses the content of past emails and an emotion analysis engine to generate a reply, taking into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[2262] Specific examples

[2263] When a user replies to an email, the server generates the following reply based on recent interactions and emotional information:

[2264] Response: Hello, I'm a little surprised by your message the other day. Could you please confirm the following points again?

[2265] Email correction

[2266] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses a sentiment analysis engine to generate suggested corrections that take the user's feelings into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[2267] Specific examples

[2268] If the user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[2269] Before: I might be late for the meeting tomorrow.

[2270] Corrected: I might be a little late for tomorrow's meeting, but I hope you understand the reason.

[2271] Email Summary

[2272] When a user summarizes a long email, the device sends the email content to the server. The server analyzes the content and extracts key points using a sentiment analysis engine. A summary is generated based on the extracted points and sent to the device for the user to review.

[2273] Specific examples

[2274] If a long email says "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[2275] Summary: There was some positive progress in the recent meeting.

[2276] In this way, the present invention provides a system that allows users to efficiently compose, reply, correct, and summarize emails. In addition, by combining it with emotion analysis technology, it accurately reflects the user's emotions, enabling more appropriate and effective email management.

[2277] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2278] Compose a new email

[2279] Step 1:

[2280] The user opens a screen to compose a new email. The user clicks the "Create new email" button on the device. This causes the device to send a request to the server to compose a new email.

[2281] Input: User clicks "Create New Email" button

[2282] Output: A request to the server to create a new email

[2283] Step 2:

[2284] The server generates input prompts for "recipient," "subject," and "body" and sends them to the terminal. The terminal displays an input form to the user.

[2285] Input: Triggering prompt generation on the server side

[2286] Output: Input prompt sent to terminal

[2287] Step 3:

[2288] The user enters the "recipient," "subject," and "body" and presses the send button. The device sends the input data to the server.

[2289] Input: "To", "Subject", and "Body" entered by the user

[2290] Output: Input data sent to the server

[2291] Step 4:

[2292] The server analyzes the received data using NLP and sentiment analysis technologies. Specifically, the NLP model analyzes the text and understands its context and structure, while the sentiment engine determines the user's sentiment.

[2293] Input: Data entered by the user ("To", "Subject", "Body")

[2294] Data processing: Text analysis using NLP models, sentiment analysis using emotion engines

[2295] Output: Analysis results

[2296] Step 5:

[2297] Based on the analysis results, the server generates an appropriate subject line and body text. Specifically, it combines an NLP model and an emotion engine to automatically generate email content that reflects the user's intentions and emotions.

[2298] Input: Analysis results

[2299] Data processing: Email content generation using NLP models and emotion engines

[2300] Output: Auto-generated subject and body

[2301] Step 6:

[2302] The server sends the generated email to the terminal, and the terminal displays the contents of the generated email to the user so that the user can check and edit them.

[2303] Input: Auto-generated subject and body

[2304] Output: Email content sent to the terminal

[2305] Create a reply email

[2306] Step 1:

[2307] When the user presses the reply button for the email they want to reply to, the device sends a request to the server to create a reply email.

[2308] Input: The act of the user pressing the reply button

[2309] Output: Reply mail creation request to the server

[2310] Step 2:

[2311] The server retrieves the user's past email correspondence, retrieves the email to be replied to and the past conversation history, and prepares for analysis.

[2312] Input: The request sent to the server

[2313] Output: Retrieved email history

[2314] Step 3:

[2315] The server uses NLP and emotion analysis technologies to analyze the content and emotional information of past emails. Specifically, it uses NLP to analyze the structure of the text and an emotion engine to determine the importance of the emotion.

[2316] Input: Retrieved past email history

[2317] Data processing: Text and sentiment analysis using NLP models and sentiment engines

[2318] Output: Analysis results

[2319] Step 4:

[2320] The server generates an appropriate reply based on the analysis results, automatically generating a reply email that reflects the user's past emotional state.

[2321] Input: Analysis results

[2322] Data processing: Generate reply content using NLP models and emotion engines

[2323] Output: Auto-generated reply

[2324] Step 5:

[2325] The server sends the generated reply email to the terminal, and the terminal displays the generated reply content to the user so that the user can check and edit it.

[2326] Input: Auto-generated reply

[2327] Output: Reply email sent to the device

[2328] Email correction

[2329] Step 1:

[2330] The user opens an input screen to request corrections to the email content. The user enters the email body they want to correct and presses the send button. The device then sends this to the server.

[2331] Input: The email body entered by the user

[2332] Output: Correction request to the server

[2333] Step 2:

[2334] The server analyzes the email body received to detect grammatical and stylistic errors, using NLP models and a sentiment analysis engine.

[2335] Input: The email body entered by the user

[2336] Data processing: Error detection using NLP models and emotion engines

[2337] Output: Error detection results

[2338] Step 3:

[2339] The server generates suggestions for correcting grammar and style errors, providing sentiment-sensitive suggestions based on the analysis results.

[2340] Input: Error detection result

[2341] Data processing: Generating revision suggestions using NLP models and emotion engines

[2342] Output: Revision proposal

[2343] Step 4:

[2344] The server sends the proposed revisions to the terminal, which displays them to the user so that the user can confirm and apply them.

[2345] Input: Proposed fix

[2346] Output: Suggested fixes sent to the device

[2347] Email Summary

[2348] Step 1:

[2349] The user opens an input screen to request a summary of a long email. The user enters the email text they want to summarize and presses the send button. The terminal then sends this to the server.

[2350] Input: A long email body entered by the user

[2351] Output: Summary request to server

[2352] Step 2:

[2353] The server analyzes the email text to be summarized and extracts key points using NLP technology and a sentiment analysis engine.

[2354] Input: A long email body entered by the user

[2355] Data processing: Text analysis and key point extraction using NLP models and sentiment engines

[2356] Output: Analysis results

[2357] Step 3:

[2358] The server generates a summary based on the analysis results, taking into account emotional information and creating a concise summary.

[2359] Input: Analysis results

[2360] Data processing: Generating summaries using NLP models and emotion engines

[2361] Output: Auto-generated summary

[2362] Step 4:

[2363] The server sends the generated summary to the terminal, which displays the summary to the user for the user to review.

[2364] Input: Auto-generated summary

[2365] Output: Summary sent to terminal

[2366] (Application example 2)

[2367] 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."

[2368] Conventional email management systems provide support for users to efficiently compose, reply, correct, and summarize emails, but they do not support generating emails that take user emotions into account or real-time customer support. As a result, when effective communication with customers is required, especially in physical stores, it is difficult for store staff to respond quickly and appropriately. To solve this problem, a system is needed that includes email management that takes user emotions into account and real-time customer support functions.

[2369] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically generating a subject and body of a new email based on user input, means for automatically generating the body of a reply email based on past email exchanges, means for detecting grammatical and stylistic errors in the email and suggesting corrections, means for extracting important points from the email and generating a summary, means for automatically generating replies to customer inquiries using emotion analysis technology, and means for providing real-time customer service support based on customer emotions via a smart device. This enables users to efficiently manage their emails while providing quick and accurate responses to customers in physical stores.

[2370] A "user" is a person or entity who uses the system to compose, reply to, correct, or summarize emails.

[2371] A "new email" is an email that a user creates and sends for the first time.

[2372] The "subject" is the title part that briefly indicates the content of a new email or a reply email.

[2373] The "body" is the part where the main content of the email is written.

[2374] A "server" is a computer system that processes information entered by users and performs various automatic generation and analysis.

[2375] "Natural language processing" is a technology that analyzes and understands human language, and is primarily used for automatic text generation and sentiment analysis.

[2376] "Emotion analysis technology" is a technology that analyzes a user's emotional state from text or voice and extracts that information.

[2377] "Past email exchanges" refers to the content of emails exchanged between the user and other people in the past.

[2378] A "reply email" is an email sent in response to a received email.

[2379] A "grammatical error" is an error in the language rules contained in the body of an email.

[2380] A "style error" is an error in format or expression contained within the body of an email.

[2381] "Suggested fixes" are suggestions for correcting grammatical or stylistic errors.

[2382] A "summary" is a simplified version of a long email message that extracts the important points.

[2383] A "customer" is a person who visits a physical store to make an inquiry or purchase.

[2384] A "smart device" is an electronic device that has Internet connectivity and can run various applications.

[2385] This invention relates to a system that allows users to efficiently compose, reply to, correct, and summarize emails. In particular, by combining an emotion engine, it enables email management that takes user emotions into consideration and customer service support in physical stores. This system utilizes natural language processing technology and emotion analysis technology to reduce the burden on users and achieve fast and effective email and customer service.

[2386] Compose a new email

[2387] When a user composes a new email, they follow the prompts displayed on their device to enter the recipient, subject, and message body. The device then sends this information to the server. The server then uses natural language processing technology and an emotion engine to analyze the entered information and generate an appropriate subject and message body, taking the user's emotions into account. The generated email is then sent to the device, where it can be viewed and edited by the user.

[2388] For example, if a user enters "example@example.com" in the "To" field, "About your weekend plans" in the "Subject" field, and "Hello. What are your plans for the weekend?" in the "Body" field, the server will analyze the user's sentiment and generate the following subject and body:

[2389] Subject: "Great weekend plans"

[2390] Body text: "Hello, I'd like to reminisce about our lovely conversation the other day and check in with you about your plans for the weekend. I'm looking forward to it."

[2391] Create a reply email

[2392] When a user presses the reply button for a specific email, the server retrieves the user's past email exchanges. The server analyzes the content of the past emails and uses an emotion engine to generate a reply that takes into account the user's emotions in the dialogue history. The generated reply email is sent to the terminal, where the user can view and edit it.

[2393] For example, when a user replies to an email, the server generates the following reply content based on the most recent exchange and emotional information:

[2394] Response: "Hello, I'm a bit surprised by your message. Could you please confirm the following points again?"

[2395] Email correction

[2396] When a user requests corrections to an email, the device sends the entered email text to the server. The server detects grammatical and stylistic errors and uses an emotion engine to generate suggested corrections that take the user's emotions into account. The suggested corrections are then sent to the device, where the user can review and apply them.

[2397] For example, if a user enters "I might be late for the meeting tomorrow," the server generates the following correction suggestions:

[2398] Before: "I might be late for the meeting tomorrow."

[2399] Correction: "I might be a little late for tomorrow's meeting, but I hope you understand why."

[2400] Email Summary

[2401] When a user summarizes a long email, the device sends the email content to the server, which analyzes the content and uses an emotion engine to extract key points. A summary is generated based on the extracted points and sent to the device for the user to review.

[2402] For example, if a long email reads, "Hello, regarding the meeting we had the other day, the following progress has been made...", the server will take into account the sentiment information and generate a summary like this:

[2403] Summary: "There was some positive progress in the meeting the other day."

[2404] Physical store applications

[2405] In physical stores, store clerks use smart devices (e.g., smart glasses) to assist customers. When a customer sends an inquiry email to a store clerk, the store clerk checks the email via the smart device, and the server performs sentiment analysis to generate an appropriate reply.

[2406] For example, if a customer asks, "Please tell me about new products arriving," the server will generate the following reply if it detects a positive emotion:

[2407] Response: "Thank you for your question about the new product arrivals. We expect new products to arrive this weekend."

[2408] This system allows users to manage emails more efficiently and also enables them to respond to customers more effectively in physical stores. In addition, the emotion engine enables communication that accurately reflects the emotions of users and customers.

[2409] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2410] Step 1:

[2411] The user enters the "recipient," "subject," and "body" into the terminal. The terminal sends this to the server. Input data: the recipient, subject, and body entered by the user. Output data: the input data sent to the server.

[2412] Step 2:

[2413] The server analyzes the received input data using natural language processing and sentiment analysis technologies. Input data: User's address, subject, and text. Output data: Analyzed sentiment information and natural language processing results.

[2414] Step 3:

[2415] Based on the analysis results, the server uses a generative AI model to automatically generate the subject and body of a new email. Input data: Analyzed emotional information and natural language processing results. Output data: Automatically generated subject and body.

[2416] Step 4:

[2417] The generated email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated email sent from the server. Output data: Final email edited by the user.

[2418] Step 5:

[2419] When a user presses the reply button for a specific email, the device sends that information to the server. Input data: Email information from which the reply was made. Output data: Reply information sent to the server.

[2420] Step 6:

[2421] The server acquires past email exchanges and analyzes them using emotion analysis technology. Input data: Past email exchanges. Output data: Analyzed dialogue history and emotion information.

[2422] Step 7:

[2423] Based on the analysis results, the server uses a generative AI model to automatically generate the reply email body. Input data: Analyzed dialogue history and emotional information. Output data: Automatically generated reply body.

[2424] Step 8:

[2425] The generated reply email is sent to the terminal, where the user can check it and edit it if necessary. Input data: Automatically generated reply email. Output data: Final reply email edited by the user.

[2426] Step 9:

[2427] When a user requests corrections to an email, the terminal sends the email body to the server. Input data: The email body entered by the user. Output data: The email body sent to the server.

[2428] Step 10:

[2429] The server detects grammatical and style errors and generates correction suggestions using sentiment analysis technology. Input data: User's email body. Output data: Detected error information and correction suggestions.

[2430] Step 11:

[2431] The generated correction proposal is sent to the terminal, where the user can check it and apply it as necessary. Input data: The generated correction proposal. Output data: The correction email applied by the user.

[2432] Step 12:

[2433] When a user requests a summary of an email, the terminal sends the email content to the server. Input data: Content of the long email. Output data: Content of the email sent to the server.

[2434] Step 13:

[2435] The server analyzes the email content, extracts key points using sentiment analysis technology, and generates a summary. Input data: Long email content. Output data: Extracted key points and generated summary.

[2436] Step 14:

[2437] The generated summary is sent to the terminal and confirmed by the user. Input data: Generated summary. Output data: Summary confirmed by the user.

[2438] Step 15:

[2439] When a user uses a smart device to deal with a customer in a physical store, the device receives the customer's inquiry email and sends it to the server. Input data: Customer's inquiry email. Output data: Inquiry email sent to the server.

[2440] Step 16:

[2441] The server uses emotion analysis technology to analyze customer emotions and generate appropriate replies. Input data: Customer inquiry email and emotion analysis results. Output data: Automatically generated replies.

[2442] Step 17:

[2443] The generated reply is sent to the smart device, and the store clerk responds appropriately to the customer. Input data: Automatically generated reply. Output data: Response provided by the store clerk to the customer.

[2444] 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.

[2445] 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.

[2446] 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.

[2447] 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.

[2448] 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.

[2449] 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.

[2450] 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).

[2451] 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 ...

Claims

1. A means for automatically generating a subject and body of a new email based on user input; A method to automatically generate the body of reply emails based on past email exchanges, A way to detect grammatical and stylistic errors in emails and suggest corrections; The system includes a means to extract key points from emails and generate summaries.

2. 2. The system according to claim 1, wherein the means for automatically generating the subject and body of the new email utilizes natural language processing.

3. 2. The system according to claim 1, wherein the means for automatically generating the reply email body analyzes the contents of past emails and generates the reply content taking into account a dialogue history.

4. 2. The system of claim 1, wherein the means for detecting grammatical and stylistic errors and suggesting corrections utilizes natural language processing.

5. 2. The system according to claim 1, wherein the means for extracting important points of the email and generating a summary utilizes natural language processing.

Images