System

Abstract

To provide a system.SOLUTION: A system includes means for: checking a destination domain and whether or not a title is present before transmitting a mail; checking and correcting typographical errors and omissions in content of the mail using a natural language processing system; displaying the corrected state in a preview; checking and correcting expressions relating to harassment or expressions being considerate of sustainable development goals; analyzing an emotion of a user and adjusting the expression of the mail according to the emotion; learning feedback from the user and reflecting a learning result in checking and correcting the typographical errors and omissions from next time onwards; and learning feedback from the user and reflecting the learning result in recognition of the emotion and correction of the mail from next time onwards.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

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

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the current email system, users must check for typos, omissions, and inappropriate expressions before sending an email, which is a burden for users. Furthermore, mistakes due to self-checking are unavoidable, which may result in rudeness or misunderstandings to the recipient. [Means for solving the problem]

[0005] To solve this problem, we provide a means to check the destination domain and whether or not there is a title before sending an email, a means to check and correct typos and omissions in the email content using a natural language processing system, and a means to display the corrected state in a preview. Furthermore, the natural language processing system checks and corrects harassment-related expressions and expressions that take SDGs into consideration, learns from user feedback, and reflects the results in checking and correcting typos and omissions from the next time onwards. This saves users the trouble of having to check for typos, omissions, and inappropriate expressions in their emails themselves, and helps them avoid being rude to the other person or causing misunderstandings. [Brief description of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 2 is a sequence diagram showing a process flow of the data processing system in the first embodiment of the first form example. [Figure 12] FIG. 11 is a sequence diagram showing a process flow of the data processing system in the application example 1 of the first embodiment. [Figure 13] FIG. 11 is a sequence diagram showing a process flow of the data processing system in the second embodiment of the second embodiment. [Figure 14] FIG. 13 is a sequence diagram showing a process flow of the data processing system in application example 2 of embodiment 2. [Figure 15] FIG. 13 is a sequence diagram showing a process flow of a data processing system in a third embodiment of the present invention. [Figure 16] FIG. 13 is a sequence diagram showing a process flow of the data processing system in the third application example of the third embodiment. [Figure 17] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when combined with an emotion engine. [Figure 19] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system according to the embodiment 2 of the second embodiment in the case where an emotion engine is combined. [Figure 20] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when combined with an emotion engine. [Figure 21] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the embodiment 3 of the form example 3 in the case where an emotion engine is combined. [Figure 22] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when combined with an emotion engine. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0011] 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.

[0012] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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 wide area network (WAN) and / or a local area network (LAN).

[0018] 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.

[0019] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] 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 (e.g., voice and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs voice according to instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Fig. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific process program 56 executed on the RAM 30.

[0024] 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.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores a reception output program 60. The reception output program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads out the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output processing 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] As one embodiment of the present invention, an email sending system is provided. This system has a function for checking the destination domain and whether or not there is a title when a user creates an email. Specifically, when a user inputs a destination of an email, the system automatically checks whether the destination domain is an internal domain or an external domain. Furthermore, if the title of the email has not been input, the system prompts the user to input a title.

[0029] "Example 2"

[0030] Furthermore, this system has the function of checking and correcting typos and omissions in the email content using a natural language processing system. Specifically, when a user enters the body of the email, the natural language processing system analyzes the content and checks for typos and omissions. If a typographical error or omission is found, the system corrects that part and displays the corrected state in a preview.

[0031] "Example 3"

[0032] The system also has the ability to check and correct harassment-type language and language that takes SDGs into consideration. Specifically, the natural language processing system checks the content of emails for harassment-type language and language that takes SDGs into consideration. If any such language is found, the system corrects the language and displays the corrected state in a preview.

[0033] "Example 4"

[0034] Furthermore, the system has the ability to learn from feedback from users and reflect the results in checking and correcting typos and omissions from the next time onwards. Specifically, when a user provides feedback on corrections suggested by the system, the natural language processing system learns from that feedback and reflects it in checking and correcting typos and omissions from the next time onwards. This allows the system to better adapt to the needs of the user.

[0035] The process flow of each embodiment will be described below.

[0036] "Example 1"

[0037] Step 1: The user enters the address to which the email should be sent.

[0038] Step 2: The system will automatically check whether the destination domain is an internal or external domain.

[0039] Step 3: If the email subject is not entered, the system will prompt the user to enter a subject.

[0040] "Example 2"

[0041] Step 1: The user enters the body of the email.

[0042] Step 2: The natural language processing system analyzes the content and checks for typos and omissions.

[0043] Step 3: If any typos or omissions are found, the system will correct them and display the corrected version in a preview.

[0044] "Example 3"

[0045] Step 1: The natural language processing system checks the content of the email for any harassing language or language that takes into consideration the SDGs.

[0046] Step 2: If a matching expression is found, the system corrects the expression and displays the corrected version in a preview.

[0047] "Example 4"

[0048] Step 1: The user provides feedback on the system's suggested fixes.

[0049] Step 2: The natural language processing system learns from the feedback and uses it to check for and correct typos and omissions in future sentences.

[0050] Example 1

[0051] Next, a description will be given of Example 1 of the embodiment 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".

[0052] In conventional email sending systems, users are often required to manually check the destination domain and whether the email title is correct, which is time-consuming. In addition, there is a risk of email quality declining due to insufficient checking and correction of typos, omissions, harassing expressions, and expressions that take into consideration the Sustainable Development Goals. Furthermore, there is no mechanism for reflecting user feedback in subsequent checks and corrections, which means that improvements to the system are not progressing.

[0053] 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.

[0054] In this invention, the server includes a means for the user to input the destination of the email, a means for the server to check the domain of the destination, a means for the server to check whether the email has a title, a means for the server to prompt the user to input a title, a means for checking and correcting typos and omissions in the email content using a natural language processing system, and a means for displaying the corrected state in a preview. This makes it possible to reduce the user's work and improve the quality of emails. In addition, by learning from user feedback and reflecting it in subsequent checks and corrections, the system can be continuously improved.

[0055] "User" means any individual or entity that uses the System to create and send Email.

[0056] A "server" is a computer system that accepts and processes input from users.

[0057] A "destination domain" is the portion of an email address following the @ symbol, and indicates the network to which the email is being sent.

[0058] "Email title" is the text entered as the subject of the email.

[0059] A "natural language processing system" is software or algorithms that analyze text data and check for and correct typos, omissions, and specific expressions.

[0060] "Checking for and correcting typos and omissions" refers to the process of detecting spelling mistakes and input errors in the content of an email and correcting them to the correct expression.

[0061] "Harassing language" is any word or phrase that may cause discomfort or offense to others.

[0062] "SDG-friendly expressions" are words and phrases that are ethical and socially sensitive and are in line with the Sustainable Development Goals (SDGs).

[0063] "Feedback" refers to opinions and evaluations provided by users, and is information that is useful for improving the system.

[0064] "Preview" is a display screen where the user can check the content of the email after correction.

[0065] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0066] The present invention relates to an email sending system that checks the destination domain and whether or not there is a title when a user creates an email. When a user enters the destination of an email, the system automatically checks whether the destination domain is an internal domain or an external domain. If the title of the email has not been entered, the server prompts the user to enter a title.

[0067] Hardware and Software Used

[0068] This system is realized using the following hardware and software.

[0069] Server: A computer system that handles the central processing of the mail delivery system. It uses a Python script to verify domains.

[0070] Database: A database system for storing the internal domain list, specifically using MySQL (registered trademark).

[0071] Terminal: A device on which a user creates an email. The presence or absence of a title is checked using JavaScript (registered trademark) that runs on the browser.

[0072] System Operation

[0073] When a user enters the destination of an email, the server extracts the domain part of the destination email address and checks it against the internal domain list, which is stored in the server's database. Next, if the user has not entered the subject of the email, the server prompts the user to enter a subject. This function is implemented on the browser using JavaScript.

[0074] Examples

[0075] For example, if a user enters the email address "example@external.com", the server will check that "external.com" is not in the internal domain list. Then, if the user clicks the submit button without entering a title, JavaScript will display an alert saying "Please enter a title."

[0076] Example prompts for generative AI models

[0077] "Please generate a program for an email sending system. When a user enters the destination of an email, the server checks whether the destination domain is an internal domain or an external domain, and if the subject of the email has not been entered, it prompts the user to enter a subject."

[0078] The above is an embodiment of the present invention. This system makes it possible to reduce the work for users and improve the quality of emails. In addition, the system can continuously improve by learning from user feedback and reflecting it in subsequent checks and corrections.

[0079] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0080] Program processing flow

[0081] Step 1: User enters destination for email

[0082] The user enters the destination email address on the email composition screen. For example, they enter "example@external.com." This input is the starting point for the system's processing.

[0083] Input: The email address entered by the user (e.g. example@external.com)

[0084] Output: The entered email address is sent to the server.

[0085] Step 2: The server checks the destination domain

[0086] The server extracts the domain part from the email address it receives -- specifically, the part of the email address that comes after the @ sign -- and then checks whether this domain is on its internal domain list, which is stored in the server's database.

[0087] Input: The email address entered by the user (e.g. example@external.com)

[0088] Data processing: Extraction of domain part (e.g. external.com)

[0089] Data calculation: Check against internal domain list

[0090] Output: Result of whether the domain is internal or external

[0091] Step 3: The server checks whether the email title exists

[0092] When the user clicks the send button, the server checks if the email subject field is empty. This check is done in the browser using JavaScript.

[0093] Input: Email title entered by the user (e.g. blank)

[0094] Data Calculation: Check if title field is empty

[0095] Output: Check whether a title has been entered.

[0096] Step 4: Server prompts user for title (if necessary)

[0097] If no subject is entered, the server will display an alert to the user and prompt them to enter a subject. The email will not be sent unless the user enters a subject.

[0098] Input: The result indicates that the title has not been entered.

[0099] Specific behavior: JavaScript displays an alert saying "Please enter a title"

[0100] Output: An alert prompting the user to enter a title.

[0101] Step 5: Check the email for typos and omissions using a natural language processing system and correct them.

[0102] The server passes the email content to a natural language processing system, which checks for and corrects typos and omissions. The corrected email content is then displayed to the user as a preview.

[0103] Input: Email content entered by the user

[0104] Data Calculation: Checking and correcting typos and omissions using natural language processing systems

[0105] Output: Email content after correction

[0106] Step 6: Preview the result after correction

[0107] The server then displays a preview of the revised email to the user, allowing the user to review the preview and make further edits if necessary.

[0108] Input: Email content after correction

[0109] Specific operation: The revised email content is displayed on the preview screen.

[0110] Output: A preview screen that the user can see

[0111] The above are the specific processing steps of the program of this system.

[0112] (Application example 1)

[0113] Next, a description will be given of Application Example 1 of Form 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."

[0114] In the conventional email sending system, there was insufficient checking of the destination domain and the presence or absence of a title, which often led to communication problems due to erroneous sending or not entering a title. In addition, there was a lack of functionality to notify users of security risks when sending to external domains, which increased the risk of information leakage. Furthermore, checking and correcting typos in email content was often done manually, which was inefficient.

[0115] 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.

[0116] In this invention, the server includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for notifying security risks when sending to an external domain, and means for prompting input of a title if one has not been entered. This makes it possible to prevent problems with erroneous sending of emails and not entering a title, reduce security risks when sending to an external domain, and efficiently check and correct typos and omissions in the email content.

[0117] "Checking the destination domain before sending email" means automatically checking whether the domain part of the destination email address is an internal domain or an external domain before sending the email.

[0118] "Checking whether or not there is a title" means checking whether or not the title of the email has been entered.

[0119] A "natural language processing system" is a system that analyzes text data and checks for and corrects typos, omissions, and specific expressions.

[0120] "Checking for and correcting typos and omissions" means detecting typos and omissions in the content of an email and correcting them.

[0121] "Display the corrected state in the preview" means visually displaying to the user the contents of the email after typos and omissions have been corrected.

[0122] "Notify of security risks when sending to external domains" means to inform the user of security risks when sending email to external domains.

[0123] "Prompt for input of title if not entered" means to prompt the user to input a title if the title of the email has not been entered.

[0124] A system for implementing this invention includes means for checking the destination domain before sending an email and for checking whether or not there is a title, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for notifying of security risks when sending to an external domain, and means for prompting input of a title if one has not been entered.

[0125] Hardware and Software Used

[0126] Hardware: Smartphones, servers

[0127] Software: Python, natural language processing library (e.g. spaCy), email sending library (e.g. smtplib)

[0128] System processing overview

[0129] 1. Domain Verification:

[0130] The server extracts the domain portion of the email address entered by the user and checks it against its internal domain list.

[0131] If it is an internal domain, it is identified as an "internal domain"; if it is an external domain, it is identified as an "external domain."

[0132] 2. Title Check:

[0133] The server checks whether the email subject entered by the user is empty.

[0134] If the title is not entered, a message is displayed prompting the user to enter a title.

[0135] 3. Checking and correcting typos using natural language processing:

[0136] The server uses a natural language processing system to analyze the email content and detect typos and missing characters.

[0137] Detected typos and omissions are automatically corrected and the corrected email content is generated.

[0138] 4. Preview display:

[0139] The revised email content is displayed as a preview to the user for confirmation.

[0140] 5. Security Risk Notification:

[0141] The server notifies the user that there is a security risk when sending to external domains.

[0142] The email will be sent after the user confirms.

[0143] Examples

[0144] For example, if a user attempts to send an email to "user@external.com", the server extracts the domain portion of this email address, "external.com", and compares it with the internal domain list. Because it is an external domain, the server notifies the user that there is a security risk. Also, if the email subject line has not been entered, a message is displayed prompting the user to enter one. Furthermore, if the email content contains typos or omissions, the natural language processing system detects them and automatically corrects them. The corrected email content is displayed as a preview to the user, and the email is sent after the user has confirmed it.

[0145] Examples of prompt statements

[0146] A user is trying to send an email. Check whether the destination domain is an internal domain or an external domain, and prompt the user to enter a subject if one has not been entered. Inform the user that there is a security risk when sending to an external domain.

[0147] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0148] Step 1:

[0149] The server receives the email address entered by the user. Given an email address as input, the server extracts the domain part from the email address. Specifically, it splits the email address at "@" and takes the latter part as the domain. As output, it gets the extracted domain.

[0150] Step 2:

[0151] The server checks the extracted domain against the internal domain list. Given the extracted domain and the internal domain list as input, the server checks whether the domain is included in the internal domain list. Specifically, it performs a lookup in the list and checks whether the domain is present in the list. As an output, it gives information about whether the domain is an internal or external domain.

[0152] Step 3:

[0153] The server checks whether the email title has been entered. The email title is given as input, and the server checks whether the title is empty. Specifically, it checks the length of the title string, and if the length is zero, it determines that the title has not been entered. As output, it obtains information about whether the title has been entered.

[0154] Step 4:

[0155] If the title is not entered, the server displays a message prompting the user to enter a title. Information that the title is not entered is given as input, and the server displays the message through the user interface. Specifically, a pop-up window or a notification message is displayed. As output, a notification is given to the user.

[0156] Step 5:

[0157] The server uses a natural language processing system to check the email content for typos and omissions and make corrections. The email content is given as input, and the server analyzes the text using a natural language processing library. Specifically, it applies text analysis algorithms to detect typos and omissions and make automatic corrections. The corrected email content is obtained as output.

[0158] Step 6:

[0159] The server displays the corrected email content to the user as a preview. The corrected email content is given as input, and the server displays the preview through a user interface. Specifically, the corrected text is displayed in a window or editor that displays the email content. The output is a preview display to the user.

[0160] Step 7:

[0161] The server notifies the user of the security risk when sending to an external domain. As input, information that the domain is an external domain is given, and the server displays a message to notify the user of the security risk. Specifically, it displays a warning message or an alert. As output, the user is notified of the security risk.

[0162] Step 8:

[0163] After the user confirms, the server will send the email. Given the user's confirmation information and the modified email content as input, the server will use the email sending library to send the email. Specifically, it will send the email to the mail server using the SMTP protocol. As output, the email will be sent.

[0164] Example 2

[0165] Next, a second embodiment of the second embodiment will be described. 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."

[0166] Conventional email sending systems have insufficient means for checking for typos and omissions before sending, which increases the risk of sending emails with incorrect content. Users also need to manually check the destination domain and the presence or absence of the title, which places a heavy burden on users. Furthermore, there is no checking for harassing expressions or expressions that take into consideration the Sustainable Development Goals, making it difficult to communicate appropriately.

[0167] 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.

[0168] In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omissions in the email content by a natural language processing system, a means for displaying the corrected state in a preview, a means for a user to input an email body and a terminal to send the email body to the server, a means for the server to pass the email body to the natural language processing system and for the natural language processing system to analyze the email body, a means for the server to correct typos and omissions and send the corrected email body to the terminal, and a means for the terminal to display the corrected email body in a preview. This allows the user to send an accurate email without typos or omissions, and the confirmation of the destination domain and the presence or absence of a title is also automated, reducing the burden on the user. In addition, the system also checks for harassment-related expressions and expressions that take into consideration the sustainable development goals, allowing for appropriate communication.

[0169] "Verifying destination domain before sending email" is the process of verifying whether the intended recipient domain is correct before an email is sent.

[0170] "Title check" is the process of checking whether an email has an appropriate title before it is sent.

[0171] A "natural language processing system" is a system that uses artificial intelligence technology to analyze text data and detect and correct typos, omitted words, grammatical errors, and so on.

[0172] "Checking for and correcting typos and omissions" is the process of detecting incorrect or missing characters in entered text and correcting them to their correct form.

[0173] "Displaying the corrected state in a preview" is a process of displaying the text after the typos and omissions have been corrected so that the user can visually confirm it.

[0174] "A user inputs the body of an email" refers to an action in which a user inputs the contents of an email in text format using a terminal.

[0175] "The terminal sends the email body to the server" refers to the process of sending the email body entered by the user from the terminal to the server via data communication.

[0176] "The server passes the email body to the natural language processing system" refers to the process in which the server inputs the email body it received into the natural language processing system and requests it to analyze it.

[0177] "A natural language processing system analyzes the body of an email" refers to the process in which a natural language processing system analyzes the body of an email and detects typos, omitted words, and grammatical errors.

[0178] "The server corrects typos and spelling errors" refers to the process in which the server corrects typos and spelling errors in the body of the email based on the analysis results from the natural language processing system.

[0179] "Sending the corrected email body to the terminal" refers to the process in which the server sends the corrected email body to the terminal via data communication.

[0180] "Displaying the revised email body in a preview on the terminal" refers to a process in which the terminal displays the revised email body received from the server so that the user can visually confirm it.

[0181] "Harassing language" refers to words or phrases that may offend or hurt others.

[0182] "SDG-friendly expressions" refer to words and phrases that are environmentally and socially conscious and in line with the Sustainable Development Goals (SDGs).

[0183] "Learning from user feedback" is the process by which a system collects opinions and suggestions for corrections from users and uses them to improve the system's performance.

[0184] This invention is a system for checking and correcting typos and omissions before sending an email. The system operates when a user inputs the body of an email and the terminal sends the email body to a server. The server passes the email body to a natural language processing system, which analyzes the email body. Based on the analysis results, the server corrects the typos and omissions and sends the corrected email body to the terminal. The terminal displays a preview of the corrected email body.

[0185] Hardware and Software Used

[0186] Hardware: Servers, terminals (PCs, smartphones, etc.)

[0187] Software: Natural language processing systems (e.g. SpaCy, GPT-3 (registered trademark)), email client software (e.g. Outlook, Gmail)

[0188] Processing flow

[0189] 1. The user enters the email body

[0190] The user uses the email client software on the terminal to input the body of the email, for example, "Hello, Tanaka-san. The meeting tomorrow is at 10:00. Thank you for your cooperation."

[0191] 2. The device sends the email body to the server

[0192] When the user finishes entering the email body and clicks a button to request checking for typos, the device sends the email body to the server by using an HTTP POST request.

[0193] 3. The server passes the email body to a natural language processing system

[0194] The server passes the received email text to a natural language processing system (e.g. SpaCy, GPT-3). Specifically, it sends the email text to the natural language processing system using an API request.

[0195] 4. A natural language processing system analyzes the email body

[0196] The natural language processing system analyzes the text of the received email and checks for typos and omissions. For example, it determines that "onegaishimasu" (please) should be corrected to "onegaishimasu" (please onegaishimasu).

[0197] 5. The server corrects typos

[0198] The server that receives the analysis results from the natural language processing system corrects any typos or omissions that are detected. Specifically, it performs character string manipulation to change "onegaishimasu" (please) to "onegaishimasu" (please please).

[0199] 6. The server sends the corrected email body to the device.

[0200] The server sends the revised email body to the terminal. Specifically, it returns the revised email body to the terminal as an HTTP response.

[0201] 7. Your device will display a preview of the revised email body.

[0202] The terminal displays a preview of the corrected email text received by the user. For example, the preview screen of the email client software displays "Hello, Tanaka-san. Tomorrow's meeting starts at 10:00. Thank you for your cooperation."

[0203] Examples

[0204] For example, if a user enters the following email body:

[0205] Hello, Tanaka-san.

[0206] The meeting tomorrow will start at 10:00. Thank you for your cooperation.

[0207] If this email text is passed to a natural language processing system, the system will correct "onegaishimasu" to "onegaishimasu."

[0208] The corrected email body will look like this:

[0209] Hello, Tanaka-san.

[0210] The meeting tomorrow will start at 10:00. Thank you.

[0211] Example prompts to be input to the generative AI model

[0212] Email body entered by the user:

[0213] Hello, Tanaka-san.

[0214] The meeting tomorrow will start at 10:00. Thank you for your cooperation.

[0215] Please check the body of this email for typos and correct them.

[0216] This prompt is fed into a generative AI model, which checks for and corrects typos.

[0217] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0218] Step 1:

[0219] The user enters the email body.

[0220] The user uses the email client software on the device to input the body of the email. For example, the user might type, "Hello, Tanaka-san. The meeting tomorrow is at 10:00. Thank you for your cooperation." The inputted body of the email is stored in the device's memory.

[0221] Step 2:

[0222] The device sends the email body to the server.

[0223] When the user finishes entering the email body and clicks a button to request a check for typos, the terminal sends the email body to the server. Specifically, the terminal uses an HTTP POST request to send the email body to the server. The input is the email body entered by the user, and the output is the email body sent to the server.

[0224] Step 3:

[0225] The server passes the email text to a natural language processing system

[0226] The server passes the received email body to the natural language processing system. Specifically, it sends the email body to the natural language processing system using an API request. The input is the email body sent to the server, and the output is the email body passed to the natural language processing system.

[0227] Step 4:

[0228] A natural language processing system analyzes the email body

[0229] The natural language processing system analyzes the body of the received email and checks for typos and omissions. For example, it determines that "onegaishimasu" should be corrected to "onegaishimasu." The input is the email body passed to the natural language processing system, and the output is the analysis result.

[0230] Step 5:

[0231] The server will correct any typos.

[0232] The server that receives the analysis results from the natural language processing system corrects any typos or omissions that are detected. Specifically, it performs string manipulation to correct "onegaishimasu" (please) to "onegaishimasu" (please). The input is the analysis results, and the output is the corrected email body.

[0233] Step 6:

[0234] The server sends the corrected email body to the device.

[0235] The server sends the revised email body to the terminal. Specifically, it returns the revised email body to the terminal as an HTTP response. The input is the revised email body, and the output is the revised email body sent to the terminal.

[0236] Step 7:

[0237] Your device will display a preview of the revised email body.

[0238] The terminal displays a preview of the corrected email body received by the terminal to the user. For example, the preview screen of the email client software displays "Hello, Tanaka-san. Tomorrow's meeting starts at 10:00. Thank you for your cooperation." The input is the corrected email body sent to the terminal, and the output is the corrected email body displayed to the user.

[0239] (Application example 2)

[0240] Next, a description will be given of Application Example 2 of Form 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."

[0241] In electronic payment services, if there are typos or omissions in transaction confirmation emails or customer notification emails, it can reduce user trust and compromise the security of transactions. In addition, there is also the problem that correcting typos and omissions takes time, which reduces work efficiency. Furthermore, because the history of corrections is not saved, past corrections cannot be checked, and there is a risk of repeating the same mistake.

[0242] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omissions in the email content by a natural language processing system, a means for displaying the corrected state in a preview, a means for saving the history of the correction content so that it can be checked later, and a means for the natural language processing system to check and correct typos and omissions when the text of the transaction confirmation email or customer notification email is input. This makes it possible to automatically correct typos and omissions in the transaction confirmation email or customer notification email, and to increase the user's reliability. In addition, by saving the history of the correction content, it is possible to check past correction content and reduce the risk of repeating the same mistake.

[0243] "Checking the destination domain before sending email" means checking whether the domain part of the destination email address is correct before sending the email.

[0244] "Checking for the presence or absence of a title" means checking whether or not the subject of the email has been entered.

[0245] A "natural language processing system" is a technology that enables computers to understand, analyze, and generate human language.

[0246] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in a sentence and replacing them with the correct characters.

[0247] "Display the corrected state in the preview" means that the user can visually confirm the text after typos and omissions have been corrected.

[0248] "Saving a history of modifications so that they can be checked later" means saving a record of modifications made so that the contents of the modifications can be checked later.

[0249] "Transaction confirmation emails and customer notification emails" are emails used in electronic payment services to confirm transactions and notify customers.

[0250] "Learning feedback from users and reflecting the results in checking for and correcting typos and omissions from the next time onwards" means that the system will learn the opinions and corrections provided by users and use them to detect and correct typos and omissions from the next time onwards.

[0251] As an embodiment of the present invention, a system for automatically correcting typos and omissions in transaction confirmation emails and customer notification emails in electronic payment services will be described.

[0252] System Programming

[0253] This system is implemented using Python. A library called SpaCy is used for natural language processing. Specifically, the following processes are performed:

[0254] Processing Description

[0255] Hardware

[0256] Smartphone

[0257] software

[0258] Python

[0259] SpaCy (Natural Language Processing Library)

[0260] Data processing and calculation

[0261] 1. User enters email body:

[0262] The user enters the text of the transaction confirmation email or customer notification email into the smartphone application.

[0263] 2. The server receives the email body:

[0264] The server receives the body of the email sent by the user.

[0265] 3. Analysis by natural language processing system:

[0266] The server analyzes the email body using SpaCy's Japanese model (ja_core_news_sm). It detects and corrects typos and omissions from the analysis results.

[0267] 4. Preview the email body after editing:

[0268] The server displays a preview of the revised email body on the user's smartphone.

[0269] 5. Save revision history:

[0270] The server keeps a record of the modifications made so that they can be reviewed at a later time.

[0271] Examples

[0272] If a user enters the email body as "Hello, this is a test email. There may be typos and spelling errors.", the system will correct it to "Hello, this is a test email. There may be typos and spelling errors."

[0273] Examples of prompt statements

[0274] "Please correct the typos and spelling errors in the following email body: 'Hello, this is a test email. There may be typos and spelling errors.'"

[0275] In this way, typos and omissions in transaction confirmation emails and customer notification emails in electronic payment services can be automatically corrected, increasing user trust. In addition, by saving the history of corrections, past corrections can be checked, reducing the risk of repeating the same mistake.

[0276] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0277] Step 1:

[0278] The user enters the email body.

[0279] The user inputs the text of the transaction confirmation email or customer notification email into the smartphone application. The inputted text of the email is temporarily saved within the application.

[0280] Step 2:

[0281] The terminal sends the email body to the server.

[0282] The terminal sends the email text entered by the user to the server, where it is received and prepared for analysis.

[0283] Step 3:

[0284] The server receives the email body.

[0285] The server receives the body of the email sent from the terminal, and passes the received body of the email to the natural language processing system.

[0286] Step 4:

[0287] The server analyzes the email body using a natural language processing system.

[0288] The server analyzes the email body using SpaCy's Japanese model (ja_core_news_sm). The input for the analysis is the email body, and the output is the detection result of typos and omissions. Specifically, the server tokenizes the email body and analyzes each token to detect typos and omissions.

[0289] Step 5:

[0290] The server will correct any typos or omissions.

[0291] The server corrects typos and omissions based on the analysis results. The input for correction is the detection result of typos and omissions, and the output is the corrected email body. Specifically, the server replaces the parts where typos and omissions are detected with the correct characters.

[0292] Step 6:

[0293] The server sends the revised email body to the terminal for preview display.

[0294] The server sends the corrected email body to the terminal. The input for sending is the corrected email body, and the output is a notification of completion of sending to the terminal. Specifically, the server sends the corrected email body to the terminal, and the terminal receives it.

[0295] Step 7:

[0296] The device will display a preview of the revised email body.

[0297] The terminal receives the corrected email body sent from the server and displays a preview for the user. The input for the preview display is the corrected email body, and the output is a visual display for the user. Specifically, the terminal displays the corrected email body on the screen so that the user can confirm it.

[0298] Step 8:

[0299] The server stores a history of modifications.

[0300] The server keeps a record of the modifications made. The input to the save is the modifications, and the output is a notification that the save was completed. Specifically, the server saves the modifications in a database for later review.

[0301] Example 3

[0302] Next, a third embodiment of the third embodiment will be described. 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."

[0303] In conventional email sending systems, checking and correcting typos and omissions before sending was insufficient, and checking and correcting harassing expressions and expressions that take into consideration the Sustainable Development Goals was often not done. As a result, there was a risk that emails containing misleading or inappropriate expressions would be sent. In addition, the system lacked the ability to learn from user feedback and reflect it in future corrections, which resulted in low adaptability of the system.

[0304] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0305] In this invention, the server includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for collecting feedback from users and saving it in a database, means for learning the collected feedback and reflecting it in checking and correcting typos and omissions from the next time onwards, means for checking whether the email content contains harassment-related expressions or expressions that take into consideration the Sustainable Development Goals and correcting them as necessary, and means for sending the corrected email content to the user's terminal and displaying it as a preview. This makes it possible to effectively correct typos, omissions, and inappropriate expressions before sending an email and to improve the adaptability of the system by learning from user feedback.

[0306] "Destination domain verification" is the process of verifying whether the domain name of the email recipient is correct before the email is sent.

[0307] "Checking for the presence of a title" is a process of checking whether the title of an email is set appropriately before the email is sent.

[0308] A "natural language processing system" is a system that uses artificial intelligence technology to analyze text data and understand meaning and grammar.

[0309] "Checking for and correcting typos and omissions" refers to the process of detecting incorrect or missing characters in the content of an email and correcting them to the correct form.

[0310] "Preview" refers to the process of displaying the revised email content on the screen for the user to review.

[0311] "Collecting feedback and storing in a database" is the process of collecting opinions and ratings from users and storing them in a database.

[0312] "Learning from the feedback and reflecting it in checking and correcting typos and omissions from the next time onwards" refers to the process in which the system learns from the collected feedback and reflects the results of that learning when checking and correcting the content of emails from the next time onwards.

[0313] "Checking and correcting harassing language and language that takes into consideration the Sustainable Development Goals" refers to the process of detecting language contained in emails that is harassing or contrary to the Sustainable Development Goals, and correcting it to appropriate language.

[0314] "Send to user's device and display as preview" refers to the process of sending the revised email content to the user's device and displaying it for the user to review.

[0315] The present invention is a system that effectively corrects typos, omissions, and inappropriate expressions before sending an email, and improves the adaptability of the system by learning from user feedback. A specific embodiment of this system will be described below.

[0316] Hardware and Software Used

[0317] Hardware: Servers, terminals

[0318] Software: Natural language processing systems (e.g. GPT-4 (registered trademark)), database management systems (e.g. MySQL)

[0319] System Configuration

[0320] 1. Receiving email content

[0321] User composes an email and clicks the send button.

[0322] The terminal transmits the contents of the email created by the user to the server.

[0323] The server temporarily stores the contents of the received email.

[0324] 2. Email content analysis

[0325] The server transmits the stored email content to a natural language processing system.

[0326] A natural language processing system analyzes email content and understands the structure and meaning of the text.

[0327] 3. Check for harassment-related expressions and expressions that consider the SDGs

[0328] Based on the analysis results, the natural language processing system will check whether the email contains any harassing language or language that takes into consideration the SDGs.

[0329] The server receives the verification result and determines whether or not there is a matching expression.

[0330] 4. Generate the corrections

[0331] If a matching expression is found, the natural language processing system generates new text to correct the expression.

[0332] The server receives the generated modifications and merges them with the original mail content.

[0333] 5. Preview of edits

[0334] The server sends the corrected email content to the user's terminal.

[0335] The device will display a preview of the revised email content.

[0336] 6. Collecting User Feedback

[0337] The user provides feedback on the modifications.

[0338] The terminal transmits the user's feedback to the server.

[0339] The server stores the feedback in a database.

[0340] 7. Learning Feedback

[0341] The natural language processing system learns from the collected feedback and uses it to check and correct typos and omissions in future posts.

[0342] The server stores the learning results and uses them the next time it analyzes an email.

[0343] Examples

[0344] Example 1:

[0345] What the user is trying to send: "This project is not progressing at all. Try harder."

[0346] The server receives the email content and analyzes it using a natural language processing system.

[0347] The natural language processing system determines that the expression "No progress at all" is harassing, and corrects it to "Progress seems to be slow. Let's think together about ways to improve the situation."

[0348] The server sends the revised email content to the user's device and displays it as a preview.

[0349] Users provide feedback on the corrections, saying, "This expression is appropriate."

[0350] The server collects the feedback and stores it in a database.

[0351] The natural language processing system learns from the collected feedback and incorporates it into future email revisions.

[0352] Example prompt:

[0353] "Please check whether the content of this email contains harassing language and revise it if necessary."

[0354] Please check that the content of this email includes language that takes the SDGs into consideration and revise it if necessary.

[0355] In this way, the system can effectively correct typos, omissions, and inappropriate expressions before sending an email, and can learn from user feedback to improve the adaptability of the system.The flow of the identification process in the third embodiment will be described with reference to FIG.

[0356] Step 1:

[0357] Receiving email content

[0358] User composes an email and clicks the send button.

[0359] The terminal transmits the contents of the email created by the user to the server.

[0360] The server temporarily stores the contents of the received email.

[0361] Input: User-created email content

[0362] Output: Email contents stored on the server

[0363] Specific behavior:

[0364] When the user clicks the "Send" button, the terminal sends the email contents to the server.

[0365] The server stores the contents of the received email in a database.

[0366] Step 2:

[0367] Email content analysis

[0368] The server transmits the stored email content to a natural language processing system.

[0369] A natural language processing system analyzes email content and understands the structure and meaning of the text.

[0370] Input: Email contents stored on the server

[0371] Output: Structure and meaning of parsed email content

[0372] Specific behavior:

[0373] The server sends the email content to a natural language processing system via an API.

[0374] The natural language processing system tokenizes the email content and performs grammatical analysis.

[0375] Step 3:

[0376] Check for harassment-related expressions and expressions that consider the SDGs

[0377] Based on the analysis results, the natural language processing system will check whether the email contains any harassing language or language that takes into consideration the SDGs.

[0378] The server receives the verification result and determines whether or not there is a matching expression.

[0379] Input: Structure and meaning of parsed email content

[0380] Output: Detection results for harassment-related expressions and expressions that are insensitive to the SDGs

[0381] Specific behavior:

[0382] Natural language processing systems use algorithms to detect specific keywords and phrases.

[0383] The server analyzes the detection results and determines whether remediation is necessary.

[0384] Step 4:

[0385] Generate corrections

[0386] If a matching expression is found, the natural language processing system generates new text to correct the expression.

[0387] The server receives the generated modifications and merges them with the original mail content.

[0388] Input: Detection results for harassment-related expressions and expressions that are insensitive to the SDGs

[0389] Output: The modified email content

[0390] Specific behavior:

[0391] A natural language processing system replaces the original expression with a more appropriate expression.

[0392] The server stores the revised email content as new text.

[0393] Step 5:

[0394] Preview of the changes

[0395] The server sends the corrected email content to the user's terminal.

[0396] The device will display a preview of the revised email content.

[0397] Input: Modified email content

[0398] Output: A preview displayed on the user's device.

[0399] Specific behavior:

[0400] The server sends the corrected email content to the user's terminal.

[0401] The terminal displays the corrected email content on the screen and asks the user to confirm it.

[0402] Step 6:

[0403] Collecting user feedback

[0404] The user provides feedback on the modifications.

[0405] The terminal transmits the user's feedback to the server.

[0406] The server stores the feedback in a database.

[0407] Input: User feedback

[0408] Output: Feedback stored in a database

[0409] Specific behavior:

[0410] The user enters their feedback and clicks the "Submit" button.

[0411] The terminal transmits the feedback content to the server.

[0412] The server stores the feedback in a database.

[0413] Step 7:

[0414] Learning Feedback

[0415] The natural language processing system learns from the collected feedback and uses it to check and correct typos and omissions in future posts.

[0416] The server stores the learning results and uses them the next time it analyzes an email.

[0417] Input: Feedback stored in the database

[0418] Output: The trained model parameters.

[0419] Specific behavior:

[0420] The natural language processing system uses the feedback data to update the parameters of the model.

[0421] The server uses the updated model the next time it parses the email.

[0422] (Application example 3)

[0423] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0424] Email is an important means of modern communication, but it can contain typos, omissions, harassing language, or language that does not consider the Sustainable Development Goals. This can lead to misunderstandings and trouble, and can damage the credibility of companies and individuals. Another challenge is to effectively utilize feedback from users to improve the accuracy of the system. Furthermore, there is a demand to make these functions available on mobile devices such as smartphones.

[0425] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0426] In this invention, the server includes a means for checking the destination domain and whether or not there is a title before sending an email, a means for checking and correcting typos and omissions in the email content using a natural language processing system, a means for displaying the corrected state in a preview, a means for checking and correcting harassment-related expressions and expressions that take into consideration the sustainable development goals, a means for learning feedback from users and reflecting the results in checking and correcting typos and omissions from the next time onwards, a means for operating as an application to be installed on a smartphone, a means for making appropriate correction suggestions using a generative AI model, and a means for generating prompt sentences and obtaining appropriate correction suggestions. This makes it possible to ensure that the content of the email is appropriate and prevent misunderstandings and problems from occurring.

[0427] "Checking the destination domain before sending email" means checking the domain part of the destination email address to confirm that it is a valid domain before sending email.

[0428] "Checking for the presence or absence of a title" means checking whether the title of an email is entered properly before the email is sent.

[0429] A "natural language processing system" is a technology that enables computers to understand and process human language, and is a system that checks for and corrects typos and omissions.

[0430] "Checking for and correcting typos and omissions" means checking the contents of the email for incorrect or missing characters and correcting them if necessary.

[0431] "Show revised state in preview" means displaying the contents of the email after typos, omissions, and other corrections have been made so that the user can check them.

[0432] "Checking for and correcting harassing language" means checking whether the content of an email contains any language that could be considered harassing, and correcting it if necessary.

[0433] "Checking and correcting expressions that take into consideration the Sustainable Development Goals" means checking whether the content of the email contains expressions that go against the Sustainable Development Goals (SDGs) and correcting them if necessary.

[0434] "Learning feedback from users" means that the system learns the opinions and evaluations provided by users to the system and reflects them in subsequent processing.

[0435] An "application installed on a smartphone" is a software program that runs on a smartphone and is provided in a form that is usable by the user.

[0436] A "generative AI model" is a model that uses artificial intelligence technology to generate appropriate outputs for specific tasks.

[0437] "Generating a prompt sentence and obtaining appropriate correction suggestions" means that the system generates a sentence to make correction suggestions based on the content of the input email, and uses that sentence to obtain appropriate correction suggestions.

[0438] In order to implement the present invention, the following system configuration and processing procedure are used.

[0439] System Configuration

[0440] 1. Hardware:

[0441] Server: A server with high computing power is required. It is used to analyze the content of the email and generate correction suggestions.

[0442] Device: A mobile device such as a smartphone or tablet is required. The user will use this device to send the email.

[0443] 2. Software:

[0444] Natural Language Processing System: A system to check and correct typos, omissions, harassing expressions, and expressions that are considerate of the Sustainable Development Goals, using the Hugging Face Transformers library.

[0445] Generative AI model: A model to provide appropriate correction suggestions. For example, we use the distilbert-base-uncased-finetuned-sst-2-english model.

[0446] Application: An application installed on a smartphone that users use to send emails. It transmits the contents of the email to the server and receives suggestions for corrections.

[0447] Processing flow

[0448] 1. Before sending an email, check:

[0449] The server receives the contents of the email sent from the terminal.

[0450] The server checks the destination domain and whether or not the title is included.

[0451] 2. Analysis using natural language processing:

[0452] The server analyzes the content of the email using a natural language processing system.

[0453] Check for typos, omissions, harassment, and Sustainable Development Goal-friendly language, and generate suggested revisions as needed.

[0454] 3. Generate correction suggestions:

[0455] The server uses a generative AI model to make appropriate correction suggestions.

[0456] Generate prompt sentences and get appropriate correction suggestions.

[0457] 4. Preview after correction:

[0458] The contents of the email, including suggested corrections, are sent to the device and a preview is displayed to the user.

[0459] 5. Learning from user feedback:

[0460] Users provide feedback on the proposed revisions.

[0461] The server learns from this feedback and uses it to check for and correct typos and omissions from the next time onwards.

[0462] Examples

[0463] For example, consider the case where a user tries to send an email like this:

[0464] Example of email content

[0465] This is a sample email with bad word1.

[0466] The server analyzes the email content and generates correction suggestions such as the following:

[0467] Examples of suggested revisions

[0468] This is a sample email with [Suggested correction: Change to appropriate wording].

[0469] The system will display a preview of the proposed corrections to the user, and once the user has confirmed the corrections, will send an email. By providing feedback on the proposed corrections, the system can improve the accuracy of corrections in future.

[0470] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0471] Step 1:

[0472] The server receives the contents of the email sent from the terminal.

[0473] Input: The email content that the user creates on their device.

[0474] Output: The email content is sent to the server.

[0475] Specific operation: When a user composes an email on the terminal and presses the send button, the contents of the email are sent to the server.

[0476] Step 2:

[0477] The server checks the destination domain and whether or not the title is included.

[0478] Input: The email content sent from the device.

[0479] Output: The results of the domain and title check.

[0480] What happens: The server extracts the domain part of the email address and verifies that it is a valid domain. It also checks that the subject of the email is entered correctly.

[0481] Step 3:

[0482] The server analyzes the content of the email using a natural language processing system.

[0483] Input: The email content sent from the device.

[0484] Output: Check results for typos, omissions, harassment, and expressions that take into consideration the Sustainable Development Goals.

[0485] What it does: The server uses a natural language processing system (Hugging Face's Transformers library) to analyze the content of the email and check for typos, omissions, harassment, and language that takes the Sustainable Development Goals into consideration.

[0486] Step 4:

[0487] The server uses a generative AI model to make appropriate correction suggestions.

[0488] Input: Analysis results from a natural language processing system.

[0489] Output: suggested fixes.

[0490] What happens: The server uses a generative AI model (distilbert-base-uncased-finetuned-sst-2-english) to generate appropriate correction suggestions based on the analysis results.

[0491] Step 5:

[0492] The server generates the prompt sentence and obtains appropriate correction suggestions.

[0493] Input: suggested fix.

[0494] Output: The prompt text and suggested corrections.

[0495] Specific behavior: The server generates a prompt based on the correction suggestions, and uses the prompt to get the appropriate correction suggestions.

[0496] Step 6:

[0497] The contents of the email, including suggested corrections, are sent to the device and a preview is displayed to the user.

[0498] Input: The prompt text and suggested corrections.

[0499] Output: A preview of the email content after corrections.

[0500] What it does: The server sends the email content, including suggested edits, to the device, allowing the user to preview the revised email content.

[0501] Step 7:

[0502] Users provide feedback on the proposed revisions.

[0503] Input: Preview of the email content after revision.

[0504] Output: User feedback.

[0505] What happens: The user reviews the revised email and provides feedback on the suggested changes.

[0506] Step 8:

[0507] The server learns from this feedback and uses it to check for and correct typos and omissions from the next time onwards.

[0508] Input: User feedback.

[0509] Output: The training results.

[0510] What it does: The server learns from user feedback and updates the model to reflect it in future checks and corrections of typos and omissions.

[0511] Furthermore, an emotion engine that estimates the emotion of the user may be combined. That is, the identification processing unit 290 may estimate the emotion of the user using the emotion identification model 59, and perform identification processing using the emotion of the user.

[0512] "Example 1"

[0513] One embodiment of the present invention is an email system incorporating an emotion engine. In this system, when a user composes an email, a natural language processing system and the emotion engine work together. Specifically, when the user enters the body of the email, the natural language processing system analyzes the content and checks for typos and missing characters. At the same time, the emotion engine recognizes the user's emotion and corrects the content of the email based on that emotion.

[0514] "Example 2"

[0515] The emotion engine can also control the sending of emails based on the user's emotions. For example, if the user feels negative emotions such as anger or sadness, the emotion engine can pause sending the email and prompt the user to reconsider. This can prevent the wrong emails from being sent based on emotional decisions.

[0516] "Example 3"

[0517] In addition, the emotion engine can automatically adjust the expression of the email according to the user's emotions. For example, if the user feels that he or she has positive emotions such as joy or excitement, the emotion engine automatically modifies the expression of the email to be brighter and more lively, which can better convey the user's emotions.

[0518] "Example 4"

[0519] The emotion engine can also learn from user feedback and reflect the results in future emotion recognition and email corrections. For example, if a user gives feedback such as "This email does not accurately reflect my emotions," the emotion engine can learn from that feedback and from the next time onwards, it will be able to more accurately recognize emotions and correct emails.

[0520] The process flow of each embodiment will be described below.

[0521] "Example 1"

[0522] Step 1: The user enters the body of the email.

[0523] Step 2: A natural language processing system analyzes the content of the email and checks for typos and spelling errors.

[0524] Step 3: The emotion engine recognizes the user's emotion.

[0525] Step 4: The emotion engine modifies the email content based on the emotion.

[0526] "Example 2"

[0527] Step 1: The emotion engine recognizes the user's emotion.

[0528] Step 2: If the sentiment engine determines that the user's sentiment is negative, pause sending emails.

[0529] Step 3: The emotion engine prompts the user to reconsider the email.

[0530] "Example 3"

[0531] Step 1: The emotion engine recognizes the user's emotion.

[0532] Step 2: If the sentiment engine determines that the user's sentiment is positive, it automatically modifies the email's tone to make it more upbeat and vibrant.

[0533] "Example 4"

[0534] Step 1: The user provides feedback to the emotion engine.

[0535] Step 2: The emotion engine learns from the feedback.

[0536] Step 3: The emotion engine uses the feedback to improve emotion recognition and email revisions for the next time.

[0537] Example 1

[0538] Next, a description will be given of Example 1 of the embodiment 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".

[0539] In conventional email sending systems, confirmation of the recipient's domain and the presence or absence of a title are often done manually, which can easily lead to problems due to sending by mistake or not entering a title. In addition, checking for typos and omissions in the content of the email is also done manually, which is inefficient and may leave errors. Furthermore, if the content of the email is emotional, sending it as is can cause misunderstandings and problems. A system that solves these problems is needed.

[0540] 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.

[0541] In this invention, the server includes a means for checking the destination domain before sending, a means for checking the presence or absence of a title, a means for checking the content of the email for typos and omissions using a natural language processing system, a means for recognizing the user's emotions using an emotion engine and proposing corrections to the content of the email based on the emotions, a means for displaying the corrected state in a preview, and a means for providing feedback to the user. This automatically checks the domain and title before sending an email, making it possible to prevent erroneous sending or leaving the title blank. Furthermore, the quality of the email content is improved by automatically checking for typos and omissions and proposing corrections, and the content correction suggestions by the emotion engine can prevent problems caused by sending emotional emails.

[0542] "Checking the destination domain" means analyzing the domain portion of the destination address of an email to determine whether it is an internal domain or an external domain.

[0543] "Check for title" means checking whether the subject field of the email is empty and prompting the user to enter a title if necessary.

[0544] A "natural language processing system" is a system that analyzes text data, detects typos, omitted words, and grammatical errors, and suggests corrections.

[0545] "Checking for typos and omissions" means detecting typos and omissions in the body of the email and making suggestions for correcting them.

[0546] The "emotion engine" is a system that analyzes user emotions from text data and provides appropriate feedback and suggested corrections based on those emotions.

[0547] "Show post-correction state in preview" means visually displaying to the user the contents of the email after applying suggested corrections for typos and omissions, or corrections suggested by the emotion engine.

[0548] "Providing feedback to the user" means notifying the user of the results of the system's domain check, title check, suggestions for correction of typos, and suggestions for correction based on the emotion engine, and encouraging them to take necessary action.

[0549] The present invention is an email sending system that performs the following functions automatically when a user creates an email: checking the destination domain, checking the existence of a title, checking for spelling errors using a natural language processing system, and recognizing emotions and suggesting corrections using an emotion engine. This system automatically performs these functions and provides feedback to the user when the user creates an email.

[0550] Hardware and Software Used

[0551] 1. Server

[0552] The server plays a central role in the mail transmission system. It receives input from users and performs various checks and analysis.

[0553] 2. Natural Language Processing Systems

[0554] The natural language processing system uses Google Cloud Natural Language API, which detects typos and omissions in the email text and makes correction suggestions.

[0555] 3. Emotion Engine

[0556] The emotion engine used is IBM Watson (registered trademark) Tone Analyzer, which analyzes the user's emotions from the email body and makes correction suggestions based on those emotions.

[0557] Data processing and calculation

[0558] 1. Verify the recipient's domain

[0559] The server analyzes the destination address entered by the user, extracts the domain part, and checks the extracted domain against the internal domain list to determine whether it is an internal domain or an external domain.

[0560] 2. Check if there is a title

[0561] The server checks if the email's subject field is empty. If it is, it displays a message prompting the user to enter a subject.

[0562] 3. Check for typos

[0563] The server sends the email text to a natural language processing system to detect typos and spelling errors, and generates correction suggestions for the detected typos and spelling errors and displays them to the user.

[0564] 4. Emotion recognition and correction suggestions

[0565] The server sends the email text to the emotion engine to analyze the user's emotions. Based on the analysis results, it generates suggestions to modify the emotional expressions and displays them to the user.

[0566] Examples

[0567] If a user enters a message with a destination address of "example@gmail.com", no subject line, and no body text, the server does the following:

[0568] 1. Verify the recipient's domain

[0569] The server receives "example@gmail.com", extracts the domain part "@gmail.com", checks it against the internal domain list, and determines that it is an external domain.

[0570] 2. Check if there is a title

[0571] The server sees that the title field is empty and displays the message "Please enter a title" to the user.

[0572] 3. Check for typos

[0573] The server sends the text to the Google Cloud Natural Language API to check for typos and spelling errors. For example, it checks that there are no typos in the text "Hello, this is a test email."

[0574] 4. Emotion recognition and correction suggestions

[0575] The server sends the text to IBM Watson Tone Analyzer to analyze the sentiment. For example, in response to the text, "This project is not progressing at all. I'm very frustrated," the server generates a correction suggestion, "This project seems to be progressing slowly. Let's think of ways to improve it," and displays it to the user.

[0576] Examples of prompt statements

[0577] "Please describe a system that uses a natural language processing system and an emotion engine to analyze the text entered by the user, check for typos and omissions, recognize emotions, and then correct the content of the email based on those emotions."

[0578] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0579] Step 1:

[0580] A user composes an email.

[0581] Input: The user enters the destination address, subject, and message body using an email client.

[0582] What happens: The user opens an email client (e.g. Outlook, Gmail) and composes a new email, entering "example@gmail.com" in the To field, leaving the subject blank, and typing "Hello, this is a test email" in the body.

[0583] Output: The entered destination address, subject, and body are sent to the server.

[0584] Step 2:

[0585] The server checks the destination domain.

[0586] Input: The destination address entered by the user.

[0587] Specific operation: The server receives the destination address "example@gmail.com" and extracts the domain part "@gmail.com".

[0588] Data processing: Check the extracted domains against an internal domain list.

[0589] Output: The result of determining whether the destination is an internal domain or an external domain.

[0590] Step 3:

[0591] The server checks whether the title exists.

[0592] Input: A title field entered by the user.

[0593] Specific Behavior: The server verifies that the title field is empty.

[0594] Data Operation: Check if the title field is empty.

[0595] Output: If title is empty, show the user the message "Please enter a title".

[0596] Step 4:

[0597] The server uses a natural language processing system to check the email content for typos and missing characters.

[0598] Input: The email body entered by the user.

[0599] Specific operation: The server sends the text “Hello, this is a test email” to a natural language processing system (e.g., Google Cloud Natural Language API).

[0600] Data processing: A natural language processing system analyzes the text and detects typos and omissions.

[0601] Output: Detected typos and omissions and suggested corrections.

[0602] Step 5:

[0603] The server uses an emotion engine to recognize the user's emotions and makes suggestions to modify the content of the email based on those emotions.

[0604] Input: The email body entered by the user.

[0605] Specific operation: The server sends the text to an emotion engine (e.g. IBM Watson Tone Analyzer).

[0606] Data processing: The emotion engine analyzes the text and recognizes the user's emotions.

[0607] Output: Emotion recognition results and correction suggestions. For example, for the text "This project is not progressing at all. I'm very frustrated," we generate a correction suggestion such as "This project seems to be progressing slowly. Let's think of ways to improve it."

[0608] Step 6:

[0609] The server will display the revised state in a preview.

[0610] Input: Typo suggestions and sentiment engine suggestions.

[0611] Specific operation: The server generates the email content after applying the proposed modifications.

[0612] Data calculation: Apply the suggested corrections and generate the corrected email content.

[0613] Output: A preview of the revised email content is displayed to the user.

[0614] Step 7:

[0615] The server provides feedback to the user.

[0616] Input: Domain check result, Title check result, Suggested corrections for typos, Sentiment engine suggestions.

[0617] Specific behavior: The server notifies the user of these results and prompts them to take any necessary action.

[0618] Data Calculation: Integrates each check result and correction suggestions to generate a feedback message for the user.

[0619] Output: A feedback message to the user, for example "Title required for submission to external domain. Misspellings detected. Contains emotional content."

[0620] (Application example 1)

[0621] Next, a description will be given of Application Example 1 of Form 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."

[0622] In conventional email sending systems, there was insufficient checking of the recipient's domain and whether or not there was a title, which led to problems such as erroneous sending and not entering a title. In addition, it was difficult for customer support robots to properly recognize the customer's emotions and generate responses based on them, which could lead to a decrease in customer satisfaction. Furthermore, when sending customer inquiries by email, there was a lack of functionality to check the recipient's domain and whether or not there was a title.

[0623] 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. In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an e-mail, a means for checking and correcting typos and omissions in the e-mail content by a natural language processing system, a means for displaying the corrected state in a preview, and a means for analyzing the customer's input and generating an appropriate response in a customer service robot, a means for recognizing the customer's emotions and correcting the response based on the emotions, and a means for checking the destination domain and the presence or absence of a title when sending the customer's inquiry content by e-mail. This makes it possible to prevent erroneous sending of e-mails and failure to enter a title, and enables the customer service robot to generate an appropriate response according to the customer's emotions, thereby improving customer satisfaction.

[0624] "Checking the destination domain before sending email" means automatically checking whether the domain part of the destination email address is an internal domain or an external domain before sending the email.

[0625] "Checking whether or not there is a title" means checking whether or not the title of the email has been entered before sending the email.

[0626] A "natural language processing system" is a system that analyzes text data and checks for and corrects typos and omissions.

[0627] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in text and correcting them to the correct form.

[0628] "Display the corrected state in the preview" means visually displaying to the user the contents of the email after typos and omissions have been corrected.

[0629] A "customer service robot" is a robot that responds to customer inquiries and requests in a physical store.

[0630] "Analyzing customer input and generating an appropriate response" means analyzing what the customer says to the robot and generating an appropriate reply based on that content.

[0631] "Recognizing the customer's emotions and modifying responses based on those emotions" means reading the customer's emotions from the way they speak and the content of their messages, and providing an appropriate response based on those emotions.

[0632] "When sending a customer inquiry by email, check the destination domain and whether or not a title is entered" means that when sending a customer inquiry by email to store staff, check whether the destination domain is appropriate and whether a title has been entered.

[0633] As an embodiment of the present invention, a specific implementation method of a system in a customer service robot will be described below.

[0634] First, the customer-facing robot uses hardware with a voice recognition microphone, speaker and internet connection, and software using Python, TextBlob, spaCy and smtplib.

[0635] When a customer speaks to the robot, the robot converts the voice into text data through a voice recognition microphone. This text data is analyzed using a natural language processing system (spaCy). The analyzed text data undergoes sentiment analysis using TextBlob. Based on the result of the sentiment analysis, the robot generates an appropriate response.

[0636] The generated response is communicated to the customer through the speaker. At the same time, the customer's inquiry is sent to the store staff via email. Before sending the email, the destination domain and the presence or absence of a title are checked. This makes it possible to prevent sending to the wrong address or leaving the title blank.

[0637] As a concrete example, if a customer asks the robot, "Is the product in stock?", the robot responds, "We will check the stock. Please wait a moment," and sends the inquiry to the store staff by email. At this time, the robot checks whether the destination domain is an internal domain or an external domain, and checks whether a title has been entered.

[0638] An example of a prompt would be:

[0639] "If a customer asks, 'Is the item in stock?' the robot should respond with, 'We'll check the stock. Please wait a moment,' and then email the inquiry to a store staff member."

[0640] In this way, the customer service robot can generate appropriate responses according to the customer's emotions, improving customer satisfaction. It can also prevent emails from being sent by mistake or without a title.

[0641] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0642] Step 1:

[0643] The customer speaks to the robot. The customer's voice input is converted into text data through a voice recognition microphone. The input is the customer's voice, and the output is text data.

[0644] Step 2:

[0645] The server sends the text data to a natural language processing system (spaCy) for analysis. The input is text data, and the output is the analyzed text data. Specifically, the system analyzes the grammatical structure and meaning of the text.

[0646] Step 3:

[0647] The server sends the analyzed text data to the sentiment analysis system (TextBlob) for sentiment analysis. The input is the analyzed text data, and the output is the result of sentiment analysis. The specific operation is to classify the text sentiment into positive, negative, and neutral.

[0648] Step 4:

[0649] The server generates an appropriate response based on the results of the sentiment analysis. The input is the result of the sentiment analysis, and the output is the generated response text. The specific behavior is to generate a positive response for positive sentiment, and a response containing an apology or solution for negative sentiment.

[0650] Step 5:

[0651] The server transmits the generated response text to the customer through the speaker. The input is the generated response text, and the output is the voice response. The specific operation is to convert the text into voice and play it back through the speaker.

[0652] Step 6:

[0653] The server prepares to send the customer's inquiry to the store staff via email. The input is the text of the customer's inquiry, and the output is a draft of the email. Specifically, the server generates the body of the email and sets the destination email address and subject.

[0654] Step 7:

[0655] Before sending an email, the server checks the destination domain and whether or not there is a title. The input is the email draft, and the output is the confirmation result. Specifically, it checks whether the destination domain is an internal or external domain, and whether a title has been entered.

[0656] Step 8:

[0657] If the verification result is OK, the server sends an email to the store staff. The input is a draft of the verified email, and the output is the sent email. The specific operation is to connect to the email server and send the email.

[0658] In this way, the customer service robot can generate appropriate responses according to the customer's emotions, improving customer satisfaction. It can also prevent emails from being sent by mistake or without a title.

[0659] Example 2

[0660] Next, a second embodiment of the second embodiment will be described. 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."

[0661] In conventional email sending systems, checking and correcting typos and omissions is often done manually, which is time-consuming for users. In addition, there is no way to prevent emotional emails from being sent, which can lead to problems caused by sending incorrect emails. Furthermore, there is a risk that users will unintentionally use inappropriate language because the checking and correction of harassing expressions and expressions that take into consideration the Sustainable Development Goals is not automated. A system that can solve these problems is needed.

[0662] 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.

[0663] In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omissions in the email content using a natural language processing system, a means for displaying the corrected state in a preview, and a means for analyzing the user's emotions and pausing the sending of the email and encouraging the user to reconsider if negative emotions are detected. This not only enables the user to send accurate emails without typos and omissions, but also prevents the user from sending erroneous emails based on emotional judgment. In addition, the checking and correction of harassment-related expressions and expressions that take into consideration the Sustainable Development Goals are also automated, reducing the risk of the user unintentionally using inappropriate expressions.

[0664] "Verifying the destination domain" means verifying whether the domain part of the destination address of the email is correct.

[0665] "Checking whether or not there is a title" means checking whether or not there is a title set in the email.

[0666] A "natural language processing system" is a system that analyzes text data and automatically determines typos, omissions, emotions, and the appropriateness of expressions.

[0667] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in the body of an email and correcting them to the correct form.

[0668] "Display in preview" means to display the corrected email body on the screen so that the user can visually confirm it.

[0669] "Analyzing emotions" means reading the user's emotions from the email body and determining whether those emotions are positive or negative.

[0670] "If negative emotions are detected, the sending of the email will be paused and the user will be prompted to reconsider" means that if it is determined that the user has negative emotions such as anger or sadness, the sending of the email will be temporarily stopped and the user will be prompted to reconsider sending it.

[0671] "Harassing speech" is any offensive, discriminatory or inappropriate speech that may make others feel uncomfortable.

[0672] "Expression that takes into consideration the Sustainable Development Goals" means expression that does not contradict the Sustainable Development Goals, such as environmental protection, social justice, and economic development.

[0673] "Learning feedback from users" means that the system records the opinions and corrections provided by users and reflects them in future analyses and corrections.

[0674] The present invention is directed to a mail sending system that allows users to send accurate and appropriate mail by checking for and correcting typos and omissions and analyzing emotions. A specific embodiment of this system will be described below.

[0675] Hardware and Software Used

[0676] The system uses the following hardware and software:

[0677] Server: A central processing unit for analyzing and correcting email text and for sentiment analysis. Specifically, a cloud-based server is used.

[0678] Terminal: A device on which a user types email and previews the email after editing. Specifically, this applies to PCs, smartphones, tablets, etc.

[0679] Natural language processing system: Software for checking and correcting typos and omissions in email text. Specifically, it uses the Google Cloud Natural Language API.

[0680] Sentiment engine: Software to analyze user emotions and pause sending emails if negative emotions are detected, specifically using IBM Watson Tone Analyzer.

[0681] Data processing and calculation

[0682] 1. The user enters the body of the email. For example, the user enters "Hello, it's a nice day today."

[0683] 2. The terminal sends the entered email body to the server. The terminal sends the entered text data to the server as an HTTP request.

[0684] 3. The server receives the email body and calls the natural language processing system. The server sends the received text data to the Google Cloud Natural Language API.

[0685] 4. The server analyzes the email body using a natural language processing system and receives the analysis results returned from the API.

[0686] 5. The server detects and corrects typos and omissions based on the analysis results. For example, it corrects "konnichiwa" to "konnichiwa" and "kyo" to "kyo".

[0687] 6. The server sends the corrected email body to the terminal. The server sends the corrected text data to the terminal as an HTTP response.

[0688] 7. The terminal displays a preview of the corrected email body to the user. The terminal displays the received corrected text data on the screen.

[0689] Additionally, a process of sentiment analysis is also included.

[0690] 1. The user types the body of the email, for example, "I don't want to talk to you anymore!"

[0691] 2. The terminal sends the entered email body to the server. The terminal sends the entered text data to the server as an HTTP request.

[0692] 3. The server receives the email body and calls the emotion engine. The server sends the received text data to the IBM Watson Tone Analyzer.

[0693] 4. The server analyzes the user's emotions using the emotion engine. The server receives the emotion analysis results returned from the API.

[0694] 5. The server determines that the user's emotion is negative based on the analysis results. For example, it checks the emotion scores of "anger" or "sadness."

[0695] 6. The server suspends sending of the email and sends a message to the terminal prompting the user to reconsider. The server sends the message "Do you want to send this email? Please reconsider" to the terminal as an HTTP response.

[0696] 7. The terminal displays a message to the user urging him or her to reconsider. The terminal displays the received message on the screen.

[0697] 8. The user reconsiders and confirms sending the email again. The user can either reconsider and press the send button again, or cancel the sending.

[0698] Examples of prompt statements

[0699] Example of a prompt to be input to a generative AI model:

[0700] "Please correct the typo or omission in the following email body: 'Hello, it's a lovely day today.'"

[0701] "Analyze the sentiment of the following email body text and note the negative sentiment: 'I don't want to talk to you anymore!'"

[0702] In this way, the system analyzes the email text entered by the user, corrects typos and omissions, and analyzes emotions, thereby supporting the sending of more accurate and appropriate emails.

[0703] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0704] Step 1:

[0705] The user enters the body of the email.

[0706] Input: The email body entered by the user (e.g. "Hello, it's a nice day today.")

[0707] Output: The entered email text is displayed on the terminal.

[0708] Specific operation: The user enters text into the email composition screen of the device.

[0709] Step 2:

[0710] The terminal sends the entered email text to the server.

[0711] Input: The email body entered by the user

[0712] Output: The email body is sent to the server.

[0713] Specific operation: The terminal sends the entered text data to the server as an HTTP request.

[0714] Step 3:

[0715] The server receives the email body and calls the natural language processing system.

[0716] Input: Email body sent from the device

[0717] Output: Text data sent to a natural language processing system

[0718] Specific operation: The server sends the received text data to the Google Cloud Natural Language API.

[0719] Step 4:

[0720] The server analyzes the email body using a natural language processing system.

[0721] Input: Email body sent to natural language processing system

[0722] Output: Analysis results (information on typos and omissions)

[0723] Specific operation: The server receives the analysis results returned from the API.

[0724] Step 5:

[0725] The server detects and corrects typos and omissions based on the analysis results.

[0726] Input: Analysis results (information on typos and omissions)

[0727] Output: Email body after correction

[0728] Specific operation: The server corrects the typo "konnichiwa" to "konnichiwa" and the omitted typo "kyo" to "kyo".

[0729] Step 6:

[0730] The server sends the corrected email body to the terminal.

[0731] Input: Email body after correction

[0732] Output: The revised email body sent to the device

[0733] Specific operation: The server sends the corrected text data to the terminal as an HTTP response.

[0734] Step 7:

[0735] The terminal displays a preview of the corrected email body to the user.

[0736] Input: Email body after correction

[0737] Output: The revised email body as seen by the user

[0738] Specific operation: The terminal displays the received corrected text data on the screen.

[0739] Step 8:

[0740] The server receives the email body and invokes the emotion engine.

[0741] Input: Email body sent from the device

[0742] Output: Text data sent to the emotion engine

[0743] Specific operation: The server sends the received text data to the IBM Watson Tone Analyzer.

[0744] Step 9:

[0745] The server uses an emotion engine to analyze the user's emotions.

[0746] Input: Email body sent to the sentiment engine

[0747] Output: Sentiment analysis results (sentiment score)

[0748] Specific operation: The server receives the sentiment analysis results returned from the API.

[0749] Step 10:

[0750] Based on the analysis results, the server determines that the user's emotions are negative.

[0751] Input: Sentiment analysis results (sentiment score)

[0752] Output: Flag if negative sentiment detected

[0753] Specific operation: The server checks the emotion scores of "anger" and "sadness" from the analysis results.

[0754] Step 11:

[0755] The server will suspend sending the email and send a message to the device urging you to reconsider.

[0756] Input: Flag if negative sentiment detected

[0757] Output: A message to reconsider

[0758] Specific operation: The server sends the message "Do you want to send this email? Please reconsider" to the terminal as an HTTP response.

[0759] Step 12:

[0760] The terminal displays a message to the user urging him or her to reconsider.

[0761] Input: A message encouraging you to reconsider

[0762] Output: A reminder message to the user

[0763] Specific operation: The terminal displays the received message on the screen.

[0764] Step 13:

[0765] The user reconsiders and confirms sending the email again.

[0766] Input: User's decision after reconsideration

[0767] Output: Action to send or unsend the email

[0768] Specific behavior: The user may reconsider and press the send button again, or cancel the send.

[0769] (Application example 2)

[0770] Next, a description will be given of Application Example 2 of Form 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."

[0771] Conventional email sending systems do not adequately check for and correct typos and do not control email sending based on the user's emotions, which means there is a high possibility of problems occurring due to erroneous sending or emotional email sending. In particular, sending emails with negative emotions is at high risk of causing misunderstandings and problems. This can lead to problems such as information leaks and deterioration of human relationships.

[0772] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0773] In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omitted words in the email content using a natural language processing system, a means for displaying the corrected state in a preview, and a means for analyzing the user's emotions, and for suspending the sending of the email and encouraging reconsideration if negative emotions are detected. This makes it possible to correct typos and omitted words and to control the sending of email based on emotions.

[0774] "Verifying the destination domain" means checking whether the domain part of the recipient's email address is correct before the email is sent.

[0775] "Checking for the presence or absence of a title" means checking whether or not the subject of the email has been entered before the email is sent.

[0776] A "natural language processing system" is a technology that enables computers to understand, analyze, and generate human language.

[0777] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in the body of an email and correcting them to the correct form.

[0778] "Show in preview" means that the revised email content is displayed on the screen so that the user can check it.

[0779] "Analyzing emotions" means inferring and analyzing the emotional state of a user from the email text they enter.

[0780] "Negative emotions" refers to negative feelings such as anger or sadness.

[0781] "Pause email sending" means temporarily halting email sending if negative sentiment is detected.

[0782] "Encourage reconsideration" means encouraging the user to reexamine the content of the email and make corrections if necessary.

[0783] A system for implementing this invention includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, and means for analyzing the user's emotions and, if negative emotions are detected, pausing the sending of the email and encouraging reconsideration.

[0784] Program processing explanation

[0785] The server performs data processing and calculation using the following hardware and software.

[0786] Hardware and Software Used

[0787] Natural Language Processing Model: We load a Japanese natural language processing model using the spacy library and check the email body for typos.

[0788] Emotion Engine: Analyzes the emotions in the email body using the EmotionRecognizer class.

[0789] Data processing: The email body is tokenized, and if there are any typos, suggestions for corrections are provided.

[0790] Data calculation: Based on the results of sentiment analysis, suspend sending of emails if negative sentiment is detected.

[0791] Example of processing

[0792] As the user types the text of the email, the natural language processing system checks for typos in real time and suggests corrections. The corrected email content is displayed in a preview for the user to review. In addition, the emotion engine analyzes the sentiment of the email text, and if negative sentiment is detected, the email is paused before being sent and the user is urged to reconsider.

[0793] Examples of prompt statements

[0794] If a user types an email saying, "This project isn't moving forward at all! What's going on?", the emotion engine will detect "anger" and pause the email from being sent. The user will see a message saying, "Negative emotion detected. Email sending will be paused."

[0795] In this way, it is possible to correct typos and control email sending based on emotions, thereby preventing problems caused by sending emails incorrectly or based on emotion.

[0796] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0797] Step 1:

[0798] The user enters the email body.

[0799] Input: The user enters the email body.

[0800] Specific behavior: A user enters the body of an email into a text area in an email client.

[0801] Step 2:

[0802] The terminal sends the email text to a natural language processing system to check for typos and misspellings.

[0803] Input: The email body entered by the user.

[0804] Data processing: A natural language processing system (spacy library) is used to tokenize the email body and detect typos.

[0805] Output: The email body with typos corrected.

[0806] Specific operation: The terminal sends the email text to a natural language processing system, which checks for and corrects typos. Correction suggestions are generated and presented to the user.

[0807] Step 3:

[0808] The device will display a preview of the revised email body.

[0809] Input: The email body with typos corrected.

[0810] Data processing: Convert the revised email body into a format for preview.

[0811] Output: The revised email body as displayed in the preview screen.

[0812] Specific operation: The device will display the revised email body on the preview screen so that the user can confirm it.

[0813] Step 4:

[0814] The device sends the email text to the emotion engine for emotion analysis.

[0815] Input: The revised email body.

[0816] Data calculation: Analyze the emotions in the email body using the emotion engine (EmotionRecognizer class).

[0817] Output: The result of the sentiment analysis (e.g. anger, sadness, etc.).

[0818] Specific operation: The device sends the corrected email text to the emotion engine for emotion analysis, and the analysis results are generated.

[0819] Step 5:

[0820] The device controls the sending of emails based on the results of sentiment analysis.

[0821] Input: Sentiment analysis results.

[0822] Data calculation: If negative emotions (e.g. anger, sadness) are detected, the sending of the email will be paused and the user will be prompted to reconsider.

[0823] Output: Email sending pause message.

[0824] Specific operation: The device checks the results of the emotion analysis, and if negative emotion is detected, it pauses sending of the email. The user is shown a message saying, "Negative emotion has been detected. Email sending will be paused."

[0825] Step 6:

[0826] The user reconsiders the content of the email and modifies it if necessary.

[0827] Input: Email sending pause message and revised email body.

[0828] Specific behavior: The user reviews the email, makes corrections if necessary, and then attempts to send the email again.

[0829] Step 7:

[0830] The terminal checks again to confirm the sending of the email, and if there are no problems, sends the email.

[0831] Input: The email body after reconsideration.

[0832] Data calculation: Check again for typos and perform sentiment analysis, and if there are no problems, send the email.

[0833] Output: The email sent.

[0834] Specific operation: The device checks the email body again and sends the email if there are no problems.

[0835] Example 3

[0836] Next, a third embodiment of the third embodiment will be described. 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."

[0837] Conventional email systems lacked the functionality to check the recipient's identifier or the presence or absence of a subject before sending, which led to occasional erroneous sending or missing subjects. Also, checking and correcting typos and omissions in document content was done manually, which was inefficient and placed a heavy burden on users. Furthermore, there was a lack of social consideration as there was no checking and correction of harassing expressions or expressions that take into consideration the Sustainable Development Goals. In addition, the quality of communication could decline as the expression in the document was not adjusted to reflect the user's feelings. A system that can solve these issues is needed.

[0838] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0839] In this invention, the server includes a means for confirming the identifier of the destination, a means for confirming the presence or absence of a subject, a means for confirming and correcting typos and omissions in the document content using a natural language processing system, a means for displaying the state after correction, and a means for analyzing the user's emotions and adjusting the expression of the document. This makes it possible to confirm the presence or absence of the identifier of the destination and the subject before sending, and to automatically correct typos and omissions. In addition, harassment-related expressions and expressions that take into consideration the sustainable development goals are confirmed and corrected, and the expression of the document can be adjusted to reflect the user's emotions. This allows for efficient and socially considerate communication.

[0840] The "destination identifier" is information for identifying the recipient of the email.

[0841] The "subject" is a title that briefly indicates the content of the email.

[0842] "Document content" is text information included in the body of the email.

[0843] A "natural language processing system" is a technology for analyzing text data and understanding its meaning.

[0844] A "typo" is a character string that is entered incorrectly instead of a correct character string.

[0845] An "omission" is when a necessary character is missing.

[0846] A "correction measure" is a method or device for correcting an error.

[0847] A "display means" is a method or device for visually presenting information.

[0848] "User emotion" refers to the psychological state of the user when composing an email.

[0849] The "means for analyzing emotions and adjusting the expression of a document" refers to a method or device for understanding the emotions of a user and changing the expression of an email accordingly.

[0850] "Harassing language" is any word or phrase that causes discomfort or disgust to others.

[0851] The "Sustainable Development Goals" are internationally agreed upon goals aimed at achieving sustainable development.

[0852] This invention is a system that has the functions of checking the recipient's identifier before sending an email, checking whether there is a subject, checking and correcting typos and omissions in the document content using a natural language processing system, displaying the state after correction, and analyzing the user's emotions to adjust the wording of the document.

[0853] Hardware and Software Used

[0854] The server uses the following hardware and software:

[0855] Hardware: Powerful processor, ample memory, and storage

[0856] Software: Python, NLTK library, Hugging Face Transformers library, scikit-learn, TENSORFLOW, TextBlob

[0857] A terminal is a device that allows a user to compose email and communicate with a server. This includes computers, smartphones, tablets, etc.

[0858] Program Processing

[0859] When a user composes an email and presses the send button, the device sends the email contents to the server. The server first checks whether the recipient identifier and the subject are included. It then uses a natural language processing system to analyze the document contents and detect and correct typos and omissions. This analysis uses Python's NLTK library and Hugging Face's Transformers library.

[0860] In addition, the server checks the document content for harassment and language that is sensitive to the Sustainable Development Goals, and makes corrections as necessary. The corrected document is then sent to the terminal and a preview is displayed to the user.

[0861] When a user provides feedback on the system's suggested corrections, the device sends the feedback to the server, which uses scikit-learn and TensorFlow to learn from the feedback and use it to check and correct typos and omissions in the future.

[0862] The server also uses an emotion engine to analyze the user's emotions. For this analysis, it uses an emotion analysis library (e.g., VADER or TextBlob). If the user is analyzed as having a positive emotion, the server automatically modifies the email's expression to be brighter and more lively. The modified document is sent to the terminal and previewed for the user.

[0863] When a user provides feedback on the emotion engine's suggestions, the device sends the feedback to the server, which uses TensorFlow to learn from the feedback and use it in future emotion recognition and document corrections.

[0864] Examples

[0865] For example, if a user writes "This project is completely pointless" in an email, the server will amend this to "This project has room for improvement" and display it in the preview. If a user writes "I think this project will be a success," the server will amend it to "I think this project will be a huge success!"

[0866] Examples of prompt statements

[0867] "Please check whether the content of this email contains any harassing language or language that takes into consideration the SDGs, and revise it as necessary."

[0868] "Please check this email for typos and correct them. Also, please learn from your user feedback and incorporate it into your next email."

[0869] “Adjust the content of this email according to the user’s emotions. If the user has positive emotions, make the message more upbeat.”

[0870] "Please check whether the content of this email accurately reflects the user's feelings and revise it if necessary. Also, please learn from the user's feedback and reflect it in future emails." The flow of the identification process in the third embodiment will be described with reference to FIG. 21.

[0871] Step 1:

[0872] The user composes an email and clicks the send button.

[0873] Input: The email content entered by the user

[0874] Output: The contents of the email are saved to the device.

[0875] Specific behavior: The user enters the body of the email and clicks the send button.

[0876] Step 2:

[0877] The terminal sends the contents of the email to the server.

[0878] Input: The content of the email the user creates

[0879] Output: The email content is sent to the server.

[0880] Specific operation: The terminal initiates communication to send the contents of the email to the server.

[0881] Step 3:

[0882] The server checks the presence of a destination identifier and a subject.

[0883] Input: The content of the email sent

[0884] Output: Destination identifier and subject existence check result

[0885] Specific operation: The server analyzes the header information of the email and checks whether it contains a destination identifier and a subject.

[0886] Step 4:

[0887] The server uses a natural language processing system to analyze the document content and detect and correct typos and omissions.

[0888] Input: Email body

[0889] Output: The modified email body

[0890] What it does: The server uses Python's NLTK library and Hugging Face's Transformers library to parse the body of the email, detect and correct typos and spelling errors.

[0891] Step 5:

[0892] The server will check the document content for harassment and language that takes into consideration the Sustainable Development Goals, and will make corrections if necessary.

[0893] Input: Modified email body

[0894] Output: Further modified email body

[0895] Specific actions: The server re-analyzes the body of the email to detect harassing language or language that is sensitive to the Sustainable Development Goals, and corrects it as necessary.

[0896] Step 6:

[0897] The server sends the corrected email to the terminal.

[0898] Input: Modified email body

[0899] Output: The corrected email body is sent to the device.

[0900] Specific operation: The server initiates communication to send the corrected email body to the terminal.

[0901] Step 7:

[0902] The terminal displays a preview of the corrected email to the user.

[0903] Input: The revised text of the email

[0904] Output: The email body as previewed to the user

[0905] Specific operation: The device displays the revised body of the email to the user.

[0906] Step 8:

[0907] The user provides feedback on the system's suggested modifications.

[0908] Input: User feedback

[0909] Output: Feedback is saved to your device

[0910] Specific action: The user enters feedback on the corrections and clicks the submit button.

[0911] Step 9:

[0912] The device sends the feedback to the server.

[0913] Input: User feedback

[0914] Output: Feedback is sent to the server

[0915] Specific operation: The device initiates communication to send feedback to the server.

[0916] Step 10:

[0917] The server learns the feedback and uses it to check for and correct typos and omissions in future posts.

[0918] Input: User feedback

[0919] Output: The trained model

[0920] Specific operation: The server uses scikit-learn and TensorFlow to learn the feedback and reflect it in checking and correcting typos and omissions from the next time onwards.

[0921] Step 11:

[0922] The server uses an emotion engine to analyze the user's emotions.

[0923] Input: Email body

[0924] Output: User sentiment analysis results

[0925] Specific operation: The server uses a sentiment analysis library (e.g., VADER or TextBlob) to analyze the user's sentiment from the body of the email.

[0926] Step 12:

[0927] The server adjusts the wording of the email according to the user's emotions.

[0928] Input: User sentiment analysis results

[0929] Output: Adjusted email body

[0930] Specific operation: The server adjusts the wording of the email based on the results of the user's sentiment analysis.

[0931] Step 13:

[0932] The server sends the adjusted email to the terminal.

[0933] Input: The adjusted email body

[0934] Output: The adjusted email body is sent to the device.

[0935] Specific operation: The server initiates communication to send the adjusted email body to the terminal.

[0936] Step 14:

[0937] The device displays a preview of the adjusted email to the user.

[0938] Input: Adjusted email body

[0939] Output: The email body as previewed to the user

[0940] Specific operation: The device displays the adjusted email body to the user.

[0941] Step 15:

[0942] The user provides feedback on the emotion engine's suggestions.

[0943] Input: User feedback

[0944] Output: Feedback is saved to your device

[0945] Specific action: The user enters feedback on the emotion engine's suggestion and clicks the submit button.

[0946] Step 16:

[0947] The device sends the feedback to the server.

[0948] Input: User feedback

[0949] Output: Feedback is sent to the server

[0950] Specific operation: The device initiates communication to send feedback to the server.

[0951] Step 17:

[0952] The server learns from the feedback and uses it to improve emotion recognition and document correction in future tasks.

[0953] Input: User feedback

[0954] Output: The trained model

[0955] Specific operation: The server uses TensorFlow to learn the feedback and reflect it in future emotion recognition and document corrections.

[0956] (Application example 3)

[0957] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0958] Conventional email sending systems check and correct typos and omissions, but do not check and correct harassing expressions or expressions that take into consideration the Sustainable Development Goals, and do not adjust expressions according to the user's emotions. This makes it difficult to communicate appropriately and ethically when sending messages related to transactions or payments. In addition, the system lacks the ability to learn from user feedback and reflect it in checking and correcting typos and omissions and in emotion recognition for future messages, which results in low adaptability.

[0959] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0960] In this invention, the server includes: a means for checking the destination domain and checking whether the email has a title before sending the email; a means for checking and correcting typos and omissions in the email content using a natural language processing system; a means for displaying the corrected state in a preview; a means for checking and correcting harassing expressions and expressions that take the Sustainable Development Goals into consideration; a means for analyzing the user's emotions and adjusting the email expression according to the emotions; a means for learning feedback from the user and reflecting the results in checking and correcting typos and omissions from the next time onwards; a means for learning feedback from the user and reflecting the results in emotion recognition and email correction from the next time onwards; and a means for checking and correcting harassing expressions and expressions that take the Sustainable Development Goals into consideration when sending a message related to a transaction or payment, and adjusting the expression according to the user's emotions using an emotion engine. This enables appropriate and ethical communication when sending a message related to a transaction or payment.

[0961] "Verifying the destination domain" means checking whether the domain part of the recipient's email address is correct before the email is sent.

[0962] "Checking for the presence or absence of a title" means checking whether or not the title is set appropriately in the email before the email is sent.

[0963] A "natural language processing system" is a system that analyzes text data and checks for and corrects typos and omissions.

[0964] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in the email content and correcting them to the correct form.

[0965] "Displaying in preview" means displaying the revised email content on the screen so that the user can visually confirm it.

[0966] "Harassment-type expressions" refer to expressions that cause discomfort or disgust to others.

[0967] "Expression that takes into consideration the Sustainable Development Goals" refers to expression that takes into consideration environmental, social, and economic sustainability.

[0968] "Analyzing emotions" means reading and analyzing emotions from text entered by the user.

[0969] "Adjusting email expression according to emotion" means appropriately changing the content and tone of an email based on the user's emotions.

[0970] "Learning from feedback" means that the system incorporates opinions and evaluations provided by users and reflects them in future processing.

[0971] A "transactional or payment message" is a message that contains content related to a commercial transaction or the payment of money.

[0972] An "emotion engine" is an engine that analyzes a user's emotions and adjusts the content of the text based on those emotions.

[0973] A system for carrying out the present invention has the following configuration: The server includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for checking and correcting harassing expressions and expressions that take into consideration the Sustainable Development Goals, means for analyzing a user's emotions and adjusting the expression of the email in accordance with the emotions, means for learning feedback from the user and reflecting the results in checking and correcting typos and omissions from the next time onwards, means for learning feedback from the user and reflecting the results in emotion recognition and email correction from the next time onwards, and means for checking and correcting harassing expressions and expressions that take into consideration the Sustainable Development Goals when sending a message related to a transaction or payment, and adjusting the expression in accordance with the user's emotions using an emotion engine.

[0974] Hardware and software used

[0975] Hardware: Smartphones, tablets

[0976] Software: Natural language processing libraries (e.g. spaCy, NLTK), sentiment analysis libraries (e.g. TextBlob, VADER), machine learning models (e.g. BERT, GPT-3)

[0977] Data processing and calculation

[0978] 1. Get input data:

[0979] A user enters a message within an electronic payment application.

[0980] Gets the message content in text format.

[0981] 2. Natural Language Processing:

[0982] Use a natural language processing library to analyze the message content.

[0983] Check whether the content contains any harassing language or language that takes into consideration the Sustainable Development Goals.

[0984] 3. Revision of expression:

[0985] If any such expressions are found, they will be amended to suit the needs of the user.

[0986] The revised message is displayed in a preview.

[0987] 4. Applying the Emotion Engine:

[0988] Analyze user sentiment using a sentiment analysis library.

[0989] Adjust the wording of your message depending on the user's emotions.

[0990] 5. Learning feedback:

[0991] Get feedback from users.

[0992] The feedback is trained into a machine learning model and reflected in future analyses and corrections.

[0993] Examples

[0994] Example prompts to be input to the generative AI model

[0995] User typed message: "This payment is late, please act quickly."

[0996] Emotion engine analysis result: "Anger"

[0997] New message: "We have confirmed that this payment is late. We would appreciate your prompt response."

[0998] In this way, by analyzing the messages entered by users and correcting them to appropriate expressions, communication regarding transactions and payments can be made smoother and more ethical.

[0999] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1000] Step 1:

[1001] A user enters a message within an electronic payment application.

[1002] Input: The message text entered by the user.

[1003] Output: The message text is sent to the system.

[1004] Specific operation: A user uses a smartphone or tablet to enter a message regarding a transaction or payment. The entered message is sent in text form to the system.

[1005] Step 2:

[1006] The server checks the destination domain and whether or not the title exists.

[1007] Input: message text, destination email address.

[1008] Output: Domain and title check results.

[1009] Specific operation: The server analyzes the message text and the destination email address, checks whether the domain part is correct, and whether the subject is set appropriately. The result of the check is stored as internal data.

[1010] Step 3:

[1011] The server uses a natural language processing system to check for and correct typos and omissions.

[1012] Input: The message text.

[1013] Output: The modified message text.

[1014] Specific behavior: The server uses a natural language processing library (e.g. spaCy, NLTK) to analyze the message text and detect typos and spelling errors. Detected typos and spelling errors are corrected appropriately.

[1015] Step 4:

[1016] The server will display a preview of the revised message.

[1017] Input: The modified message text.

[1018] Output: The preview screen that the user sees.

[1019] Specific operation: The corrected message text is sent to the user's device and displayed on the preview screen. The user can check the corrected message.

[1020] Step 5:

[1021] The server will check for and edit harassing language and language that takes into consideration the Sustainable Development Goals.

[1022] Input: The modified message text.

[1023] Output: Further modified message text.

[1024] Specific operation: The server uses a natural language processing library to check whether the message text contains harassing language or language that takes into consideration the Sustainable Development Goals. If any such language is found, it is modified to an appropriate language.

[1025] Step 6:

[1026] The server uses an emotion engine to analyze the user's emotions and adjust the wording of the email according to the emotions.

[1027] Input: The modified message text.

[1028] Output: Message text adjusted according to the sentiment.

[1029] Specific operation: The server analyzes the user's sentiment using a sentiment analysis library (e.g. TextBlob, VADER), and adjusts the wording of the message based on the analysis result.

[1030] Step 7:

[1031] The server learns from user feedback and uses the results to check for and correct typos and omissions in future posts.

[1032] Input: User feedback.

[1033] Output: The learning results are reflected in the next and subsequent processing.

[1034] Specific operation: The server receives the feedback provided by the user and trains a machine learning model (e.g. BERT, GPT-3). The learning results are reflected in checking and correcting typos and omissions from the next time onwards.

[1035] Step 8:

[1036] The server learns from user feedback and uses the results to improve emotion recognition and email correction in future emails.

[1037] Input: User feedback.

[1038] Output: The learning results are reflected in the next and subsequent processing.

[1039] How it works: The server receives the feedback provided by the user and trains the machine learning model. The learning results are reflected in future emotion recognition and email corrections.

[1040] Step 9:

[1041] When the server sends messages related to transactions or payments, it checks and corrects any harassing language or language that takes into consideration the Sustainable Development Goals, and uses an emotion engine to adjust the language according to the user's emotions.

[1042] Input: Message text for the transaction or payment.

[1043] Output: The message text, modified appropriately.

[1044] How it works: The server analyzes messages related to transactions and payments, checks for and modifies harassing expressions and expressions that are considerate of the Sustainable Development Goals, and uses an emotion engine to adjust the expressions according to the user's emotions.

[1045] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1046] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to 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.

[1047] Another example of generative AI is Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1048] In the above embodiment, an example was given in which the specific process was performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[1049] [Second embodiment]

[1050] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[1051] 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.

[1052] 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 wide area network (WAN) and / or a local area network (LAN).

[1053] 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.

[1054] 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 the voice according to instructions from the processor 46.

[1055] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[1056] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[1057] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1058] 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.

[1059] 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.

[1060] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1061] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1062] "Example 1"

[1063] As one embodiment of the present invention, an email sending system is provided. This system has a function for checking the destination domain and whether or not there is a title when a user creates an email. Specifically, when a user inputs a destination of an email, the system automatically checks whether the destination domain is an internal domain or an external domain. Furthermore, if the title of the email has not been input, the system prompts the user to input a title.

[1064] "Example 2"

[1065] Furthermore, this system has the function of checking and correcting typos and omissions in the email content using a natural language processing system. Specifically, when a user enters the body of the email, the natural language processing system analyzes the content and checks for typos and omissions. If a typographical error or omission is found, the system corrects that part and displays the corrected state in a preview.

[1066] "Example 3"

[1067] The system also has the ability to check and correct harassment-type language and language that takes SDGs into consideration. Specifically, the natural language processing system checks the content of emails for harassment-type language and language that takes SDGs into consideration. If any such language is found, the system corrects the language and displays the corrected state in a preview.

[1068] "Example 4"

[1069] Furthermore, the system has the ability to learn from feedback from users and reflect the results in checking and correcting typos and omissions from the next time onwards. Specifically, when a user provides feedback on corrections suggested by the system, the natural language processing system learns from that feedback and reflects it in checking and correcting typos and omissions from the next time onwards. This allows the system to better adapt to the needs of the user.

[1070] The process flow of each embodiment will be described below.

[1071] "Example 1"

[1072] Step 1: The user enters the address to which the email should be sent.

[1073] Step 2: The system will automatically check whether the destination domain is an internal or external domain.

[1074] Step 3: If the email subject is not entered, the system will prompt the user to enter a subject.

[1075] "Example 2"

[1076] Step 1: The user enters the body of the email.

[1077] Step 2: The natural language processing system analyzes the content and checks for typos and omissions.

[1078] Step 3: If any typos or omissions are found, the system will correct them and display the corrected version in a preview.

[1079] "Example 3"

[1080] Step 1: The natural language processing system checks the content of the email for any harassing language or language that takes into consideration the SDGs.

[1081] Step 2: If a matching expression is found, the system corrects the expression and displays the corrected version in a preview.

[1082] "Example 4"

[1083] Step 1: The user provides feedback on the system's suggested fixes.

[1084] Step 2: The natural language processing system learns from the feedback and uses it to check for and correct typos and omissions in future sentences.

[1085] Example 1

[1086] Next, a description will be given of Example 1 of Form 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".

[1087] In conventional email sending systems, users are often required to manually check the destination domain and whether the email title is correct, which is time-consuming. In addition, there is a risk of email quality declining due to insufficient checking and correction of typos, omissions, harassing expressions, and expressions that take into consideration the Sustainable Development Goals. Furthermore, there is no mechanism for reflecting user feedback in subsequent checks and corrections, which means that improvements to the system are not progressing.

[1088] 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.

[1089] In this invention, the server includes a means for the user to input the destination of the email, a means for the server to check the domain of the destination, a means for the server to check whether the email has a title, a means for the server to prompt the user to input a title, a means for checking and correcting typos and omissions in the email content using a natural language processing system, and a means for displaying the corrected state in a preview. This makes it possible to reduce the user's work and improve the quality of emails. In addition, by learning from user feedback and reflecting it in subsequent checks and corrections, the system can be continuously improved.

[1090] "User" means any individual or entity that uses the System to create and send Email.

[1091] A "server" is a computer system that accepts and processes input from users.

[1092] A "destination domain" is the portion of an email address following the @ symbol, and indicates the network to which the email is being sent.

[1093] "Email title" is the text entered as the subject of the email.

[1094] A "natural language processing system" is software or algorithms that analyze text data and check for and correct typos, omissions, and specific expressions.

[1095] "Checking for and correcting typos and omissions" refers to the process of detecting spelling mistakes and input errors in the content of an email and correcting them to the correct expression.

[1096] "Harassing language" is any word or phrase that may cause discomfort or offense to others.

[1097] "SDG-friendly expressions" are words and phrases that are ethical and socially sensitive and are in line with the Sustainable Development Goals (SDGs).

[1098] "Feedback" refers to opinions and evaluations provided by users, and is information that is useful for improving the system.

[1099] "Preview" is a display screen where the user can check the content of the email after correction.

[1100] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[1101] The present invention relates to an email sending system that checks the destination domain and whether or not there is a title when a user creates an email. When a user enters the destination of an email, the system automatically checks whether the destination domain is an internal domain or an external domain. If the title of the email has not been entered, the server prompts the user to enter a title.

[1102] Hardware and Software Used

[1103] This system is realized using the following hardware and software.

[1104] Server: A computer system that handles the central processing of the mail delivery system. It uses a Python script to verify domains.

[1105] Database: A database system for storing the internal domain list, specifically MySQL.

[1106] Terminal: The device on which the user composes the email. JavaScript running on the browser is used to check whether the title is included.

[1107] System Operation

[1108] When a user enters the destination of an email, the server extracts the domain part of the destination email address and checks it against the internal domain list, which is stored in the server's database. Next, if the user has not entered the subject of the email, the server prompts the user to enter a subject. This function is implemented on the browser using JavaScript.

[1109] Examples

[1110] For example, if a user enters the email address "example@external.com", the server will check that "external.com" is not in the internal domain list. Then, if the user clicks the submit button without entering a title, JavaScript will display an alert saying "Please enter a title."

[1111] Example prompts for generative AI models

[1112] "Please generate a program for an email sending system. When a user enters the destination of an email, the server checks whether the destination domain is an internal domain or an external domain, and if the subject of the email has not been entered, it prompts the user to enter a subject."

[1113] The above is an embodiment of the present invention. This system makes it possible to reduce the work for users and improve the quality of emails. In addition, the system can continuously improve by learning from user feedback and reflecting it in subsequent checks and corrections.

[1114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1115] Program processing flow

[1116] Step 1: User enters destination for email

[1117] The user enters the destination email address on the email composition screen. For example, they enter "example@external.com." This input is the starting point for the system's processing.

[1118] Input: The email address entered by the user (e.g. example@external.com)

[1119] Output: The entered email address is sent to the server.

[1120] Step 2: The server checks the destination domain

[1121] The server extracts the domain part from the email address it receives -- specifically, the part of the email address that comes after the @ sign -- and then checks whether this domain is on its internal domain list, which is stored in the server's database.

[1122] Input: The email address entered by the user (e.g. example@external.com)

[1123] Data processing: Extraction of domain part (e.g. external.com)

[1124] Data calculation: Check against internal domain list

[1125] Output: Result of whether the domain is internal or external

[1126] Step 3: The server checks whether the email title exists

[1127] When the user clicks the send button, the server checks if the email subject field is empty. This check is done in the browser using JavaScript.

[1128] Input: Email title entered by the user (e.g. blank)

[1129] Data Calculation: Check if title field is empty

[1130] Output: Check whether a title has been entered.

[1131] Step 4: Server prompts user for title (if necessary)

[1132] If no subject is entered, the server will display an alert to the user and prompt them to enter a subject. The email will not be sent unless the user enters a subject.

[1133] Input: The result indicates that the title has not been entered.

[1134] Specific behavior: JavaScript displays an alert saying "Please enter a title"

[1135] Output: An alert prompting the user to enter a title.

[1136] Step 5: Check the email for typos and omissions using a natural language processing system and correct them.

[1137] The server passes the email content to a natural language processing system, which checks for and corrects typos and omissions. The corrected email content is then displayed to the user as a preview.

[1138] Input: Email content entered by the user

[1139] Data Calculation: Checking and correcting typos and omissions using natural language processing systems

[1140] Output: Email content after correction

[1141] Step 6: Preview the result after correction

[1142] The server then displays a preview of the revised email to the user, allowing the user to review the preview and make further edits if necessary.

[1143] Input: Email content after correction

[1144] Specific operation: The revised email content is displayed on the preview screen.

[1145] Output: A preview screen that the user can see

[1146] The above are the specific processing steps of the program of this system.

[1147] (Application example 1)

[1148] Next, a description will be given of Application Example 1 of Form 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".

[1149] In the conventional email sending system, there was insufficient checking of the destination domain and the presence or absence of a title, which often led to communication problems due to erroneous sending or not entering a title. In addition, there was a lack of functionality to notify users of security risks when sending to external domains, which increased the risk of information leakage. Furthermore, checking and correcting typos in email content was often done manually, which was inefficient.

[1150] 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.

[1151] In this invention, the server includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for notifying security risks when sending to an external domain, and means for prompting input of a title if one has not been entered. This makes it possible to prevent problems with erroneous sending of emails and not entering a title, reduce security risks when sending to an external domain, and efficiently check and correct typos and omissions in the email content.

[1152] "Checking the destination domain before sending email" means automatically checking whether the domain part of the destination email address is an internal domain or an external domain before sending the email.

[1153] "Checking whether or not there is a title" means checking whether or not the title of the email has been entered.

[1154] A "natural language processing system" is a system that analyzes text data and checks for and corrects typos, omissions, and specific expressions.

[1155] "Checking for and correcting typos and omissions" means detecting typos and omissions in the content of an email and correcting them.

[1156] "Display the corrected state in the preview" means visually displaying to the user the contents of the email after typos and omissions have been corrected.

[1157] "Notify of security risks when sending to external domains" means to inform the user of security risks when sending email to external domains.

[1158] "Prompt for input of title if not entered" means to prompt the user to input a title if the title of the email has not been entered.

[1159] A system for implementing this invention includes means for checking the destination domain before sending an email and for checking whether or not there is a title, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for notifying of security risks when sending to an external domain, and means for prompting input of a title if one has not been entered.

[1160] Hardware and Software Used

[1161] Hardware: Smartphones, servers

[1162] Software: Python, natural language processing library (e.g. spaCy), email sending library (e.g. smtplib)

[1163] System processing overview

[1164] 1. Domain Verification:

[1165] The server extracts the domain portion of the email address entered by the user and checks it against its internal domain list.

[1166] If it is an internal domain, it is identified as an "internal domain"; if it is an external domain, it is identified as an "external domain."

[1167] 2. Title Check:

[1168] The server checks whether the email subject entered by the user is empty.

[1169] If the title is not entered, a message is displayed prompting the user to enter a title.

[1170] 3. Checking and correcting typos using natural language processing:

[1171] The server uses a natural language processing system to analyze the email content and detect typos and missing characters.

[1172] Detected typos and omissions are automatically corrected and the corrected email content is generated.

[1173] 4. Preview display:

[1174] The revised email content is displayed as a preview to the user for confirmation.

[1175] 5. Security Risk Notification:

[1176] The server notifies the user that there is a security risk when sending to external domains.

[1177] The email will be sent after the user confirms.

[1178] Examples

[1179] For example, if a user attempts to send an email to "user@external.com", the server extracts the domain portion of this email address, "external.com", and compares it with the internal domain list. Because it is an external domain, the server notifies the user that there is a security risk. Also, if the email subject line has not been entered, a message is displayed prompting the user to enter one. Furthermore, if the email content contains typos or omissions, the natural language processing system detects them and automatically corrects them. The corrected email content is displayed as a preview to the user, and the email is sent after the user has confirmed it.

[1180] Examples of prompt statements

[1181] A user is trying to send an email. Check whether the destination domain is an internal domain or an external domain, and prompt the user to enter a subject if one has not been entered. Inform the user that there is a security risk when sending to an external domain.

[1182] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1183] Step 1:

[1184] The server receives the email address entered by the user. Given an email address as input, the server extracts the domain part from the email address. Specifically, it splits the email address at "@" and takes the latter part as the domain. As output, it gets the extracted domain.

[1185] Step 2:

[1186] The server checks the extracted domain against the internal domain list. Given the extracted domain and the internal domain list as input, the server checks whether the domain is included in the internal domain list. Specifically, it performs a lookup in the list and checks whether the domain is present in the list. As an output, it gives information about whether the domain is an internal or external domain.

[1187] Step 3:

[1188] The server checks whether the email title has been entered. The email title is given as input, and the server checks whether the title is empty. Specifically, it checks the length of the title string, and if the length is zero, it determines that the title has not been entered. As output, it obtains information about whether the title has been entered.

[1189] Step 4:

[1190] If the title is not entered, the server displays a message prompting the user to enter a title. Information that the title is not entered is given as input, and the server displays the message through the user interface. Specifically, a pop-up window or a notification message is displayed. As output, a notification is given to the user.

[1191] Step 5:

[1192] The server uses a natural language processing system to check the email content for typos and omissions and make corrections. The email content is given as input, and the server analyzes the text using a natural language processing library. Specifically, it applies text analysis algorithms to detect typos and omissions and make automatic corrections. The corrected email content is obtained as output.

[1193] Step 6:

[1194] The server displays the corrected email content to the user as a preview. The corrected email content is given as input, and the server displays the preview through a user interface. Specifically, the corrected text is displayed in a window or editor that displays the email content. The output is a preview display to the user.

[1195] Step 7:

[1196] The server notifies the user of the security risk when sending to an external domain. As input, information that the domain is an external domain is given, and the server displays a message to notify the user of the security risk. Specifically, it displays a warning message or an alert. As output, the user is notified of the security risk.

[1197] Step 8:

[1198] After the user confirms, the server will send the email. Given the user's confirmation information and the modified email content as input, the server will use the email sending library to send the email. Specifically, it will send the email to the mail server using the SMTP protocol. As output, the email will be sent.

[1199] Example 2

[1200] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[1201] Conventional email sending systems have insufficient means for checking for typos and omissions before sending, which increases the risk of sending emails with incorrect content. Users also need to manually check the destination domain and the presence or absence of the title, which places a heavy burden on users. Furthermore, there is no checking for harassing expressions or expressions that take into consideration the Sustainable Development Goals, making it difficult to communicate appropriately.

[1202] 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.

[1203] In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omissions in the email content by a natural language processing system, a means for displaying the corrected state in a preview, a means for a user to input an email body and a terminal to send the email body to the server, a means for the server to pass the email body to the natural language processing system and for the natural language processing system to analyze the email body, a means for the server to correct typos and omissions and send the corrected email body to the terminal, and a means for the terminal to display the corrected email body in a preview. This allows the user to send an accurate email without typos or omissions, and the confirmation of the destination domain and the presence or absence of a title is also automated, reducing the burden on the user. In addition, the system also checks for harassment-related expressions and expressions that take into consideration the sustainable development goals, allowing for appropriate communication.

[1204] "Verifying the destination domain before sending email" is the process of verifying whether the intended recipient domain is correct before an email is sent.

[1205] "Title check" is the process of checking whether an email has an appropriate title before it is sent.

[1206] A "natural language processing system" is a system that uses artificial intelligence technology to analyze text data and detect and correct typos, omitted words, grammatical errors, and so on.

[1207] "Checking for and correcting typos and omissions" is the process of detecting incorrect or missing characters in entered text and correcting them to their correct form.

[1208] "Displaying the corrected state in a preview" is a process of displaying the text after the typos and omissions have been corrected so that the user can visually confirm it.

[1209] "A user inputs the body of an email" refers to an action in which a user inputs the contents of an email in text format using a terminal.

[1210] "The terminal sends the email body to the server" refers to the process of sending the email body entered by the user from the terminal to the server via data communication.

[1211] "The server passes the email body to the natural language processing system" refers to the process in which the server inputs the email body it received into the natural language processing system and requests it to analyze it.

[1212] "A natural language processing system analyzes the body of an email" refers to the process in which a natural language processing system analyzes the body of an email and detects typos, omitted words, and grammatical errors.

[1213] "The server corrects typos and spelling errors" refers to the process in which the server corrects typos and spelling errors in the body of the email based on the analysis results from the natural language processing system.

[1214] "Sending the corrected email body to the terminal" refers to the process in which the server sends the corrected email body to the terminal via data communication.

[1215] "Displaying the revised email body in a preview on the terminal" refers to a process in which the terminal displays the revised email body received from the server so that the user can visually confirm it.

[1216] "Harassing language" refers to words or phrases that may offend or hurt others.

[1217] "SDG-friendly expressions" refer to words and phrases that are environmentally and socially conscious and in line with the Sustainable Development Goals (SDGs).

[1218] "Learning from user feedback" is the process by which a system collects opinions and suggestions for corrections from users and uses them to improve the system's performance.

[1219] This invention is a system for checking and correcting typos and omissions before sending an email. The system operates when a user inputs the body of an email and the terminal sends the email body to a server. The server passes the email body to a natural language processing system, which analyzes the email body. Based on the analysis results, the server corrects the typos and omissions and sends the corrected email body to the terminal. The terminal displays a preview of the corrected email body.

[1220] Hardware and Software Used

[1221] Hardware: Servers, terminals (PCs, smartphones, etc.)

[1222] Software: Natural language processing systems (e.g. SpaCy, GPT-3), email client software (e.g. Outlook, Gmail)

[1223] Processing flow

[1224] 1. The user enters the email body

[1225] The user uses the email client software on the terminal to input the body of the email, for example, "Hello, Tanaka-san. The meeting tomorrow is at 10:00. Thank you for your cooperation."

[1226] 2. The device sends the email body to the server

[1227] When the user finishes entering the email body and clicks a button to request checking for typos, the device sends the email body to the server by using an HTTP POST request.

[1228] 3. The server passes the email body to a natural language processing system

[1229] The server passes the received email text to a natural language processing system (e.g. SpaCy, GPT-3). Specifically, it sends the email text to the natural language processing system using an API request.

[1230] 4. A natural language processing system analyzes the email body

[1231] The natural language processing system analyzes the text of the received email and checks for typos and omissions. For example, it determines that "onegaishimasu" (please) should be corrected to "onegaishimasu" (please onegaishimasu).

[1232] 5. The server corrects typos

[1233] The server that receives the analysis results from the natural language processing system corrects any typos or omissions that are detected. Specifically, it performs character string manipulation to change "onegaishimasu" (please) to "onegaishimasu" (please please).

[1234] 6. The server sends the corrected email body to the device.

[1235] The server sends the revised email body to the terminal. Specifically, it returns the revised email body to the terminal as an HTTP response.

[1236] 7. Your device will display a preview of the revised email body.

[1237] The terminal displays a preview of the corrected email text received by the user. For example, the preview screen of the email client software displays "Hello, Tanaka-san. Tomorrow's meeting starts at 10:00. Thank you for your cooperation."

[1238] Examples

[1239] For example, if a user enters the following email body:

[1240] Hello, Tanaka-san.

[1241] The meeting tomorrow will start at 10:00. Thank you for your cooperation.

[1242] If this email text is passed to a natural language processing system, the system will correct "onegaishimasu" to "onegaishimasu."

[1243] The corrected email body will look like this:

[1244] Hello, Tanaka-san.

[1245] The meeting tomorrow will start at 10:00. Thank you.

[1246] Example prompts to be input to the generative AI model

[1247] Email body entered by the user:

[1248] Hello, Tanaka-san.

[1249] The meeting tomorrow will start at 10:00. Thank you for your cooperation.

[1250] Please check the body of this email for typos and correct them.

[1251] This prompt is fed into a generative AI model, which checks for and corrects typos.

[1252] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1253] Step 1:

[1254] The user enters the email body.

[1255] The user uses the email client software on the device to input the body of the email. For example, the user might type, "Hello, Tanaka-san. The meeting tomorrow is at 10:00. Thank you for your cooperation." The inputted body of the email is stored in the device's memory.

[1256] Step 2:

[1257] The device sends the email body to the server.

[1258] When the user finishes entering the email body and clicks a button to request a check for typos, the terminal sends the email body to the server. Specifically, the terminal uses an HTTP POST request to send the email body to the server. The input is the email body entered by the user, and the output is the email body sent to the server.

[1259] Step 3:

[1260] The server passes the email text to a natural language processing system

[1261] The server passes the received email body to the natural language processing system. Specifically, it sends the email body to the natural language processing system using an API request. The input is the email body sent to the server, and the output is the email body passed to the natural language processing system.

[1262] Step 4:

[1263] A natural language processing system analyzes the email body

[1264] The natural language processing system analyzes the body of the received email and checks for typos and omissions. For example, it determines that "onegaishimasu" should be corrected to "onegaishimasu." The input is the email body passed to the natural language processing system, and the output is the analysis result.

[1265] Step 5:

[1266] The server will correct any typos.

[1267] The server that receives the analysis results from the natural language processing system corrects any typos or omissions that are detected. Specifically, it performs string manipulation to correct "onegaishimasu" (please) to "onegaishimasu" (please). The input is the analysis results, and the output is the corrected email body.

[1268] Step 6:

[1269] The server sends the corrected email body to the device.

[1270] The server sends the revised email body to the terminal. Specifically, it returns the revised email body to the terminal as an HTTP response. The input is the revised email body, and the output is the revised email body sent to the terminal.

[1271] Step 7:

[1272] Your device will display a preview of the revised email body.

[1273] The terminal displays a preview of the corrected email body received by the terminal to the user. For example, the preview screen of the email client software displays "Hello, Tanaka-san. Tomorrow's meeting starts at 10:00. Thank you for your cooperation." The input is the corrected email body sent to the terminal, and the output is the corrected email body displayed to the user.

[1274] (Application example 2)

[1275] Next, a description will be given of Application Example 2 of Form 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".

[1276] In electronic payment services, if there are typos or omissions in transaction confirmation emails or customer notification emails, it can reduce user trust and compromise the security of transactions. In addition, there is also the problem that correcting typos and omissions takes time, which reduces work efficiency. Furthermore, because the history of corrections is not saved, past corrections cannot be checked, and there is a risk of repeating the same mistake.

[1277] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omissions in the email content by a natural language processing system, a means for displaying the corrected state in a preview, a means for saving the history of the correction content so that it can be checked later, and a means for the natural language processing system to check and correct typos and omissions when the text of the transaction confirmation email or customer notification email is input. This makes it possible to automatically correct typos and omissions in the transaction confirmation email or customer notification email, and to increase the user's reliability. In addition, by saving the history of the correction content, it is possible to check past correction content and reduce the risk of repeating the same mistake.

[1278] "Checking the destination domain before sending email" means checking whether the domain part of the destination email address is correct before sending the email.

[1279] "Checking for the presence or absence of a title" means checking whether or not the subject of the email has been entered.

[1280] A "natural language processing system" is a technology that enables computers to understand, analyze, and generate human language.

[1281] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in a sentence and replacing them with the correct characters.

[1282] "Display the corrected state in the preview" means that the user can visually confirm the text after typos and omissions have been corrected.

[1283] "Saving a history of modifications so that they can be checked later" means saving a record of modifications made so that the contents of the modifications can be checked later.

[1284] "Transaction confirmation emails and customer notification emails" are emails used in electronic payment services to confirm transactions and notify customers.

[1285] "Learning feedback from users and reflecting the results in checking for and correcting typos and omissions from the next time onwards" means that the system will learn the opinions and corrections provided by users and use them to detect and correct typos and omissions from the next time onwards.

[1286] As an embodiment of the present invention, a system for automatically correcting typos and omissions in transaction confirmation emails and customer notification emails in electronic payment services will be described.

[1287] System Programming

[1288] This system is implemented using Python. A library called SpaCy is used for natural language processing. Specifically, the following processes are performed:

[1289] Processing Description

[1290] Hardware

[1291] Smartphone

[1292] software

[1293] Python

[1294] SpaCy (Natural Language Processing Library)

[1295] Data processing and calculation

[1296] 1. User enters email body:

[1297] The user enters the text of the transaction confirmation email or customer notification email into the smartphone application.

[1298] 2. The server receives the email body:

[1299] The server receives the body of the email sent by the user.

[1300] 3. Analysis by natural language processing system:

[1301] The server analyzes the email body using SpaCy's Japanese model (ja_core_news_sm). It detects and corrects typos and omissions from the analysis results.

[1302] 4. Preview the email body after editing:

[1303] The server displays a preview of the revised email body on the user's smartphone.

[1304] 5. Save revision history:

[1305] The server keeps a record of the modifications made so that they can be reviewed at a later time.

[1306] Examples

[1307] If a user enters the email body as "Hello, this is a test email. There may be typos and spelling errors.", the system will correct it to "Hello, this is a test email. There may be typos and spelling errors."

[1308] Examples of prompt statements

[1309] "Please correct the typos and spelling errors in the following email body: 'Hello, this is a test email. There may be typos and spelling errors.'"

[1310] In this way, typos and omissions in transaction confirmation emails and customer notification emails in electronic payment services can be automatically corrected, increasing user trust. In addition, by saving the history of corrections, past corrections can be checked, reducing the risk of repeating the same mistake.

[1311] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1312] Step 1:

[1313] The user enters the email body.

[1314] The user inputs the text of the transaction confirmation email or customer notification email into the smartphone application. The inputted text of the email is temporarily saved within the application.

[1315] Step 2:

[1316] The terminal sends the email body to the server.

[1317] The terminal sends the email text entered by the user to the server, where it is received and prepared for analysis.

[1318] Step 3:

[1319] The server receives the email body.

[1320] The server receives the body of the email sent from the terminal, and passes the received body of the email to the natural language processing system.

[1321] Step 4:

[1322] The server analyzes the email body using a natural language processing system.

[1323] The server analyzes the email body using SpaCy's Japanese model (ja_core_news_sm). The input for the analysis is the email body, and the output is the detection result of typos and omissions. Specifically, the server tokenizes the email body and analyzes each token to detect typos and omissions.

[1324] Step 5:

[1325] The server will correct any typos or omissions.

[1326] The server corrects typos and omissions based on the analysis results. The input for correction is the detection result of typos and omissions, and the output is the corrected email body. Specifically, the server replaces the parts where typos and omissions are detected with the correct characters.

[1327] Step 6:

[1328] The server sends the revised email body to the terminal for preview display.

[1329] The server sends the corrected email body to the terminal. The input for sending is the corrected email body, and the output is a notification of completion of sending to the terminal. Specifically, the server sends the corrected email body to the terminal, and the terminal receives it.

[1330] Step 7:

[1331] The device will display a preview of the revised email body.

[1332] The terminal receives the corrected email body sent from the server and displays a preview for the user. The input for the preview display is the corrected email body, and the output is a visual display for the user. Specifically, the terminal displays the corrected email body on the screen so that the user can confirm it.

[1333] Step 8:

[1334] The server stores a history of modifications.

[1335] The server keeps a record of the modifications made. The input to the save is the modifications, and the output is a notification that the save was completed. Specifically, the server saves the modifications in a database for later review.

[1336] Example 3

[1337] Next, a third embodiment of the third embodiment will be described. 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".

[1338] In conventional email sending systems, checking and correcting typos and omissions before sending was insufficient, and checking and correcting harassing expressions and expressions that take into consideration the Sustainable Development Goals was often not done. As a result, there was a risk that emails containing misleading or inappropriate expressions would be sent. In addition, the system lacked the ability to learn from user feedback and reflect it in future corrections, which resulted in low adaptability of the system.

[1339] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1340] In this invention, the server includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for collecting feedback from users and saving it in a database, means for learning the collected feedback and reflecting it in checking and correcting typos and omissions from the next time onwards, means for checking whether the email content contains harassment-related expressions or expressions that take into consideration the Sustainable Development Goals and correcting them as necessary, and means for sending the corrected email content to the user's terminal and displaying it as a preview. This makes it possible to effectively correct typos, omissions, and inappropriate expressions before sending an email and to improve the adaptability of the system by learning from user feedback.

[1341] "Destination domain verification" is the process of verifying whether the domain name of the email recipient is correct before the email is sent.

[1342] "Checking for the presence of a title" is a process of checking whether the title of an email is set appropriately before the email is sent.

[1343] A "natural language processing system" is a system that uses artificial intelligence technology to analyze text data and understand meaning and grammar.

[1344] "Checking for and correcting typos and omissions" refers to the process of detecting incorrect or missing characters in the content of an email and correcting them to the correct form.

[1345] "Preview" refers to the process of displaying the revised email content on the screen for the user to review.

[1346] "Collecting feedback and storing in a database" is the process of collecting opinions and ratings from users and storing them in a database.

[1347] "Learning from the feedback and reflecting it in checking and correcting typos and omissions from the next time onwards" refers to the process in which the system learns from the collected feedback and reflects the results of that learning when checking and correcting the content of emails from the next time onwards.

[1348] "Checking and correcting harassing language and language that takes into consideration the Sustainable Development Goals" refers to the process of detecting language contained in emails that is harassing or contrary to the Sustainable Development Goals, and correcting it to appropriate language.

[1349] "Send to user's device and display as preview" refers to the process of sending the revised email content to the user's device and displaying it for the user to review.

[1350] The present invention is a system that effectively corrects typos, omissions, and inappropriate expressions before sending an email, and improves the adaptability of the system by learning from user feedback. A specific embodiment of this system will be described below.

[1351] Hardware and Software Used

[1352] Hardware: Servers, terminals

[1353] Software: Natural language processing systems (e.g. GPT-4), database management systems (e.g. MySQL)

[1354] System Configuration

[1355] 1. Receiving email content

[1356] User composes an email and clicks the send button.

[1357] The terminal transmits the contents of the email created by the user to the server.

[1358] The server temporarily stores the contents of the received email.

[1359] 2. Email content analysis

[1360] The server transmits the stored email content to a natural language processing system.

[1361] A natural language processing system analyzes email content and understands the structure and meaning of the text.

[1362] 3. Check for harassment-related expressions and expressions that consider the SDGs

[1363] Based on the analysis results, the natural language processing system will check whether the email contains any harassing language or language that takes into consideration the SDGs.

[1364] The server receives the verification result and determines whether or not there is a matching expression.

[1365] 4. Generate the corrections

[1366] If a matching expression is found, the natural language processing system generates new text to correct the expression.

[1367] The server receives the generated modifications and merges them with the original mail content.

[1368] 5. Preview of edits

[1369] The server sends the corrected email content to the user's terminal.

[1370] The device will display a preview of the revised email content.

[1371] 6. Collecting User Feedback

[1372] The user provides feedback on the modifications.

[1373] The terminal transmits the user's feedback to the server.

[1374] The server stores the feedback in a database.

[1375] 7. Learning feedback

[1376] The natural language processing system learns from the collected feedback and uses it to check and correct typos and omissions in future posts.

[1377] The server stores the learning results and uses them the next time it analyzes an email.

[1378] Examples

[1379] Example 1:

[1380] What the user is trying to send: "This project is not progressing at all. Try harder."

[1381] The server receives the email content and analyzes it using a natural language processing system.

[1382] The natural language processing system determines that the expression "No progress at all" is harassing, and corrects it to "Progress seems to be slow. Let's think together about ways to improve the situation."

[1383] The server sends the revised email content to the user's device and displays it as a preview.

[1384] Users provide feedback on the corrections, saying, "This expression is appropriate."

[1385] The server collects the feedback and stores it in a database.

[1386] The natural language processing system learns from the collected feedback and incorporates it into future email revisions.

[1387] Example prompt:

[1388] "Please check whether the content of this email contains harassing language and revise it if necessary."

[1389] Please check that the content of this email includes language that takes the SDGs into consideration and revise it if necessary.

[1390] In this way, the system can effectively correct typos, omissions, and inappropriate expressions before sending an email, and can learn from user feedback to improve the adaptability of the system.The flow of the identification process in the third embodiment will be described with reference to FIG.

[1391] Step 1:

[1392] Receiving email content

[1393] User composes an email and clicks the send button.

[1394] The terminal transmits the contents of the email created by the user to the server.

[1395] The server temporarily stores the contents of the received email.

[1396] Input: User-created email content

[1397] Output: Email contents stored on the server

[1398] Specific behavior:

[1399] When the user clicks the "Send" button, the terminal sends the email contents to the server.

[1400] The server stores the contents of the received email in a database.

[1401] Step 2:

[1402] Email content analysis

[1403] The server transmits the stored email content to a natural language processing system.

[1404] A natural language processing system analyzes email content and understands the structure and meaning of the text.

[1405] Input: Email contents stored on the server

[1406] Output: Structure and meaning of parsed email content

[1407] Specific behavior:

[1408] The server sends the email content to a natural language processing system via an API.

[1409] The natural language processing system tokenizes the email content and performs grammatical analysis.

[1410] Step 3:

[1411] Check for harassment-related expressions and expressions that consider the SDGs

[1412] Based on the analysis results, the natural language processing system will check whether the email contains any harassing language or language that takes into consideration the SDGs.

[1413] The server receives the verification result and determines whether or not there is a matching expression.

[1414] Input: Structure and meaning of parsed email content

[1415] Output: Detection results for harassment-related expressions and expressions that are insensitive to the SDGs

[1416] Specific behavior:

[1417] Natural language processing systems use algorithms to detect specific keywords and phrases.

[1418] The server analyzes the detection results and determines whether remediation is necessary.

[1419] Step 4:

[1420] Generate corrections

[1421] If a matching expression is found, the natural language processing system generates new text to correct the expression.

[1422] The server receives the generated modifications and merges them with the original mail content.

[1423] Input: Detection results for harassment-related expressions and expressions that are insensitive to the SDGs

[1424] Output: The modified email content

[1425] Specific behavior:

[1426] A natural language processing system replaces the original expression with a more appropriate expression.

[1427] The server stores the revised email content as new text.

[1428] Step 5:

[1429] Preview of the changes

[1430] The server sends the corrected email content to the user's terminal.

[1431] The device will display a preview of the revised email content.

[1432] Input: Modified email content

[1433] Output: A preview displayed on the user's device.

[1434] Specific behavior:

[1435] The server sends the corrected email content to the user's terminal.

[1436] The terminal displays the corrected email content on the screen and asks the user to confirm it.

[1437] Step 6:

[1438] Collecting user feedback

[1439] The user provides feedback on the modifications.

[1440] The terminal transmits the user's feedback to the server.

[1441] The server stores the feedback in a database.

[1442] Input: User feedback

[1443] Output: Feedback stored in a database

[1444] Specific behavior:

[1445] The user enters their feedback and clicks the "Submit" button.

[1446] The terminal transmits the feedback content to the server.

[1447] The server stores the feedback in a database.

[1448] Step 7:

[1449] Learning Feedback

[1450] The natural language processing system learns from the collected feedback and uses it to check and correct typos and omissions in future posts.

[1451] The server stores the learning results and uses them the next time it analyzes an email.

[1452] Input: Feedback stored in the database

[1453] Output: The trained model parameters.

[1454] Specific behavior:

[1455] The natural language processing system uses the feedback data to update the parameters of the model.

[1456] The server uses the updated model the next time it parses the email.

[1457] (Application example 3)

[1458] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1459] Email is an important means of modern communication, but it can contain typos, omissions, harassing language, or language that does not consider the Sustainable Development Goals. This can lead to misunderstandings and trouble, and can damage the credibility of companies and individuals. Another challenge is to effectively utilize feedback from users to improve the accuracy of the system. Furthermore, there is a demand to make these functions available on mobile devices such as smartphones.

[1460] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1461] In this invention, the server includes a means for checking the destination domain and whether or not there is a title before sending an email, a means for checking and correcting typos and omissions in the email content using a natural language processing system, a means for displaying the corrected state in a preview, a means for checking and correcting harassment-related expressions and expressions that take into consideration the sustainable development goals, a means for learning feedback from users and reflecting the results in checking and correcting typos and omissions from the next time onwards, a means for operating as an application to be installed on a smartphone, a means for making appropriate correction suggestions using a generative AI model, and a means for generating prompt sentences and obtaining appropriate correction suggestions. This makes it possible to ensure that the content of the email is appropriate and prevent misunderstandings and problems from occurring.

[1462] "Checking the destination domain before sending email" means checking the domain part of the destination email address to confirm that it is a valid domain before sending email.

[1463] "Checking for the presence or absence of a title" means checking whether the title of an email is entered properly before the email is sent.

[1464] A "natural language processing system" is a technology that enables computers to understand and process human language, and is a system that checks for and corrects typos and omissions.

[1465] "Checking for and correcting typos and omissions" means checking the contents of the email for incorrect or missing characters and correcting them if necessary.

[1466] "Show revised state in preview" means displaying the contents of the email after typos, omissions, and other corrections have been made so that the user can check them.

[1467] "Checking for and correcting harassing language" means checking whether the content of an email contains any language that could be considered harassing, and correcting it if necessary.

[1468] "Checking and correcting expressions that take into consideration the Sustainable Development Goals" means checking whether the content of the email contains expressions that go against the Sustainable Development Goals (SDGs) and correcting them if necessary.

[1469] "Learning feedback from users" means that the system learns the opinions and evaluations provided by users to the system and reflects them in subsequent processing.

[1470] An "application installed on a smartphone" is a software program that runs on a smartphone and is provided in a form that is usable by the user.

[1471] A "generative AI model" is a model that uses artificial intelligence technology to generate appropriate outputs for specific tasks.

[1472] "Generating a prompt sentence and obtaining appropriate correction suggestions" means that the system generates a sentence to make correction suggestions based on the content of the input email, and uses that sentence to obtain appropriate correction suggestions.

[1473] In order to implement the present invention, the following system configuration and processing procedure are used.

[1474] System Configuration

[1475] 1. Hardware:

[1476] Server: A server with high computing power is required. It is used to analyze the content of the email and generate correction suggestions.

[1477] Device: A mobile device such as a smartphone or tablet is required. The user will use this device to send the email.

[1478] 2. Software:

[1479] Natural Language Processing System: A system to check and correct typos, omissions, harassing expressions, and expressions that are considerate of the Sustainable Development Goals, using the Hugging Face Transformers library.

[1480] Generative AI model: A model to provide appropriate correction suggestions. For example, we use the distilbert-base-uncased-finetuned-sst-2-english model.

[1481] Application: An application installed on a smartphone that users use to send emails. It transmits the contents of the email to the server and receives suggestions for corrections.

[1482] Processing flow

[1483] 1. Before sending an email, check:

[1484] The server receives the contents of the email sent from the terminal.

[1485] The server checks the destination domain and whether or not the title is included.

[1486] 2. Analysis using natural language processing:

[1487] The server analyzes the content of the email using a natural language processing system.

[1488] Check for typos, omissions, harassment, and Sustainable Development Goal-friendly language, and generate suggested revisions as needed.

[1489] 3. Generate correction suggestions:

[1490] The server uses a generative AI model to make appropriate correction suggestions.

[1491] Generate prompt sentences and get appropriate correction suggestions.

[1492] 4. Preview after correction:

[1493] The contents of the email, including suggested corrections, are sent to the device and a preview is displayed to the user.

[1494] 5. Learning from user feedback:

[1495] Users provide feedback on the proposed revisions.

[1496] The server learns from this feedback and uses it to check for and correct typos and omissions from the next time onwards.

[1497] Examples

[1498] For example, consider the case where a user tries to send an email like this:

[1499] Example of email content

[1500] This is a sample email with bad word1.

[1501] The server analyzes the email content and generates correction suggestions such as the following:

[1502] Examples of suggested revisions

[1503] This is a sample email with [Suggested correction: Change to appropriate wording].

[1504] The system will display a preview of the proposed corrections to the user, and once the user has confirmed the corrections, will send an email. By providing feedback on the proposed corrections, the system can improve the accuracy of corrections in future.

[1505] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1506] Step 1:

[1507] The server receives the contents of the email sent from the terminal.

[1508] Input: The email content that the user creates on their device.

[1509] Output: The email content is sent to the server.

[1510] Specific operation: When a user composes an email on the terminal and presses the send button, the contents of the email are sent to the server.

[1511] Step 2:

[1512] The server checks the destination domain and whether or not the title is included.

[1513] Input: The email content sent from the device.

[1514] Output: The results of the domain and title check.

[1515] What happens: The server extracts the domain part of the email address and verifies that it is a valid domain. It also checks that the subject of the email is entered correctly.

[1516] Step 3:

[1517] The server analyzes the content of the email using a natural language processing system.

[1518] Input: The email content sent from the device.

[1519] Output: Check results for typos, omissions, harassment, and expressions that take into consideration the Sustainable Development Goals.

[1520] What it does: The server uses a natural language processing system (Hugging Face's Transformers library) to analyze the content of the email and check for typos, omissions, harassment, and language that takes the Sustainable Development Goals into consideration.

[1521] Step 4:

[1522] The server uses a generative AI model to make appropriate correction suggestions.

[1523] Input: Analysis results from a natural language processing system.

[1524] Output: suggested fixes.

[1525] What happens: The server uses a generative AI model (distilbert-base-uncased-finetuned-sst-2-english) to generate appropriate correction suggestions based on the analysis results.

[1526] Step 5:

[1527] The server generates the prompt sentence and obtains appropriate correction suggestions.

[1528] Input: suggested fix.

[1529] Output: The prompt text and suggested corrections.

[1530] Specific behavior: The server generates a prompt based on the correction suggestions, and uses the prompt to get the appropriate correction suggestions.

[1531] Step 6:

[1532] The contents of the email, including suggested corrections, are sent to the device and a preview is displayed to the user.

[1533] Input: The prompt text and suggested corrections.

[1534] Output: A preview of the email content after corrections.

[1535] What it does: The server sends the email content, including suggested edits, to the device, allowing the user to preview the revised email content.

[1536] Step 7:

[1537] Users provide feedback on the proposed revisions.

[1538] Input: Preview of the email content after revision.

[1539] Output: User feedback.

[1540] What happens: The user reviews the revised email and provides feedback on the suggested changes.

[1541] Step 8:

[1542] The server learns from this feedback and uses it to check for and correct typos and omissions from the next time onwards.

[1543] Input: User feedback.

[1544] Output: The training results.

[1545] What it does: The server learns from user feedback and updates the model to reflect it in future checks and corrections of typos and omissions.

[1546] In addition, 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.

[1547] "Example 1"

[1548] One embodiment of the present invention is an email system incorporating an emotion engine. In this system, when a user composes an email, a natural language processing system and the emotion engine work together. Specifically, when the user enters the body of the email, the natural language processing system analyzes the content and checks for typos and missing characters. At the same time, the emotion engine recognizes the user's emotion and corrects the content of the email based on that emotion.

[1549] "Example 2"

[1550] The emotion engine can also control the sending of emails based on the user's emotions. For example, if the user feels negative emotions such as anger or sadness, the emotion engine can pause sending the email and prompt the user to reconsider. This can prevent the wrong emails from being sent based on emotional decisions.

[1551] "Example 3"

[1552] In addition, the emotion engine can automatically adjust the expression of the email according to the user's emotions. For example, if the user feels that he or she has positive emotions such as joy or excitement, the emotion engine automatically modifies the expression of the email to be brighter and more lively, which can better convey the user's emotions.

[1553] "Example 4"

[1554] The emotion engine can also learn from user feedback and reflect the results in future emotion recognition and email corrections. For example, if a user gives feedback such as "This email does not accurately reflect my emotions," the emotion engine can learn from that feedback and from the next time onwards, it will be able to more accurately recognize emotions and correct emails.

[1555] The process flow of each embodiment will be described below.

[1556] "Example 1"

[1557] Step 1: The user enters the body of the email.

[1558] Step 2: A natural language processing system analyzes the content of the email and checks for typos and spelling errors.

[1559] Step 3: The emotion engine recognizes the user's emotion.

[1560] Step 4: The emotion engine modifies the email content based on the emotion.

[1561] "Example 2"

[1562] Step 1: The emotion engine recognizes the user's emotion.

[1563] Step 2: If the sentiment engine determines that the user's sentiment is negative, pause sending emails.

[1564] Step 3: The emotion engine prompts the user to reconsider the email.

[1565] "Example 3"

[1566] Step 1: The emotion engine recognizes the user's emotion.

[1567] Step 2: If the sentiment engine determines that the user's sentiment is positive, it automatically modifies the email's tone to make it more upbeat and vibrant.

[1568] "Example 4"

[1569] Step 1: The user provides feedback to the emotion engine.

[1570] Step 2: The emotion engine learns from the feedback.

[1571] Step 3: The emotion engine uses the feedback to improve emotion recognition and email revisions for the next time.

[1572] Example 1

[1573] Next, a description will be given of Example 1 of Form 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".

[1574] In conventional email sending systems, confirmation of the recipient's domain and the presence or absence of a title are often done manually, which can easily lead to problems due to sending by mistake or not entering a title. In addition, checking for typos and omissions in the content of the email is also done manually, which is inefficient and may leave errors. Furthermore, if the content of the email is emotional, sending it as is can cause misunderstandings and problems. A system that solves these problems is needed.

[1575] 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.

[1576] In this invention, the server includes a means for checking the destination domain before sending, a means for checking the presence or absence of a title, a means for checking the content of the email for typos and omissions using a natural language processing system, a means for recognizing the user's emotions using an emotion engine and proposing corrections to the content of the email based on the emotions, a means for displaying the corrected state in a preview, and a means for providing feedback to the user. This automatically checks the domain and title before sending an email, making it possible to prevent erroneous sending or leaving the title blank. Furthermore, the quality of the email content is improved by automatically checking for typos and omissions and proposing corrections, and the content correction suggestions by the emotion engine can prevent problems caused by sending emotional emails.

[1577] "Checking the destination domain" means analyzing the domain portion of the destination address of an email to determine whether it is an internal domain or an external domain.

[1578] "Check for title" means checking whether the subject field of the email is empty and prompting the user to enter a title if necessary.

[1579] A "natural language processing system" is a system that analyzes text data, detects typos, omitted words, and grammatical errors, and suggests corrections.

[1580] "Checking for typos and omissions" means detecting typos and omissions in the body of the email and making suggestions for correcting them.

[1581] The "emotion engine" is a system that analyzes user emotions from text data and provides appropriate feedback and suggested corrections based on those emotions.

[1582] "Show post-correction state in preview" means visually displaying to the user the contents of the email after applying suggested corrections for typos and omissions, or corrections suggested by the emotion engine.

[1583] "Providing feedback to the user" means notifying the user of the results of the system's domain check, title check, suggestions for correction of typos, and suggestions for correction based on the emotion engine, and encouraging them to take necessary action.

[1584] The present invention is an email sending system that performs the following functions automatically when a user creates an email: checking the destination domain, checking the existence of a title, checking for spelling errors using a natural language processing system, and recognizing emotions and suggesting corrections using an emotion engine. This system automatically performs these functions and provides feedback to the user when the user creates an email.

[1585] Hardware and Software Used

[1586] 1. Server

[1587] The server plays a central role in the mail transmission system. It receives input from users and performs various checks and analysis.

[1588] 2. Natural Language Processing Systems

[1589] The Google Cloud Natural Language API is used as a natural language processing system that detects typos and omissions in the email text and makes correction suggestions.

[1590] 3. Emotion Engine

[1591] The emotion engine uses IBM Watson Tone Analyzer, which analyzes the user's emotions from the email text and makes correction suggestions based on those emotions.

[1592] Data processing and calculation

[1593] 1. Verify the recipient's domain

[1594] The server analyzes the destination address entered by the user, extracts the domain part, and checks the extracted domain against the internal domain list to determine whether it is an internal domain or an external domain.

[1595] 2. Check if there is a title

[1596] The server checks if the email's subject field is empty. If it is, it displays a message prompting the user to enter a subject.

[1597] 3. Check for typos

[1598] The server sends the email text to a natural language processing system to detect typos and spelling errors, and generates correction suggestions for the detected typos and spelling errors and displays them to the user.

[1599] 4. Emotion recognition and correction suggestions

[1600] The server sends the email text to the emotion engine to analyze the user's emotions. Based on the analysis results, it generates suggestions to modify the emotional expressions and displays them to the user.

[1601] Examples

[1602] If a user enters a message with a destination address of "example@gmail.com", no subject line, and no body text, the server does the following:

[1603] 1. Verify the recipient's domain

[1604] The server receives "example@gmail.com", extracts the domain part "@gmail.com", checks it against the internal domain list, and determines that it is an external domain.

[1605] 2. Check if there is a title

[1606] The server sees that the title field is empty and displays the message "Please enter a title" to the user.

[1607] 3. Check for typos

[1608] The server sends the text to the Google Cloud Natural Language API to check for typos and spelling errors. For example, it checks that there are no typos in the text "Hello, this is a test email."

[1609] 4. Emotion recognition and correction suggestions

[1610] The server sends the text to IBM Watson Tone Analyzer to analyze the sentiment. For example, in response to the text, "This project is not progressing at all. I'm very frustrated," the server generates a correction suggestion, "This project seems to be progressing slowly. Let's think of ways to improve it," and displays it to the user.

[1611] Examples of prompt statements

[1612] "Please describe a system that uses a natural language processing system and an emotion engine to analyze the text entered by the user, check for typos and omissions, recognize emotions, and then correct the content of the email based on those emotions."

[1613] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1614] Step 1:

[1615] A user composes an email.

[1616] Input: The user enters the destination address, subject, and message body using an email client.

[1617] What happens: The user opens an email client (e.g. Outlook, Gmail) and composes a new email, entering "example@gmail.com" in the To field, leaving the subject blank, and typing "Hello, this is a test email" in the body.

[1618] Output: The entered destination address, subject, and body are sent to the server.

[1619] Step 2:

[1620] The server checks the destination domain.

[1621] Input: The destination address entered by the user.

[1622] Specific operation: The server receives the destination address "example@gmail.com" and extracts the domain part "@gmail.com".

[1623] Data processing: Check the extracted domains against an internal domain list.

[1624] Output: The result of determining whether the destination is an internal domain or an external domain.

[1625] Step 3:

[1626] The server checks whether the title exists.

[1627] Input: A title field entered by the user.

[1628] Specific Behavior: The server verifies that the title field is empty.

[1629] Data Operation: Check if the title field is empty.

[1630] Output: If title is empty, show the user the message "Please enter a title".

[1631] Step 4:

[1632] The server uses a natural language processing system to check the email content for typos and missing characters.

[1633] Input: The email body entered by the user.

[1634] Specific operation: The server sends the text “Hello, this is a test email” to a natural language processing system (e.g., Google Cloud Natural Language API).

[1635] Data processing: A natural language processing system analyzes the text and detects typos and omissions.

[1636] Output: Detected typos and omissions and suggested corrections.

[1637] Step 5:

[1638] The server uses an emotion engine to recognize the user's emotions and makes suggestions to modify the content of the email based on those emotions.

[1639] Input: The email body entered by the user.

[1640] Specific operation: The server sends the text to an emotion engine (e.g. IBM Watson Tone Analyzer).

[1641] Data processing: The emotion engine analyzes the text and recognizes the user's emotions.

[1642] Output: Emotion recognition results and correction suggestions. For example, for the text "This project is not progressing at all. I'm very frustrated," we generate a correction suggestion such as "This project seems to be progressing slowly. Let's think of ways to improve it."

[1643] Step 6:

[1644] The server will display the revised state in a preview.

[1645] Input: Typo suggestions and sentiment engine suggestions.

[1646] Specific operation: The server generates the email content after applying the proposed modifications.

[1647] Data calculation: Apply the suggested corrections and generate the corrected email content.

[1648] Output: A preview of the revised email content is displayed to the user.

[1649] Step 7:

[1650] The server provides feedback to the user.

[1651] Input: Domain check result, Title check result, Suggested corrections for typos, Sentiment engine suggestions.

[1652] Specific behavior: The server notifies the user of these results and prompts them to take any necessary action.

[1653] Data Calculation: Integrates each check result and correction suggestions to generate a feedback message for the user.

[1654] Output: A feedback message to the user, for example "Title required for submission to external domain. Misspellings detected. Contains emotional content."

[1655] (Application example 1)

[1656] Next, a description will be given of Application Example 1 of Form 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".

[1657] In conventional email sending systems, there was insufficient checking of the recipient's domain and whether or not there was a title, which led to problems such as erroneous sending and not entering a title. In addition, it was difficult for customer support robots to properly recognize the customer's emotions and generate responses based on them, which could lead to a decrease in customer satisfaction. Furthermore, when sending customer inquiries by email, there was a lack of functionality to check the recipient's domain and whether or not there was a title.

[1658] 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. In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an e-mail, a means for checking and correcting typos and omissions in the e-mail content by a natural language processing system, a means for displaying the corrected state in a preview, and a means for analyzing the customer's input and generating an appropriate response in a customer service robot, a means for recognizing the customer's emotions and correcting the response based on the emotions, and a means for checking the destination domain and the presence or absence of a title when sending the customer's inquiry content by e-mail. This makes it possible to prevent erroneous sending of e-mails and failure to enter a title, and enables the customer service robot to generate an appropriate response according to the customer's emotions, thereby improving customer satisfaction.

[1659] "Checking the destination domain before sending email" means automatically checking whether the domain part of the destination email address is an internal domain or an external domain before sending the email.

[1660] "Checking whether or not there is a title" means checking whether or not the title of the email has been entered before sending the email.

[1661] A "natural language processing system" is a system that analyzes text data and checks for and corrects typos and omissions.

[1662] "Checking for and correcting typos and omissions" refers to detecting incorrect or missing characters in text and correcting them to the correct form.

[1663] "Display the corrected state in the preview" means visually displaying to the user the contents of the email after typos and omissions have been corrected.

[1664] A "customer service robot" is a robot that responds to customer inquiries and requests in a physical store.

[1665] "Analyzing customer input and generating an appropriate response" means analyzing what the customer says to the robot and generating an appropriate reply based on that content.

[1666] "Recognizing the customer's emotions and modifying responses based on those emotions" means reading the customer's emotions from the way they speak and the content of their messages, and providing an appropriate response based on those emotions.

[1667] "When sending a customer inquiry by email, check the destination domain and whether or not a title is entered" means that when sending a customer inquiry by email to store staff, check whether the destination domain is appropriate and whether a title has been entered.

[1668] As an embodiment of the present invention, a specific implementation method of a system in a customer service robot will be described below.

[1669] First, the customer-facing robot uses hardware with a voice recognition microphone, speaker and internet connection, and software using Python, TextBlob, spaCy and smtplib.

[1670] When a customer speaks to the robot, the robot converts the voice into text data through a voice recognition microphone. This text data is analyzed using a natural language processing system (spaCy). The analyzed text data undergoes sentiment analysis using TextBlob. Based on the result of the sentiment analysis, the robot generates an appropriate response.

[1671] The generated response is communicated to the customer through the speaker. At the same time, the customer's inquiry is sent to the store staff via email. Before sending the email, the destination domain and the presence or absence of a title are checked. This makes it possible to prevent sending to the wrong address or leaving the title blank.

[1672] As a concrete example, if a customer asks the robot, "Is the product in stock?", the robot responds, "We will check the stock. Please wait a moment," and sends the inquiry to the store staff by email. At this time, the robot checks whether the destination domain is an internal domain or an external domain, and checks whether a title has been entered.

[1673] An example of a prompt would be:

[1674] "If a customer asks, 'Is the item in stock?' the robot should respond with, 'We'll check the stock. Please wait a moment,' and then email the inquiry to a store staff member."

[1675] In this way, the customer service robot can generate appropriate responses according to the customer's emotions, improving customer satisfaction. It can also prevent emails from being sent by mistake or without a title.

[1676] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1677] Step 1:

[1678] The customer speaks to the robot. The customer's voice input is converted into text data through a voice recognition microphone. The input is the customer's voice, and the output is text data.

[1679] Step 2:

[1680] The server sends the text data to a natural language processing system (spaCy) for analysis. The input is text data, and the output is the analyzed text data. Specifically, the system analyzes the grammatical structure and meaning of the text.

[1681] Step 3:

[1682] The server sends the analyzed text data to the sentiment analysis system (TextBlob) for sentiment analysis. The input is the analyzed text data, and the output is the result of sentiment analysis. The specific operation is to classify the text sentiment into positive, negative, and neutral.

[1683] Step 4:

[1684] The server generates an appropriate response based on the results of the sentiment analysis. The input is the result of the sentiment analysis, and the output is the generated response text. The specific behavior is to generate a positive response for positive sentiment, and a response containing an apology or solution for negative sentiment.

[1685] Step 5:

[1686] The server transmits the generated response text to the customer through the speaker. The input is the generated response text, and the output is the voice response. The specific operation is to convert the text into voice and play it back through the speaker.

[1687] Step 6:

[1688] The server prepares to send the customer's inquiry to the store staff via email. The input is the text of the customer's inquiry, and the output is a draft of the email. Specifically, the server generates the body of the email and sets the destination email address and subject.

[1689] Step 7:

[1690] Before sending an email, the server checks the destination domain and whether or not there is a title. The input is the email draft, and the output is the confirmation result. Specifically, it checks whether the destination domain is an internal or external domain, and whether a title has been entered.

[1691] Step 8:

[1692] If the verification result is OK, the server sends an email to the store staff. The input is a draft of the verified email, and the output is the sent email. The specific operation is to connect to the email server and send the email.

[1693] In this way, the customer service robot can generate appropriate responses according to the customer's emotions, improving customer satisfaction. It can also prevent emails from being sent by mistake or without a title.

[1694] Example 2

[1695] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[1696] In conventional email sending systems, checking and correcting typos and omissions is often done manually, which is time-consuming for users. In addition, there is no way to prevent emotional emails from being sent, which can lead to problems caused by sending incorrect emails. Furthermore, there is a risk that users will unintentionally use inappropriate language because the checking and correction of harassing expressions and expressions that take into consideration the Sustainable Development Goals is not automated. A system that can solve these problems is needed.

[1697] 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.

[1698] In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omissions in the email content using a natural language processing system, a means for displaying the corrected state in a preview, and a means for analyzing the user's emotions and pausing the sending of the email and encouraging the user to reconsider if negative emotions are detected. This not only enables the user to send accurate emails without typos and omissions, but also prevents the user from sending erroneous emails based on emotional judgment. In addition, the checking and correction of harassment-related expressions and expressions that take into consideration the Sustainable Development Goals are also automated, reducing the risk of the user unintentionally using inappropriate expressions.

[1699] "Verifying the destination domain" means verifying whether the domain part of the destination address of the email is correct.

[1700] "Checking whether or not there is a title" means checking whether or not there is a title set in the email.

[1701] A "natural language processing system" is a system that analyzes text data and automatically determines typos, omissions, emotions, and the appropriateness of expressions.

[1702] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in the body of an email and correcting them to the correct form.

[1703] "Display in preview" means to display the corrected email body on the screen so that the user can visually confirm it.

[1704] "Analyzing emotions" means reading the user's emotions from the email body and determining whether those emotions are positive or negative.

[1705] "If negative emotions are detected, the sending of the email will be paused and the user will be prompted to reconsider" means that if it is determined that the user has negative emotions such as anger or sadness, the sending of the email will be temporarily stopped and the user will be prompted to reconsider sending it.

[1706] "Harassing speech" is any offensive, discriminatory or inappropriate speech that may make others feel uncomfortable.

[1707] "Expression that takes into consideration the Sustainable Development Goals" means expression that does not contradict the Sustainable Development Goals, such as environmental protection, social justice, and economic development.

[1708] "Learning feedback from users" means that the system records the opinions and corrections provided by users and reflects them in future analyses and corrections.

[1709] The present invention is directed to a mail sending system that allows users to send accurate and appropriate mail by checking for and correcting typos and omissions and analyzing emotions. A specific embodiment of this system will be described below.

[1710] Hardware and Software Used

[1711] The system uses the following hardware and software:

[1712] Server: A central processing unit for analyzing and correcting email text and for sentiment analysis. Specifically, a cloud-based server is used.

[1713] Terminal: A device on which a user types email and previews the email after editing. Specifically, this applies to PCs, smartphones, tablets, etc.

[1714] Natural language processing system: Software for checking and correcting typos and omissions in email text. Specifically, it uses the Google Cloud Natural Language API.

[1715] Sentiment Engine: Software to analyze user emotions and pause sending emails if negative emotions are detected, specifically using IBM Watson Tone Analyzer.

[1716] Data processing and calculation

[1717] 1. The user enters the body of the email. For example, the user enters "Hello, it's a nice day today."

[1718] 2. The terminal sends the entered email body to the server. The terminal sends the entered text data to the server as an HTTP request.

[1719] 3. The server receives the email body and calls the natural language processing system. The server sends the received text data to the Google Cloud Natural Language API.

[1720] 4. The server analyzes the email body using a natural language processing system and receives the analysis results returned from the API.

[1721] 5. The server detects and corrects typos and omissions based on the analysis results. For example, it corrects "konnichiwa" to "konnichiwa" and "kyo" to "kyo".

[1722] 6. The server sends the corrected email body to the terminal. The server sends the corrected text data to the terminal as an HTTP response.

[1723] 7. The terminal displays a preview of the corrected email body to the user. The terminal displays the received corrected text data on the screen.

[1724] Additionally, a process of sentiment analysis is also included.

[1725] 1. The user types the body of the email, for example, "I don't want to talk to you anymore!"

[1726] 2. The terminal sends the entered email body to the server. The terminal sends the entered text data to the server as an HTTP request.

[1727] 3. The server receives the email body and calls the emotion engine. The server sends the received text data to the IBM Watson Tone Analyzer.

[1728] 4. The server analyzes the user's emotions using the emotion engine. The server receives the emotion analysis results returned from the API.

[1729] 5. The server determines that the user's emotion is negative based on the analysis results. For example, it checks the emotion scores of "anger" or "sadness."

[1730] 6. The server suspends sending of the email and sends a message to the terminal prompting the user to reconsider. The server sends the message "Do you want to send this email? Please reconsider" to the terminal as an HTTP response.

[1731] 7. The terminal displays a message to the user urging him or her to reconsider. The terminal displays the received message on the screen.

[1732] 8. The user reconsiders and confirms sending the email again. The user can either reconsider and press the send button again, or cancel the sending.

[1733] Examples of prompt statements

[1734] Example of a prompt to be input to a generative AI model:

[1735] "Please correct the typo or omission in the following email body: 'Hello, it's a lovely day today.'"

[1736] "Analyze the sentiment of the following email body text and note the negative sentiment: 'I don't want to talk to you anymore!'"

[1737] In this way, the system analyzes the email text entered by the user, corrects typos and omissions, and analyzes emotions, thereby supporting the sending of more accurate and appropriate emails.

[1738] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1739] Step 1:

[1740] The user enters the body of the email.

[1741] Input: The email body entered by the user (e.g. "Hello, it's a nice day today.")

[1742] Output: The entered email text is displayed on the terminal.

[1743] Specific operation: The user enters text into the email composition screen of the device.

[1744] Step 2:

[1745] The terminal sends the entered email text to the server.

[1746] Input: The email body entered by the user

[1747] Output: The email body is sent to the server.

[1748] Specific operation: The terminal sends the entered text data to the server as an HTTP request.

[1749] Step 3:

[1750] The server receives the email body and calls the natural language processing system.

[1751] Input: Email body sent from the device

[1752] Output: Text data sent to a natural language processing system

[1753] Specific operation: The server sends the received text data to the Google Cloud Natural Language API.

[1754] Step 4:

[1755] The server analyzes the email body using a natural language processing system.

[1756] Input: Email body sent to natural language processing system

[1757] Output: Analysis results (information on typos and omissions)

[1758] Specific operation: The server receives the analysis results returned from the API.

[1759] Step 5:

[1760] The server detects and corrects typos and omissions based on the analysis results.

[1761] Input: Analysis results (information on typos and omissions)

[1762] Output: Email body after correction

[1763] Specific operation: The server corrects the typo "konnichiwa" to "konnichiwa" and the omitted typo "kyo" to "kyo".

[1764] Step 6:

[1765] The server sends the corrected email body to the terminal.

[1766] Input: Email body after correction

[1767] Output: The revised email body sent to the device

[1768] Specific operation: The server sends the corrected text data to the terminal as an HTTP response.

[1769] Step 7:

[1770] The terminal displays a preview of the corrected email body to the user.

[1771] Input: Email body after correction

[1772] Output: The revised email body as seen by the user

[1773] Specific operation: The terminal displays the received corrected text data on the screen.

[1774] Step 8:

[1775] The server receives the email body and invokes the emotion engine.

[1776] Input: Email body sent from the device

[1777] Output: Text data sent to the emotion engine

[1778] Specific operation: The server sends the received text data to the IBM Watson Tone Analyzer.

[1779] Step 9:

[1780] The server uses an emotion engine to analyze the user's emotions.

[1781] Input: Email body sent to the sentiment engine

[1782] Output: Sentiment analysis results (sentiment score)

[1783] Specific operation: The server receives the sentiment analysis results returned from the API.

[1784] Step 10:

[1785] Based on the analysis results, the server determines that the user's emotions are negative.

[1786] Input: Sentiment analysis results (sentiment score)

[1787] Output: Flag if negative sentiment detected

[1788] Specific operation: The server checks the emotion scores of "anger" and "sadness" from the analysis results.

[1789] Step 11:

[1790] The server will suspend sending the email and send a message to the device urging you to reconsider.

[1791] Input: Flag if negative sentiment detected

[1792] Output: A message to reconsider

[1793] Specific operation: The server sends the message "Do you want to send this email? Please reconsider" to the terminal as an HTTP response.

[1794] Step 12:

[1795] The terminal displays a message to the user urging him or her to reconsider.

[1796] Input: A message encouraging you to reconsider

[1797] Output: A reminder message to the user

[1798] Specific operation: The terminal displays the received message on the screen.

[1799] Step 13:

[1800] The user reconsiders and confirms sending the email again.

[1801] Input: The user's decision after reconsideration

[1802] Output: Action to send or unsend the email

[1803] Specific behavior: The user may reconsider and press the send button again, or cancel the send.

[1804] (Application example 2)

[1805] Next, a description will be given of Application Example 2 of Form 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".

[1806] Conventional email sending systems do not adequately check for and correct typos and do not control email sending based on the user's emotions, which means there is a high possibility of problems occurring due to erroneous sending or emotional email sending. In particular, sending emails with negative emotions is at high risk of causing misunderstandings and problems. This can lead to problems such as information leaks and deterioration of human relationships.

[1807] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1808] In this invention, the server includes a means for checking the destination domain and the presence or absence of a title before sending an email, a means for checking and correcting typos and omitted words in the email content using a natural language processing system, a means for displaying the corrected state in a preview, and a means for analyzing the user's emotions, and for suspending the sending of the email and encouraging reconsideration if negative emotions are detected. This makes it possible to correct typos and omitted words and to control the sending of email based on emotions.

[1809] "Verifying the destination domain" means checking whether the domain part of the recipient's email address is correct before the email is sent.

[1810] "Checking for the presence or absence of a title" means checking whether or not the subject of the email has been entered before the email is sent.

[1811] A "natural language processing system" is a technology that enables computers to understand, analyze, and generate human language.

[1812] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in the body of an email and correcting them to the correct form.

[1813] "Show in preview" means that the revised email content is displayed on the screen so that the user can check it.

[1814] "Analyzing emotions" means inferring and analyzing the emotional state of a user from the email text they enter.

[1815] "Negative emotions" refers to negative feelings such as anger or sadness.

[1816] "Pause email sending" means temporarily halting email sending if negative sentiment is detected.

[1817] "Encourage reconsideration" means encouraging the user to reexamine the content of the email and make corrections if necessary.

[1818] A system for implementing this invention includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, and means for analyzing the user's emotions and, if negative emotions are detected, pausing the sending of the email and encouraging reconsideration.

[1819] Program processing explanation

[1820] The server performs data processing and calculation using the following hardware and software.

[1821] Hardware and Software Used

[1822] Natural Language Processing Model: We load a Japanese natural language processing model using the spacy library and check the email body for typos.

[1823] Emotion Engine: Analyzes the emotions in the email body using the EmotionRecognizer class.

[1824] Data processing: The email body is tokenized, and if there are any typos, suggestions for corrections are provided.

[1825] Data calculation: Based on the results of sentiment analysis, suspend sending of emails if negative sentiment is detected.

[1826] Example of processing

[1827] As the user types the text of the email, the natural language processing system checks for typos in real time and suggests corrections. The corrected email content is displayed in a preview for the user to review. In addition, the emotion engine analyzes the sentiment of the email text, and if negative sentiment is detected, the email is paused before being sent and the user is urged to reconsider.

[1828] Examples of prompt statements

[1829] If a user types an email saying, "This project isn't moving forward at all! What's going on?", the emotion engine will detect "anger" and pause the email from being sent. The user will see a message saying, "Negative emotion detected. Email sending will be paused."

[1830] In this way, it is possible to correct typos and control email sending based on emotions, thereby preventing problems caused by sending emails incorrectly or based on emotion.

[1831] The flow of the identification process in the application example 2 will be described with reference to FIG.

[1832] Step 1:

[1833] The user enters the email body.

[1834] Input: The user enters the email body.

[1835] Specific behavior: A user enters the body of an email into a text area in an email client.

[1836] Step 2:

[1837] The terminal sends the email text to a natural language processing system to check for typos and misspellings.

[1838] Input: The email body entered by the user.

[1839] Data processing: A natural language processing system (spacy library) is used to tokenize the email body and detect typos.

[1840] Output: The email body with typos corrected.

[1841] Specific operation: The terminal sends the email text to a natural language processing system, which checks for and corrects typos. Correction suggestions are generated and presented to the user.

[1842] Step 3:

[1843] The device will display a preview of the revised email body.

[1844] Input: The email body with typos corrected.

[1845] Data processing: Convert the revised email body into a format for preview.

[1846] Output: The revised email body as displayed in the preview screen.

[1847] Specific operation: The device will display the revised email body on the preview screen so that the user can confirm it.

[1848] Step 4:

[1849] The device sends the email text to the emotion engine for emotion analysis.

[1850] Input: The revised email body.

[1851] Data calculation: Analyze the emotions in the email body using the emotion engine (EmotionRecognizer class).

[1852] Output: The result of the sentiment analysis (e.g. anger, sadness, etc.).

[1853] Specific operation: The device sends the corrected email text to the emotion engine for emotion analysis, and the analysis results are generated.

[1854] Step 5:

[1855] The device controls the sending of emails based on the results of sentiment analysis.

[1856] Input: Sentiment analysis results.

[1857] Data calculation: If negative emotions (e.g. anger, sadness) are detected, the sending of the email will be paused and the user will be prompted to reconsider.

[1858] Output: Email sending pause message.

[1859] Specific operation: The device checks the results of the emotion analysis, and if negative emotion is detected, it pauses sending of the email. The user is shown a message saying, "Negative emotion has been detected. Email sending will be paused."

[1860] Step 6:

[1861] The user reconsiders the content of the email and modifies it if necessary.

[1862] Input: Email sending pause message and revised email body.

[1863] Specific behavior: The user reviews the email, makes corrections if necessary, and then attempts to send the email again.

[1864] Step 7:

[1865] The terminal checks again to confirm the sending of the email, and if there are no problems, sends the email.

[1866] Input: The email body after reconsideration.

[1867] Data calculation: Check again for typos and perform sentiment analysis, and if there are no problems, send the email.

[1868] Output: The email sent.

[1869] Specific operation: The device checks the email body again and sends the email if there are no problems.

[1870] Example 3

[1871] Next, a third embodiment of the third embodiment will be described. 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".

[1872] Conventional email systems lacked the functionality to check the recipient's identifier or the presence or absence of a subject before sending, which led to occasional erroneous sending or missing subjects. Also, checking and correcting typos and omissions in document content was done manually, which was inefficient and placed a heavy burden on users. Furthermore, there was a lack of social consideration as there was no checking and correction of harassing expressions or expressions that take into consideration the Sustainable Development Goals. In addition, the quality of communication could decline as the expression in the document was not adjusted to reflect the user's feelings. A system that can solve these issues is needed.

[1873] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1874] In this invention, the server includes a means for confirming the identifier of the destination, a means for confirming the presence or absence of a subject, a means for confirming and correcting typos and omissions in the document content using a natural language processing system, a means for displaying the state after correction, and a means for analyzing the user's emotions and adjusting the expression of the document. This makes it possible to confirm the presence or absence of the identifier of the destination and the subject before sending, and to automatically correct typos and omissions. In addition, harassment-related expressions and expressions that take into consideration the sustainable development goals are confirmed and corrected, and the expression of the document can be adjusted to reflect the user's emotions. This allows for efficient and socially considerate communication.

[1875] The "destination identifier" is information for identifying the recipient of the email.

[1876] The "subject" is a title that briefly indicates the content of the email.

[1877] "Document content" is text information included in the body of the email.

[1878] A "natural language processing system" is a technology for analyzing text data and understanding its meaning.

[1879] A "typo" is a character string that is entered incorrectly instead of a correct character string.

[1880] An "omission" is when a necessary character is missing.

[1881] A "correction measure" is a method or device for correcting an error.

[1882] A "display means" is a method or device for visually presenting information.

[1883] "User emotion" refers to the psychological state of the user when composing an email.

[1884] The "means for analyzing emotions and adjusting the expression of a document" refers to a method or device for understanding the emotions of a user and changing the expression of an email accordingly.

[1885] "Harassing language" is any word or phrase that causes discomfort or disgust to others.

[1886] The "Sustainable Development Goals" are internationally agreed upon goals aimed at achieving sustainable development.

[1887] This invention is a system that has the functions of checking the recipient's identifier before sending an email, checking whether there is a subject, checking and correcting typos and omissions in the document content using a natural language processing system, displaying the state after correction, and analyzing the user's emotions to adjust the wording of the document.

[1888] Hardware and Software Used

[1889] The server uses the following hardware and software:

[1890] Hardware: Powerful processor, ample memory, and storage

[1891] Software: Python, NLTK library, Hugging Face Transformers library, scikit-learn, TensorFlow, TextBlob

[1892] A terminal is a device that allows a user to compose email and communicate with a server. This includes computers, smartphones, tablets, etc.

[1893] Program Processing

[1894] When a user composes an email and presses the send button, the device sends the email contents to the server. The server first checks whether the recipient identifier and the subject are included. It then uses a natural language processing system to analyze the document contents and detect and correct typos and omissions. This analysis uses Python's NLTK library and Hugging Face's Transformers library.

[1895] In addition, the server checks the document content for harassment and language that is sensitive to the Sustainable Development Goals, and makes corrections as necessary. The corrected document is then sent to the terminal and a preview is displayed to the user.

[1896] When a user provides feedback on the system's suggested corrections, the device sends the feedback to the server, which uses scikit-learn and TensorFlow to learn from the feedback and use it to check and correct typos and omissions in the future.

[1897] The server also uses an emotion engine to analyze the user's emotions. For this analysis, it uses an emotion analysis library (e.g., VADER or TextBlob). If the user is analyzed as having a positive emotion, the server automatically modifies the email's expression to be brighter and more lively. The modified document is sent to the terminal and previewed for the user.

[1898] When a user provides feedback on the emotion engine's suggestions, the device sends the feedback to the server, which uses TensorFlow to learn from the feedback and use it in future emotion recognition and document corrections.

[1899] Examples

[1900] For example, if a user writes "This project is completely pointless" in an email, the server will amend this to "This project has room for improvement" and display it in the preview. If a user writes "I think this project will be a success," the server will amend it to "I think this project will be a huge success!"

[1901] Examples of prompt statements

[1902] "Please check whether the content of this email contains any harassing language or language that takes into consideration the SDGs, and revise it as necessary."

[1903] "Please check this email for typos and correct them. Also, please learn from your user feedback and incorporate it into your next email."

[1904] “Adjust the content of this email according to the user’s emotions. If the user has positive emotions, make the message more upbeat.”

[1905] "Please check whether the content of this email accurately reflects the user's feelings and revise it if necessary. Also, please learn from the user's feedback and reflect it in future emails." The flow of the identification process in the third embodiment will be described with reference to FIG. 21.

[1906] Step 1:

[1907] The user composes an email and clicks the send button.

[1908] Input: The email content entered by the user

[1909] Output: The contents of the email are saved to the device.

[1910] Specific behavior: The user enters the body of the email and clicks the send button.

[1911] Step 2:

[1912] The terminal sends the contents of the email to the server.

[1913] Input: The content of the email the user creates

[1914] Output: The email content is sent to the server.

[1915] Specific operation: The terminal initiates communication to send the contents of the email to the server.

[1916] Step 3:

[1917] The server checks the presence of a destination identifier and a subject.

[1918] Input: The content of the email sent

[1919] Output: Destination identifier and subject existence check result

[1920] Specific operation: The server analyzes the header information of the email and checks whether it contains a destination identifier and a subject.

[1921] Step 4:

[1922] The server uses a natural language processing system to analyze the document content and detect and correct typos and omissions.

[1923] Input: Email body

[1924] Output: The modified email body

[1925] What it does: The server uses Python's NLTK library and Hugging Face's Transformers library to parse the body of the email, detect and correct typos and spelling errors.

[1926] Step 5:

[1927] The server will check the document content for harassment and language that takes into consideration the Sustainable Development Goals, and will make corrections if necessary.

[1928] Input: Modified email body

[1929] Output: Further modified email body

[1930] Specific actions: The server re-analyzes the body of the email to detect harassing language or language that is sensitive to the Sustainable Development Goals, and corrects it as necessary.

[1931] Step 6:

[1932] The server sends the corrected email to the terminal.

[1933] Input: Modified email body

[1934] Output: The corrected email body is sent to the device.

[1935] Specific operation: The server initiates communication to send the corrected email body to the terminal.

[1936] Step 7:

[1937] The terminal displays a preview of the corrected email to the user.

[1938] Input: The revised text of the email

[1939] Output: The email body as previewed to the user

[1940] Specific operation: The device displays the revised body of the email to the user.

[1941] Step 8:

[1942] The user provides feedback on the system's suggested modifications.

[1943] Input: User feedback

[1944] Output: Feedback is saved to your device

[1945] Specific action: The user enters feedback on the corrections and clicks the submit button.

[1946] Step 9:

[1947] The device sends the feedback to the server.

[1948] Input: User feedback

[1949] Output: Feedback is sent to the server

[1950] Specific operation: The device initiates communication to send feedback to the server.

[1951] Step 10:

[1952] The server learns the feedback and uses it to check for and correct typos and omissions in future posts.

[1953] Input: User feedback

[1954] Output: The trained model

[1955] Specific operation: The server uses scikit-learn and TensorFlow to learn the feedback and reflect it in checking and correcting typos and omissions from the next time onwards.

[1956] Step 11:

[1957] The server analyzes the user's emotions using an emotion engine.

[1958] Input: Email body

[1959] Output: User sentiment analysis results

[1960] Specific operation: The server uses a sentiment analysis library (e.g., VADER or TextBlob) to analyze the user's sentiment from the body of the email.

[1961] Step 12:

[1962] The server adjusts the wording of the email according to the user's emotions.

[1963] Input: User sentiment analysis results

[1964] Output: Adjusted email body

[1965] Specific operation: The server adjusts the wording of the email based on the results of the user's sentiment analysis.

[1966] Step 13:

[1967] The server sends the adjusted email to the terminal.

[1968] Input: The adjusted email body

[1969] Output: The adjusted email body is sent to the device.

[1970] Specific operation: The server initiates communication to send the adjusted email body to the terminal.

[1971] Step 14:

[1972] The device displays a preview of the adjusted email to the user.

[1973] Input: Adjusted email body

[1974] Output: The email body as previewed to the user

[1975] Specific operation: The device displays the adjusted email body to the user.

[1976] Step 15:

[1977] The user provides feedback on the emotion engine's suggestions.

[1978] Input: User feedback

[1979] Output: Feedback is saved to your device

[1980] Specific action: The user enters feedback on the emotion engine's suggestion and clicks the submit button.

[1981] Step 16:

[1982] The device sends the feedback to the server.

[1983] Input: User feedback

[1984] Output: Feedback is sent to the server

[1985] Specific operation: The device initiates communication to send feedback to the server.

[1986] Step 17:

[1987] The server learns from the feedback and uses it to improve emotion recognition and document correction in future tasks.

[1988] Input: User feedback

[1989] Output: The trained model

[1990] Specific operation: The server uses TensorFlow to learn the feedback and reflect it in future emotion recognition and document corrections.

[1991] (Application example 3)

[1992] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1993] Conventional email sending systems check and correct typos and omissions, but do not check and correct harassing expressions or expressions that take into consideration the Sustainable Development Goals, and do not adjust expressions according to the user's emotions. This makes it difficult to communicate appropriately and ethically when sending messages related to transactions or payments. In addition, the system lacks the ability to learn from user feedback and reflect it in checking and correcting typos and omissions and in emotion recognition for future messages, which results in low adaptability.

[1994] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1995] In this invention, the server includes: a means for checking the destination domain and checking whether the email has a title before sending the email; a means for checking and correcting typos and omissions in the email content using a natural language processing system; a means for displaying the corrected state in a preview; a means for checking and correcting harassing expressions and expressions that take the Sustainable Development Goals into consideration; a means for analyzing the user's emotions and adjusting the email expression according to the emotions; a means for learning feedback from the user and reflecting the results in checking and correcting typos and omissions from the next time onwards; a means for learning feedback from the user and reflecting the results in emotion recognition and email correction from the next time onwards; and a means for checking and correcting harassing expressions and expressions that take the Sustainable Development Goals into consideration when sending a message related to a transaction or payment, and adjusting the expression according to the user's emotions using an emotion engine. This enables appropriate and ethical communication when sending a message related to a transaction or payment.

[1996] "Verifying the destination domain" means checking whether the domain part of the recipient's email address is correct before the email is sent.

[1997] "Checking for the presence or absence of a title" means checking whether or not the title is set appropriately in the email before the email is sent.

[1998] A "natural language processing system" is a system that analyzes text data and checks for and corrects typos and omissions.

[1999] "Checking for and correcting typos and omissions" means detecting incorrect or missing characters in the email content and correcting them to the correct form.

[2000] "Displaying in preview" means displaying the revised email content on the screen so that the user can visually confirm it.

[2001] "Harassment-type expressions" refer to expressions that cause discomfort or disgust to others.

[2002] "Expression that takes into consideration the Sustainable Development Goals" refers to expression that takes into consideration environmental, social, and economic sustainability.

[2003] "Analyzing emotions" means reading and analyzing emotions from text entered by the user.

[2004] "Adjusting email expression according to emotion" means appropriately changing the content and tone of an email based on the user's emotions.

[2005] "Learning from feedback" means that the system incorporates opinions and evaluations provided by users and reflects them in future processing.

[2006] A "transactional or payment message" is a message that contains content related to a commercial transaction or the payment of money.

[2007] An "emotion engine" is an engine that analyzes a user's emotions and adjusts the content of the text based on those emotions.

[2008] A system for carrying out the present invention has the following configuration: The server includes means for checking the destination domain and whether or not there is a title before sending an email, means for checking and correcting typos and omissions in the email content using a natural language processing system, means for displaying the corrected state in a preview, means for checking and correcting harassing expressions and expressions that take into consideration the Sustainable Development Goals, means for analyzing a user's emotions and adjusting the expression of the email in accordance with the emotions, means for learning feedback from the user and reflecting the results in checking and correcting typos and omissions from the next time onwards, means for learning feedback from the user and reflecting the results in emotion recognition and email correction from the next time onwards, and means for checking and correcting harassing expressions and expressions that take into consideration the Sustainable Development Goals when sending a message related to a transaction or payment, and adjusting the expression in accordance with the user's emotions using an emotion engine.

[2009] Hardware and software used

[2010] Hardware: Smartphones, tablets

[2011] Software: Natural language processing libraries (e.g. spaCy, NLTK), sentiment analysis libraries (e.g. TextBlob, VADER), machine learning models (e.g. BERT, GPT-3)

[2012] Data processing and calculation

[2013] 1. Get input data:

[2014] A user enters a message within an electronic payment application.

[2015] Gets the message content in text format.

[2016] 2. Natural Language Processing:

[2017] Use a natural language processing library to analyze the message content.

[2018] Check whether the content contains any harassing language or language that takes into consideration the Sustainable Development Goals.

[2019] 3. Revision of expression:

[2020] If any such expressions are found, they will be amended to suit the needs of the user.

[2021] The revised message is displayed in a preview.

[2022] 4. Applying the Emotion Engine:

[2023] Analyze user sentiment using a sentiment analysis library.

[2024] Adjust the wording of your message depending on the user's emotions.

[2025] 5. Learning feedback:

[2026] Get feedback from users.

[2027] The feedback is trained into a machine learning model and reflected in future analyses and corrections.

[2028] Examples

[2029] Example prompts to be input to the generative AI model

[2030] User typed message: "This payment is late, please act quickly."

[2031] Emotion engine analysis result: "Anger"

[2032] New message: "We have confirmed that this payment is late. We would appreciate your prompt response."

[2033] In this way, by analyzing the messages entered by users and correcting them to appropriate expressions, communication regarding transactions and payments can be made smoother and more ethical.

[2034] The flow of the specific process in the application example 3 will be described with reference to FIG.

[2035] Step 1:

[2036] A user enters a message within an electronic payment application.

[2037] Input: The message text entered by the user.

[2038] Output: The message text is sent to the system.

[2039] Specific operation: A user uses a smartphone or tablet to enter a message regarding a transaction or payment. The entered message is sent in text form to the system.

[2040] Step 2:

[2041] The server checks the destination domain and whether or not the title exists.

[2042] Input: message text, destination email address.

[2043] Output: Domain and title check results.

[2044] Specific operation: The server analyzes the message text and the destination email address, checks whether the domain part is correct, and whether the subject is set appropriately. The result of the check is stored as internal data.

[2045] Step 3:

[2046] The server uses a natural language processing system to check for and correct typos and omissions.

[2047] Input: The message text.

[2048] Output: The modified message text.

[2049] Specific behavior: The server uses a natural language processing library (e.g. spaCy, NLTK) to analyze the message text and detect typos and spelling errors. Detected typos and spelling errors are corrected appropriately.

[2050] Step 4:

[2051] The server will display a preview of the revised message.

[2052] Input: The modified message text.

[2053] Output: The preview screen that the user sees.

[2054] Specific operation: The corrected message text is sent to the user's device and displayed on the preview screen. The user can check the corrected message.

[2055] Step 5:

[2056] The server will check for and edit harassing language and language that takes into consideration the Sustainable Development Goals.

[2057] Input: The modified message text.

[2058] Output: Further modified message text.

[2059] Specific operation: The server uses a natural language processing library to check whether the message text contains harassing language or language that takes into consideration the Sustainable Development Goals. If any such language is found, it is modified to an appropriate language.

[2060] Step 6:

[2061] The server uses an emotion engine to analyze the user's emotions and adjust the wording of the email according to the emotions.

[2062] Input: The modified message text.

[2063] Output: Message text adjusted according to the sentiment.

[2064] Specific operation: The server analyzes the user's sentiment using a sentiment analysis library (e.g. TextBlob, VADER), and adjusts the wording of the message based on the analysis result.

[2065] Step 7:

[2066] The server learns from user feedback and uses the results to check for and correct typos and omissions in future posts.

[2067] Input: User feedback.

[2068] Output: The learning results are reflected in the next and subsequent processing.

[2069] Specific operation: The server receives the feedback provided by the user and trains a machine learning model (e.g. BERT, GPT-3). The learning results are reflected in checking and correcting typos and omissions from the next time onwards.

[2070] Step 8:

[2071] The server learns from user feedback and uses the results to improve emotion recognition and email correction in future emails.

[2072] Input: User feedback.

[2073] Output: The learning results are reflected in the next and subsequent processing.

[2074] How it works: The server receives the feedback provided by the user and trains the machine learning model. The learning results are reflected in future emotion recognition and email corrections.

[2075] Step 9:

[2076] When the server sends messages related to transactions or payments, it checks and corrects any harassing language or language that takes into consideration the Sustainable Development Goals, and uses an emotion engine to adjust the language according to the user's emotions.

[2077] Input: Message text for the transaction or payment.

[2078] Output: The message text, modified appropriately.

[2079] How it works: The server analyzes messages related to transactions and payments, checks for and modifies harassing expressions and expressions that are considerate of the Sustainable Development Goals, and uses an emotion engine to adjust the expressions according to the user's emotions.

[2080] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2081] 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> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to 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.

[2082] Another example of generative AI is Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[2083] 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 smart glasses 214.

[2084] [Third embodiment]

[2085] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[2086] 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.

[2087] 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 wide area network (WAN) and / or a local area network (LAN).

[2088] 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.

[2089] 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 the voice according to instructions from the processor 46.

[2090] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[2091] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[2092] Fig. 6 shows an example of 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.

[2093] 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.

[2094] 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.

[2095] In the headset type terminal 314, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[2096] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[2097] "Example 1"

[2098] As one embodiment of the present invention, an email sending system is provided. This system has a function for checking the destination domain and whether or not there is a title when a user creates an email. Specifically, when a user inputs a destination of an email, the system automatically checks whether the destination domain is an internal domain or an external domain. Furthermore, if the title of the email has not been input, the system prompts the user to input a title.

[2099] "Example 2"

[2100] Furthermore, this system has the function of checking and correcting typos and omissions in the email content using a natural language processing system. Specifically, when a user enters the body of the email, the natural language processing system analyzes the content and checks for typos and omissions. If a typographical error or omission is found, the system corrects that part and displays the corrected state in a preview.

[2101] "Example 3"

[2102] The system also has the ability to check and correct harassment-type language and language that takes SDGs into consideration. Specifically, the natural language processing system checks the content of emails for harassment-type language and language that takes SDGs into consideration. If any such language is found, the system corrects the language and displays the corrected state in a preview.

[2103] "Example 4"

[2104] Furthermore, the system has the ability to learn from feedback from users and reflect the results in checking and correcting typos and omissions from the next time onwards. Specifically, when a user provides feedback on corrections suggested by the system, the natural language processing system learns from that feedback and reflects it in checking and correcting typos and omissions from the next time onwards. This allows the system to better adapt to the needs of the user.

[2105] The process flow of each embodiment will be described below.

[2106] "Example 1"

[2107] Step 1: The user enters the address to which the email should be sent.

[2108] Step 2: The system will automatically check whether the destination domain is an internal or external domain.

[2109] Step 3: If the email subject is not entered, the system will prompt the user to enter a subject.

[2110] "Example 2"

[2111] Step 1: The user enters the body of the email.

[2112] Step 2: The natural language processing system analyzes the content and checks for typos and omissions.

[2113] Step 3: If any typos or omissions are found, the system will correct them and display the corrected version in a preview.

[2114] "Example 3"

[2115] Step 1: The natural language processing system checks the content of the email for any harassing language or language that takes into consideration the SDGs.

[2116] Step 2: If a matching expression is found, the system corrects the expression and displays the corrected version in a preview.

[2117] "Example 4"

[2118] Step 1: The user provides feedback on the system's suggested fixes.

[2119] Step 2: The natural language processing system learns from the feedback and uses it to check for and correct typos and omissions in future sentences.

[2120] Example 1

[2121] Next, a description will be given of Example 1 of the embodiment 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".

[2122] In conventional email sending systems, users are often required to manually check the destination domain and whether the email title is correct, which is time-consuming. In addition, there is a risk of email quality declining due to insufficient checking and correction of typos, omissions, harassing expressions, and expressions that take into consideration the Sustainable Development Goals. Furthermore, there is no mechanism for reflecting user feedback in subsequent checks and corrections, which means that improvements to the system are not progressing.

[2123] 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.

[2124] In this invention, the server includes a means for the user to input the destination of the email, a means for the server to check the domain of the destination, a means for the server to check whether the email has a title, a means for the server to prompt the user to input a title, a means for checking and correcting typos and omissions in the email content using a natural language processing system, and a means for displaying the corrected state in a preview. This makes it possible to reduce the user's work and improve the quality of emails. In addition, by learning from user feedback and reflecting it in subsequent checks and corrections, the system can be continuously improved.

[2125] "User" means any individual or entity that uses the System to create and send Email.

[2126] A "server" is a computer system that accepts and processes input from users.

[2127] A "destination domain" is the portion of an email address following the @ symbol, and indicates the network to which the email is being sent.

[2128] "Email title" is the text entered as the subject of the email.

[2129] A "natural language processing system" is software or algorithms that analyze text data and check for and correct typos, omissions, and specific expressions.

[2130] "Checking for and correcting typos and omissions" refers to the process of detecting spelling mistakes and input errors in the content of an email and correcting them to the correct expression.

[2131] "Harassing language" is any word or phrase that may cause discomfort or offense to others.

[2132] "SDG-friendly expressions" are words and phrases that are ethical and socially sensitive and are in line with the Sustainable Development Goals (SDGs).

[2133] "Feedback" refers to opinions and evaluations provided by users, and is information that is useful for improving the system.

[2134] "Preview" is a display screen where the user can check the content of the email after correction.

[2135] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[2136] The present invention relates to an email sending system that checks the destination domain and whether or not there is a title when a user creates an email. When a user enters the destination of an email, the system automatically checks whether the destination domain is an internal domain or an external domain. If the title of the email has not been entered, the server prompts the user to enter a title.

[2137] Hardware and Software Used

[2138] This system is realized using the following hardware and software.

[2139] Server: A computer system that handles the central processing of the mail delivery system. It uses a Python script to verify domains.

[2140] Database: A database system for storing the internal domain list, specifically MySQL.

[2141] Terminal: The device on which the user composes the email. JavaScript running on the browser is used to check whether the title is included.

[2142] System Operation

[2143] When a user enters the destination of an email, the server extracts the domain part of the destination email address and checks it against the internal domain list, which is stored in the server's database. Next, if the user has not entered the subject of the email, the server prompts the user to enter a subject. This function is implemented on the browser using JavaScript.

[2144] Examples

[2145] For example, if a user enters the email address "example@external.com", the server will check that "external.com" is not in the internal domain list. Then, if the user clicks the submit button without entering a title, JavaScript will display an alert saying "Please enter a title."

[2146] Example prompts for generative AI models

[2147] "Please generate a program for an email sending system. When a user enters the destination of an email, the server checks whether the destination domain is an internal domain or an external domain, and if the subject of the email has not been entered, it prompts the user to enter a subject."

[2148] The above is an embodiment of the present invention. This system makes it possible to reduce the work for users and improve the quality of emails. In addition, the system can continuously improve by learning from user feedback and reflecting it in subsequent checks and corrections.

[2149] The flow of the identification process in the first embodiment will be described with reference to FIG.

[2150] Program processing flow

[2151] Step 1: User enters destination for email

[2152] The user enters the destination email address on the email composition screen. For example, they enter "example@external.com." This input is the starting point for the system's processing.

[2153] Input: The email address entered by the user (e.g. example@external.com)

[2154] Output: The entered email address is sent to the server.

[2155] Step 2: The server checks the destination domain

[2156] The server extracts the domain part from the email address it receives -- specifically, the part of the email address that comes after the @ sign -- and then it checks whether this domain is on its internal domain list, which is stored in the server's database.

[2157] Input: The email address entered by the user (e.g. example@external.com)

[2158] Data processing: Extraction of domain part (e.g. external.com)

[2159] Data calculation: Check against internal domain list

[2160] Output: Result of whether the domain is internal or external

[2161] Step 3: The server checks whether the email title exists

[2162] When the user clicks the send button, the server checks if the email subject field is empty. This check is done in the browser using JavaScript.

[2163] Input: Email title entered by the user (e.g. blank)

[2164] Data Calculation: Check if title field is empty

[2165] Output: Check whether a title has been entered.

[2166] Step 4: Server prompts user for title (if necessary)

[2167] If no subject is entered, the server will display an alert to the user and prompt them to enter a subject. The email will not be sent unless the user enters a subject.

[2168] Input: The result indicates that the title has not been entered.

[2169] Specific behavior: JavaScript displays an alert saying "Please enter a title"

[2170] Output: An alert prompting the user to enter a title.

[2171] Step 5: Check the email for typos and omissions using a natural language processing system and correct them.

[2172] The server passes the email content to a natural language processing system, which checks for and corrects typos and omissions. The corrected email content is then displayed to the user as a preview.

[2173] Input: Email content entered by the user

[2174] Data Calculation: Checking and correcting typos and omissions using natura...

Claims

1. A method to check the destination domain and whether the subject is present before sending an email, A means for checking and correcting typos and omissions in the contents of emails using a natural language processing system; A means for displaying the corrected state in a preview; A means to review and amend language that is harassing or considers the Sustainable Development Goals; and A means for analyzing a user's emotions and adjusting the expression of the email in accordance with the emotions; A means for learning feedback from the user and reflecting the learning results in checking and correcting typos and omissions in the next and subsequent times; A means for learning the feedback from the user and reflecting the learning result in subsequent emotion recognition and email corrections; A means for checking and correcting harassment-related or sustainable development goal-conscious language when sending a transaction or payment message, and adjusting the language according to the user's emotions using an emotion engine; A system including:

2. The system of claim 1 , wherein the natural language processing system checks and corrects expressions related to harassment or expressions that take into consideration the Sustainable Development Goals.

3. The system according to claim 1 , wherein the natural language processing system learns from the feedback from the user and reflects the learning results in checking and correcting typos and omissions from the next time onwards.

Images