System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving a sentence input by a user; means for analyzing the received sentence to identify stylistic and idiomatic peculiarities; means for generating a sentence with enhanced anonymity by removing the stylistic and idiomatic peculiarities based on a result of the analysis; means for transmitting the generated sentence to a terminal of the user; means for allowing the user to confirm and modify the generated sentence; and means for storing and publishing the modified final sentence in a database.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 an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

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

[0004] In CGM (Consumer Generated Media) services, it is common for users to use multiple accounts to avoid the risk of being identified by their particular writing style or phrasing. However, users often feel reluctant to post content for fear of being identified, which is one of the factors hindering the development of the service. Therefore, a system is needed that can remove particular writing styles and phrasing habits from text and generate text with increased anonymity so that users can use the service with peace of mind. [Means for solving the problem]

[0005] The present invention provides a system that receives text entered by a user, analyzes the text to identify writing style and phrasing habits, and generates text with enhanced anonymity based on the analysis results. Specifically, the system includes a means for receiving text entered by a user, a means for analyzing the received text to identify writing style and phrasing habits, a means for removing writing style and phrasing habits based on the analysis results and generating text with enhanced anonymity, a means for transmitting the generated text to the user's terminal, a means for the user to check and correct the generated text, and a means for storing the corrected final text in a database and making it public. The system also includes a means for anonymizing specific words and phrases in the generated text to increase the anonymity of the generated text, and a means for transmitting the text corrected by the user back to a server, and storing and making public the final text.

[0006] "User" refers to an individual or organization that posts content using the CGM Service.

[0007] "Text" refers to natural language text entered by the user in a language they are comfortable writing.

[0008] "Terminal" refers to a device such as a computer or smartphone used by a user.

[0009] "Server" refers to a computer system on a network that receives, analyzes, and processes data sent by users.

[0010] "Means for receiving" refers to the interface or module for receiving text entered by the user.

[0011] "Means for analyzing" refers to algorithms or programs that identify the stylistic and phrasing characteristics of the received text.

[0012] "Feature extraction" refers to the process of identifying distinctive stylistic and phrasing patterns in a user's writing.

[0013] "Means of generation" refers to algorithms or programs that extract characteristics of writing style and phrasing and then generate text with increased anonymity.

[0014] "Means for sending" refers to the interface or module for sending the generated text to the user's terminal via the network.

[0015] "Means for reviewing and correcting" refers to an interface or program that allows the user to review the generated text and make corrections as necessary.

[0016] "Database" refers to an information system for permanently storing posted text and data.

[0017] "Publication means" refers to the interface or program that allows other users to view the revised final text.

[0018] "Anonymization" refers to the process of transforming specific words or phrases in a text so that they cannot be identified. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] 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.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] 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.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] 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.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The present invention is a system that generates texts with enhanced anonymity by removing specific writing styles and phrasing habits in order to provide an environment in which users can post with peace of mind on CGM services. The system of the present invention analyzes texts entered by users and utilizes a generative AI model to remove specific writing styles and phrasing characteristics.

[0041] Below, the program processing of this system is explained in natural language, with specific examples included.

[0042] Program processing flow

[0043] 1. User Input

[0044] Users access the CGM service using their device and enter the text they want to post in the input form. For example, they might enter, "I went to a delicious cafe. The coffee was amazing!"

[0045] 2. Sending text

[0046] The terminal sends the text entered by the user to the server as is. The entered text is sent to the server using an HTTP request.

[0047] 3. Analysis and generation using AI models

[0048] The server passes the received text to a text analysis module. The text analysis module extracts features to identify the style and phrasing of the text. After feature extraction is complete, the analysis results are passed to a generative AI model. The generative AI model generates new text from which the identified style and phrasing features have been removed to increase anonymity. For example, it generates "I went to a new cafe. The coffee was very delicious."

[0049] 4. Sending the generated text

[0050] The server sends the generated text to the terminal, which returns it to the terminal as an HTTP response.

[0051] 5. User Review and Correction

[0052] The terminal displays the generated sentence received from the server to the user. The user can check the generated sentence and further modify the expressions and phrases as necessary. For example, the user can modify "very" to "really" to write "I went to a new cafe. The coffee was really delicious."

[0053] 6. Final Post

[0054] The user then sends the finalized text back to the server, which saves the revised final text in the CGM service database and makes it public. The published final text can then be viewed by other users.

[0055] Specific processing examples

[0056] Case: User B's post

[0057] User Input:

[0058] User B posts on social media, "I went to a new restaurant. The food was delicious."

[0059] Sending text:

[0060] The terminal sends this text to the server.

[0061] AI model analysis and generation:

[0062] The server receives the sentence, analyzes it with a text analysis module to identify stylistic and phrasing habits, and then the generative AI model generates the sentence, "I enjoyed eating at a new restaurant. The food was delicious."

[0063] Sending generated text:

[0064] The server sends the generated text to the terminal.

[0065] User review and correction:

[0066] User B checks the generated sentence and corrects "very" to "very." The final sentence is confirmed as "I enjoyed eating at the new restaurant. The food was delicious."

[0067] Last Post:

[0068] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[0069] This system allows users to post to the CGM service with peace of mind, avoiding identification based on specific writing styles or phrasing, which is expected to increase the number of users and page views of the service.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The user uses a device to access the input form of the CGM service and enters the text they want to post. For example, they might enter, "Today, we went to a new park and had a picnic. It was so much fun!"

[0073] Step 2:

[0074] When a user clicks the "Submit" button on the input form, the device sends the entered text to the server using an HTTP request, with the user's input data being sent to the server in the appropriate format.

[0075] Step 3:

[0076] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify its style and phrasing, specifically by analyzing the sentence structure and extracting specific phrases and expressions.

[0077] Step 4:

[0078] Based on the analysis results, the server passes the data to a generative AI model. The generative AI model takes the analysis results as input, removes stylistic and phrasing habits, and generates new sentences with increased anonymity. This process uses an algorithm that converts the sentences into different expressions while preserving the original meaning. For example, it generates the sentence, "Today we had a picnic at a new park. We had a lot of fun."

[0079] Step 5:

[0080] The server sends the generated new text to the device as an HTTP response, which includes the generated text and is displayed to the user.

[0081] Step 6:

[0082] The user checks the generated sentence on the device. If the user feels that a correction is necessary, the user corrects the generated sentence using the device interface. For example, changing "very" to "really." The final sentence is completed: "Today, we had a picnic at the new park. We really enjoyed it."

[0083] Step 7:

[0084] When the user confirms the final sentence, he clicks the "Confirm" button, and the terminal sends the corrected final sentence to the server again, again using an HTTP request.

[0085] Step 8:

[0086] The server receives the final text and stores it in a database, including any edits made by the user.

[0087] Step 9:

[0088] The server then publishes the final text on the CGM service, where it can be viewed by other users.

[0089] In this way, after going through a series of processes, the user's text is converted into a form that makes it difficult to identify and is then safely posted to the CGM service.

[0090] Example 1

[0091] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0092] In conventional CGM services, there is a risk that users can be identified by their unique writing style and phrasing in the text they post. This makes it difficult for users to post with confidence, which could lead to a decline in service usage. The lack of an appropriate system for increasing anonymity is an issue.

[0093] 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.

[0094] In this invention, the server includes: means for receiving text entered by a user; means for analyzing the received text and identifying writing style and phrasing habits; means for removing writing style and phrasing habits and generating text with increased anonymity based on the analysis results; means for sending the generated text to the user's terminal; means for the user to check and correct the generated text; means for saving and publishing the corrected final text in a database; a text analysis module for extracting features to identify specific writing style and phrasing habits; and generation AI model means for generating text with increased anonymity. This allows users to avoid being identified by their particular writing style or phrasing and to post to CGM services with peace of mind.

[0095] "User" refers to an individual or corporation that uses the CGM Service.

[0096] "Terminal" is a general term for the device used by a User to access and operate the CGM Service.

[0097] "Server" refers to a computer system that manages processing and data related to CGM services.

[0098] "Text" refers to the text data entered by a user to post on the CGM service.

[0099] "Means for receiving" refers to the function by which the server receives text sent by the user.

[0100] "Means for analysis" refers to the function for identifying the style and phrasing of the received text.

[0101] "Means of identification" refers to the function for extracting features in a text based on the analysis results.

[0102] "Means of generation" refers to the function for creating new text with increased anonymity based on the analysis results.

[0103] "Means for sending" refers to the function for returning the generated text to the user's device.

[0104] "Means for reviewing and correcting" refers to a function that allows the user to review the generated text and make corrections if necessary.

[0105] "Means for saving and publishing" refers to the functionality for saving the final revised text in a database and making it available for other users to view.

[0106] "Feature extraction" refers to the automated process of identifying specific stylistic or phrasing patterns in a piece of text.

[0107] "Text analysis module" refers to a software component for analyzing received text and extracting features.

[0108] A "generative AI model" refers to an artificial intelligence system that generates new text by removing identified stylistic and phrasing habits.

[0109] "CGM service" is a general term for online platforms that allow users to post and share content.

[0110] This invention is a system that allows users to generate sentences with enhanced anonymity by removing their specific writing style and phrasing. This system consists of the following main components:

[0111] User Input

[0112] A user accesses a CGM (Consumer Generated Media) service using a device and inputs a sentence into an input form. For example, consider a user who inputs the sentence, "I went to a delicious cafe. The coffee was great!"

[0113] Sending text

[0114] The device sends the text entered by the user to the server as an HTTP POST request, which is sent in JSON format and contains the text data.

[0115] Server analysis

[0116] The server passes the received text to a text analysis module, which uses natural language processing techniques (e.g., Spacy or NLTK) to extract features to identify the text's stylistic and phrasing habits.

[0117] Generation by generative AI model

[0118] The server passes the sentences after feature extraction to a generative AI model (such as OpenAI's GPT-3). The generative AI model generates new sentences that remove the identified stylistic and phrasing habits to enhance anonymity. For example, the sentence "I went to a delicious cafe. The coffee was great!" is converted to "I went to a new cafe. The coffee was really good."

[0119] Sending generated text

[0120] The server returns the generated text to the terminal as an HTTP response, with the appropriate Content-Type set in the response header.

[0121] User review and correction

[0122] The device displays the generated sentence to the user. The user can review the sentence and make corrections as necessary. For example, the user can change "very" to "really" to say, "I went to a new cafe. The coffee was really delicious."

[0123] Last Post

[0124] The user then sends the finalized sentence back to the server. The server saves the final, revised sentence in the CGM service's database and makes it publicly available for other users to view. For example, the published sentence might be, "I went to a new cafe. The coffee was really delicious."

[0125] This system allows users to post to CGM services with confidence, avoiding identification based on specific writing styles or phrasing. Below are some example prompts:

[0126] Example prompt sentence:

[0127] "Please make the following sentence more anonymous by removing certain styles and phrasing: We went to a new restaurant. The food was delicious."

[0128] By operating this system properly, it is expected that the number of service users and page views will increase.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] Step 1:

[0131] A user accesses the CGM service using a device and enters a sentence into the input form. For example, they might enter the sentence, "I went to a delicious cafe. The coffee was amazing!" This input becomes the starting point for the next process.

[0132] Step 2:

[0133] The device converts the input text into JSON format and sends it to the server as the payload of an HTTP POST request. Specifically, it generates a POST request and sends the input text data to the server. This request becomes the input for the analysis process.

[0134] Step 3:

[0135] The server analyzes the received HTTP POST request and extracts the text data contained therein, which becomes the input for the text analysis module.

[0136] Step 4:

[0137] The server runs a text analysis module and performs feature extraction to identify the style and phrasing of the text. Specifically, it analyzes the text using natural language processing techniques (such as Spacy or NLTK) and generates a feature vector. This feature vector becomes the input for the generative AI model.

[0138] Step 5:

[0139] The server passes the feature vector to a generative AI model, which generates new sentences. The generative AI model (such as OpenAI's GPT-3) removes the identified stylistic and phrasing habits and generates new sentences with increased anonymity. For example, the input "I went to a delicious cafe. The coffee was great!" yields the output "I went to a new cafe. The coffee was delicious." This new sentence becomes the input for the next step.

[0140] Step 6:

[0141] The server converts the generated new text into JSON format and returns it to the terminal as an HTTP response. Specifically, it generates an HTTP response and sends it with the generated text included in the payload. This response data becomes the input for the terminal.

[0142] Step 7:

[0143] The terminal parses the received HTTP response, extracts the generated text, and displays it to the user. The user can check the new text on the screen and make corrections in the text box if necessary. This corrected text becomes the input for the final post.

[0144] Step 8:

[0145] The user then sends the revised text back to the server. Specifically, the device converts the revised text back into JSON format and sends it to the server as an HTTP POST request. This request becomes the input for the final saving process.

[0146] Step 9:

[0147] The server saves the final text after receiving the corrections in a database and makes it available on the CGM service. Specifically, it performs a saving process, stores the text in a database, and makes it available for other users to view. This data is the final output.

[0148] This allows users to avoid being identified by their particular writing style or phrasing, allowing them to post to CGM services with peace of mind.

[0149] (Application example 1)

[0150] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0151] In conventional CGM services and electronic messaging systems, users often face the risk of being identified through their unique writing style and phrasing, and anonymity is often not guaranteed. Furthermore, phishing attacks are on the rise in email and messaging systems, increasing the likelihood that users will be victimized by fraudulent messages. This makes protecting user privacy and improving security even more important.

[0152] 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.

[0153] In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify stylistic and phrasing habits, means for generating text with increased anonymity by removing the stylistic and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, means for converting the stylistic and phrasing characteristics of the electronic message into standard expressions to prevent phishing attacks using received electronic messages, means for recording the converted electronic message in a log file, and means for transmitting the converted electronic message to the user. This not only increases the anonymity of text entered by the user but also makes it possible to prevent phishing attacks using electronic messages.

[0154] "User-entered text" refers to the text content entered by a user through a terminal or system.

[0155] "Means of receiving" refers to the function by which a server or system receives text sent by a user.

[0156] "Means of analyzing and identifying writing style and phrasing habits" refers to a function that uses machine learning models and natural language processing to analyze the writing style and unique expressions of input text and extract features.

[0157] The "means for generating sentences with enhanced anonymity" is a function that removes peculiar writing styles and phrases based on the characteristics obtained through analysis, and generates sentences in a standard writing style.

[0158] "Means for sending the generated text to the user's device" refers to a function that sends the highly anonymous text generated by the server back to the user's device.

[0159] "Means for checking and correcting" is a function that provides an interface for users to view the generated text and correct the content as necessary.

[0160] "Means for saving in a database and publishing" refers to the function of saving the final revised text on a server and publishing it in a form that other users can access.

[0161] "Means for converting stylistic and phrasing characteristics of electronic messages into standard expressions to prevent phishing attacks" refers to a function that converts the stylistic and idiosyncratic expressions of received electronic messages into standard business expressions, thereby preventing attacks before they occur.

[0162] "Means for recording to a log file" means a function that keeps a record of converted electronic messages for auditing and analysis.

[0163] A "means for sending an electronic message to a user" is a function that sends a converted secure electronic message to a user's device.

[0164] This invention is a system for preventing phishing attacks by anonymizing text so that users cannot be identified based on their unique writing style or phrasing habits. To realize this system, a program including the following means is required.

[0165] Program processing overview

[0166] 1. A means of receiving text entered by the user

[0167] Users access the system using a terminal and enter the text they want to post or the text of an email they received into the input form. For example, they could enter, "Please send us your bank details quickly! You'll receive a reward!"

[0168] 2. A means of analyzing received text and identifying stylistic and phrasing habits

[0169] The server receives the text sent by the user and analyzes it using Natural Language Processing (NLP) technology, which identifies stylistic and expressive features.

[0170] 3. A method for generating text with increased anonymity

[0171] The server uses a generative AI model to remove stylistic and phrasing quirks and convert the sentence into standard expressions. The generative AI model uses, for example, the OpenAI API. "Please send us your bank details quickly. We'll give you a reward!" is converted to "Could you please provide us with the following bank details as soon as possible? Thank you."

[0172] 4. A means of sending the generated text to the user's device

[0173] The server then sends the generated standard text back to the user's device using an HTTP response.

[0174] 5. A way for users to review and correct the generated text

[0175] The user can view the generated text received from the server and make further corrections as necessary. The user can freely edit the text displayed on the terminal.

[0176] 6. A means of storing and publishing the final revised text in a database

[0177] Once the user has finalized the text, it is sent back to the server and stored in the database, after which it is made publicly accessible to other users.

[0178] 7. Means of converting stylistic and phrasing features of electronic messages into standard expressions to prevent phishing attacks

[0179] The server converts the style and idioms of received emails into standard business style, particularly to prevent phishing attacks, for example converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[0180] 8. Means for recording converted electronic messages in a log file

[0181] All conversion processes are recorded in log files for later auditing and analysis, facilitating future problem solving and improvement planning.

[0182] 9. Means of sending the converted electronic message to the user

[0183] The server then sends the converted secure email to the user's device, again via an HTTP response.

[0184] Specific examples

[0185] As a specific example of use, if a security department staff member receives a suspicious email such as "Please send your bank information quickly. You will receive a reward!", the system will operate as follows.

[0186] Example prompt for a generative AI model:

[0187] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[0188] Submit your bank details quickly and you'll be rewarded!

[0189] This prompt provides users with standardized text generated by an AI model (e.g., "Could you please provide the following banking information as soon as possible? Thank you."), helping to prevent phishing attacks. The system achieves high levels of anonymity and security through communication between the server and the device, natural language processing, and the use of generative AI models.

[0190] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0191] Step 1:

[0192] The user inputs a sentence into the terminal. The input sentence is something like "Please send your bank information quickly. You will receive a reward!". The terminal sends this sentence to the server via an HTTP request.

[0193] Step 2:

[0194] The server analyzes the text received from the user. To do this, the text is passed to a natural language processing module to identify stylistic and phrasing patterns. This analysis is performed using text feature extraction algorithms.

[0195] Step 3:

[0196] Based on the analysis results, the server uses a generative AI model to generate new sentences that remove stylistic and phrasing habits. Specifically, the server passes the following prompt sentence as input to the generative AI model:

[0197] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[0198] Submit your bank details quickly and you'll be rewarded!

[0199] The output will be a secure sentence such as "Could you please provide the following bank details as soon as possible? Thank you."

[0200] Step 4:

[0201] The generated text is sent from the server to the terminal as an HTTP response, and the terminal displays the generated text to the user.

[0202] Step 5:

[0203] The user checks the generated text and makes any necessary corrections. The corrections are then sent back to the server as an HTTP request from the terminal.

[0204] Step 6:

[0205] The server saves the final corrected sentence in the database. The saved data is the final corrected sentence. For example, "Could you please provide the following bank information as soon as possible? Thank you." is saved in the database.

[0206] Step 7:

[0207] The server analyzes received electronic messages and converts stylistic and phrasing features into standard expressions to prevent phishing attacks, for example, converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[0208] Step 8:

[0209] The converted electronic messages are logged in a log file. The server stores the conversion process in a log file (e.g. "sanitized_emails_log.txt") for audit purposes.

[0210] Step 9:

[0211] Finally, the converted electronic message is sent from the server to the user's terminal and displayed to the user, thereby protecting the user from phishing attacks.

[0212] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0213] This invention relates to a system that generates text with enhanced anonymity by removing idiosyncratic writing styles and phrasing, in order to provide an environment in which users can post content safely on CGM services. It also includes an emotion engine that recognizes the emotions of user input and generated text and incorporates them into the system.

[0214] Below, the program processing of this system is explained in natural language, with specific examples included.

[0215] Program processing flow

[0216] 1. User Input

[0217] Users access the CGM service through their devices and enter the text they want to post in the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[0218] 2. Sending text

[0219] When the user clicks the "Send" button, the device sends the entered text to the server via an HTTP request.

[0220] 3. Analysis and generation using AI models and emotion engines

[0221] The server first passes the received text to a text analysis module, which identifies the text's style and phrasing. The analysis results and the text are then passed to an emotion engine to recognize the text's emotion.

[0222] Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate a more anonymous sentence. This process preserves the emotion of the original sentence while removing the stylistic and phrasing characteristics, resulting in a new sentence. For example, "Today was a really great day! I had a great time with my friends."

[0223] 4. Sending the generated text

[0224] The server sends the generated new text to the user's device as an HTTP response, where it is displayed.

[0225] 5. User Review and Correction

[0226] The user can review the generated sentence and make corrections as needed. The emotion engine recognizes and preserves emotions even as the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the sentence becomes, "Today was a wonderful day! I had a wonderful time with my friends."

[0227] 6. Final Post

[0228] Once the user has confirmed the final sentence, the device sends it back to the server, which then stores the final sentence in a database and makes it available on the CGM service.

[0229] Specific processing examples

[0230] Case: User B's post

[0231] User Input:

[0232] User B posts on social media, "I went to a new cafe. The coffee was delicious and the atmosphere was fun."

[0233] Sending text:

[0234] The terminal sends this text to the server.

[0235] Analysis and generation by AI model and emotion engine:

[0236] The server analyzes the text using a text analysis module to identify characteristics of the writing style and phrasing. Next, an emotion engine recognizes the emotion of the text. For example, positive emotion is recognized. The generative AI model reflects this emotion information and generates a response such as, "I visited a new cafe. The coffee was delicious and the atmosphere was very pleasant."

[0237] Sending generated text:

[0238] The server sends the generated text to the terminal.

[0239] User review and correction:

[0240] User B checks the generated sentence and corrects "very" to "extremely", completing the final sentence: "I visited a new cafe. The coffee was delicious and the atmosphere was very enjoyable."

[0241] Last Post:

[0242] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[0243] This system allows users to avoid being identified by their particular writing style or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of the emotion engine is expected to improve the user experience by maintaining naturalness in writing and consistency in emotions.

[0244] The processing flow will be explained below.

[0245] Step 1:

[0246] The user uses a device to access the input form of the CGM service and enters the sentence they want to post. For example, they might enter, "Today was beautiful weather, so I had a picnic in the park with my family. It was really fun!"

[0247] Step 2:

[0248] When a user clicks the "Submit" button on the input form, the terminal sends the input text to the server using an HTTP request. This request includes the user's input text.

[0249] Step 3:

[0250] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify stylistic and phrasing habits, such as the use of specific punctuation marks or frequently used phrases.

[0251] Step 4:

[0252] The server passes the analysis results to the emotion engine, which recognizes the emotion of the input text. For example, it determines that expressions such as "great weather" and "it was fun" contain positive emotions.

[0253] Step 5:

[0254] The emotion engine returns the recognized emotion information to the server, which then prepares the analysis results and emotion information to be input into the generative AI model.

[0255] Step 6:

[0256] The server inputs the analysis results and emotional information into the generative AI model. Based on this information, the generative AI model removes stylistic and phrasing habits and generates sentences with increased anonymity while preserving the original emotion. For example, it generates a sentence like, "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was a lot of fun."

[0257] Step 7:

[0258] The server sends the generated new text to the device as an HTTP response, which includes the generated text.

[0259] Step 8:

[0260] The device displays the generated text received from the server to the user, who then checks the text to see if it expresses the appropriate emotion.

[0261] Step 9:

[0262] The user can make necessary corrections to the generated sentences, for example, changing "very" to "extremely" to write "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was really fun."

[0263] Step 10:

[0264] When the user completes the edits and clicks the "Confirm" button, the device sends the final edited text back to the server, again using an HTTP request.

[0265] Step 11:

[0266] The server receives the final text and stores it in the CGM service's database, where it can be viewed by other users.

[0267] Step 12:

[0268] The server will then make the final text publicly available for other users to view on the CGM service, which will encourage users to feel more comfortable posting content and increase overall user numbers and engagement for the service.

[0269] In this way, the system of the present invention generates natural sentences that reflect emotions while eliminating particular writing styles and phrasing habits, providing an environment in which users can post to CGM services with peace of mind.

[0270] Example 2

[0271] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0272] In modern CGM services, users are required to maintain anonymity and post with peace of mind, avoiding identification based on specific writing styles or phrasing. However, existing systems have shortcomings in ensuring this anonymity, which can lead to issues with users not feeling secure enough. Furthermore, it is difficult to generate text that accurately reflects the sentiment of the user, which can lead to a poor user experience.

[0273] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text and identifying idiosyncrasies of writing style and phrasing, means for removing idiosyncrasies of writing style and phrasing based on the analysis results and generating text with increased anonymity, means for recognizing the emotion of the generated text and retaining that emotion, means for transmitting the generated text to the user's terminal, means for the user to confirm and correct the generated text, and means for saving the corrected final text in a database and making it public. This makes it possible to generate natural-looking text that reflects emotion while preventing identification based on idiosyncratic writing style and phrasing, allowing users to post their text to the CGM service with confidence.

[0274] "User" means a user who accesses the CGM Service and posts or views content.

[0275] A "terminal" is an electronic device used by a user, such as a computer, smartphone, or tablet.

[0276] A "server" is a computer system that receives a user's request, processes the data in accordance with the request, and returns the data.

[0277] "Text" refers to text data entered or generated by a user.

[0278] "Analysis" is a process for identifying the style and phrasing of the input text.

[0279] "Identifying" means identifying the specific stylistic and phrasing patterns in a piece of writing.

[0280] "Generation" involves removing identified stylistic and phrasing habits to create new sentences.

[0281] "Emotion recognition" means determining emotions such as positive, negative, or neutral from the content of a sentence.

[0282] "Preserving emotion" means maintaining the emotion of the original sentence in the generated new sentence.

[0283] "Send" means to transfer data from the server to the user's terminal.

[0284] "Verification" means that the user checks the generated text and confirms its content.

[0285] "Correction" means that the user makes changes to the generated text.

[0286] "Saving" means recording the final corrected text in the database.

[0287] "Publishing" means making the saved final text available for other users to view.

[0288] "Anonymity" means keeping users difficult to identify.

[0289] A "generative AI model" is an artificial intelligence program that uses natural language processing technology to generate new sentences.

[0290] A "prompt sentence" is text that gives instructions to the generative AI model on the sentence to output.

[0291] This invention is a system that can generate text with enhanced anonymity by removing a user's particular writing style and phrasing habits, and aims to provide an environment in which users can post content safely, particularly in CGM services. By combining a generative AI model and an emotion engine, this system neutralizes the writing style while recognizing and preserving the emotion of the text, ensuring anonymity.

[0292] First, the user accesses the CGM service using a device and enters the text they want to post into the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[0293] Next, when the user clicks the "Send" button, the device sends the input text as an HTTP request to the server, which then passes the received text to a text analysis module to identify stylistic and phrasing patterns. This analysis is performed using a Natural Language Processing (NLP) library (e.g., spaCy or NLTK).

[0294] Based on the analysis results, the server passes the sentence to an emotion engine, which uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to recognize the sentiment of the sentence, such as positive, negative, or neutral. Based on the obtained sentiment information and the identified stylistic features, the server uses a generative AI model (e.g., OpenAI's GPT-3) to generate a sentence with increased anonymity. This generation process uses a prompt such as: "Please neutralize the stylistic features while preserving the sentiment based on the original sentence."

[0295] The generated sentence might be something like, "Today was a really great day! I had a great time with my friends." This generated sentence is sent from the server to the user's device as an HTTP response.

[0296] The user can review the generated sentences on their device and make corrections as necessary. The emotion engine recognizes and retains emotions even when the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the user can write, "Today was a wonderful day! I had a wonderful time with my friends."

[0297] Finally, when the user confirms the final sentence, the device sends this sentence back to the server, which stores the final sentence in a database and makes it available on the CGM service.

[0298] This system allows users to avoid being identified by specific writing styles or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of an emotion engine and generative AI model maintains the naturalness of writing and consistency of emotion, which is expected to improve the user experience.

[0299] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0300] Step 1: The user accesses the CGM service through their device and enters the text they want to post in the input form. Specifically, they enter the text "Today was a great day! I had a great time with my friends" into the text box. The input is text data, and the output is the input text recorded on the device.

[0301] Step 2: When the user clicks the "Submit" button, the terminal sends the entered text to the server via an HTTP request. Specifically, the terminal packages the user's input in JSON format and sends it to the server's API endpoint as an HTTP POST request. The input is the entered text, and the output is the JSON data sent to the server.

[0302] Step 3: The server passes the received text to a text analysis module, which performs processing to identify stylistic and phrasing patterns. Specifically, the server uses a Natural Language Processing (NLP) library (e.g., spaCy or NLTK) to tokenize the text and extract contextual and stylistic features. The input is the JSON data sent in the HTTP request, and the output is a data structure containing the analysis results.

[0303] Step 4: The server passes the analysis results and the text to the emotion engine to recognize the sentiment of the text. Specifically, the server uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to identify the sentiment of the text, such as positive, negative, or neutral. The input is the text analysis result data, and the output is data with sentiment labels.

[0304] Step 5: Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate sentences with enhanced anonymity. Specifically, the server inputs a prompt to the generative AI model: "Please neutralize the style while preserving the emotion based on the original sentence." and obtains the generated sentence. The input is the original sentence data with emotion labels and the prompt sentence, and the output is the generated new sentence.

[0305] Step 6: The server sends the generated new text to the user's device as an HTTP response. Specifically, the generated text is included in the body of the HTTP response in JSON format and sent to the device. The input is the generated new text data, and the output is the JSON data sent to the device.

[0306] Step 7: The user checks the generated sentence on the terminal and corrects it if necessary. Specifically, the user looks at the generated sentence displayed on the terminal and corrects it, for example, from "really" to "very." The input is the generated sentence, and the output is the sentence corrected by the user.

[0307] Step 8: After the user confirms the final text edits, the device resends the text to the server. Specifically, the revised text is repackaged in JSON format and sent to the server's API endpoint as an HTTP POST request. The input is the text edited by the user, and the output is the JSON data resent to the server.

[0308] Step 9: The server saves the revised final sentence in the database and publishes it on the CGM service. Specifically, the server inserts the final sentence into the appropriate table in the database and sets it up for publication so that other users can view it. The input is the revised final sentence data, and the output is the data saved in the database and the sentence published on the CGM service.

[0309] (Application example 2)

[0310] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0311] Current online content generation services pose a risk of identifying individuals through their unique writing style and phrasing. Furthermore, when users want to express their feelings frankly, there is a lack of ways to generate text that maintains their emotions while also increasing anonymity. Therefore, there is a need to provide an environment where users can post reviews and comments with peace of mind.

[0312] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify writing style and phrasing habits, means for generating text with increased anonymity by removing writing style and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, emotion analysis means for analyzing the emotion of the text entered by the user, means for generating a prompt text based on the analyzed emotion information and generating a text that reflects the emotion while maintaining anonymity, means for anonymizing specific words and phrases to increase the anonymity of the generated text, and means for transmitting the final corrected text back to the user's terminal. This allows users to safely post text that reflects their emotions while avoiding identification based on their unique writing style and phrasing.

[0313] "Means for receiving text entered by a user" refers to a process or device that allows the server to receive text data sent from the user's terminal.

[0314] "Means for analyzing received text and identifying writing style and phrasing" refers to a process or program that uses natural language processing technology to analyze input text and extract characteristics of its writing style and expression.

[0315] "Means for generating text with enhanced anonymity by removing stylistic and phrasing characteristics" refers to a technology that uses the analysis results to generate new text with enhanced anonymity by eliminating specific stylistic and phrasing characteristics.

[0316] "Means for transmitting the generated text to the user's terminal" refers to a communication process or device for returning the text generated by the server to the user's terminal.

[0317] The "means for the user to check and correct the generated text" refers to an interface and function that allows the user to check the generated text and correct it if necessary.

[0318] "Means for storing and publishing the final revised text in a database" refers to the process or device by which the final revised text by the user is recorded in a database and published in a form accessible to other users.

[0319] "Sentiment analysis means for analyzing the sentiment of text entered by a user" is a natural language processing technology for analyzing the sentiment of input text and identifying sentiment categories such as positive or negative.

[0320] "A means for generating prompt sentences based on analyzed emotional information and generating sentences that reflect the emotions while maintaining anonymity" is a technology that generates sentences that maintain anonymity while retaining the emotions of the original sentences based on the results of emotional analysis.

[0321] "Means for anonymizing specific words and phrases to increase the anonymity of generated text" refers to a technology that increases anonymity by extracting words and phrases that could identify individuals from text and converting them into other words.

[0322] The "means for retransmitting the final revised text to the user's terminal" refers to a process or device that retransmits the final version of the text that the user has completed the revisions to the user's terminal.

[0323] System Program

[0324] A system for realizing this application example includes the following means.

[0325] Hardware and Software

[0326] Hardware: Smartphones, servers

[0327] Software: Python, Hugging Face Transformers library

[0328] Data processing and calculation

[0329] This system sends text entered by a user on a smartphone to a server, which then analyzes the text and identifies writing style and phrasing habits. It also uses a sentiment analysis tool to analyze the sentiment of the entered text and generates a prompt based on that sentiment information. A generative AI model then generates a sentence that reflects the sentiment while maintaining anonymity, and sends the sentence back to the user's device.

[0330] Processing flow

[0331] 1. User input:

[0332] Users use their smartphones to enter reviews into the food delivery app, which are then sent to the server as HTTP requests.

[0333] 2. Parsing on the server:

[0334] When a review arrives at the server, it is first analyzed by a text analysis module to identify its stylistic and phrasing characteristics, and then by a sentiment analysis module to analyze the review's sentiment (e.g., positive, negative, etc.).

[0335] 3. Sentence generation:

[0336] Based on this analysis, a generative AI model is used to generate prompt sentences that reflect the sentiment while maintaining anonymity. For example, a review with a positive sentiment, "The pizza I ordered was so delicious I ate two right away!", can be converted to "The pizza I ordered was so delicious I ate two right away!"

[0337] 4. User Verification and Correction:

[0338] The generated text is then sent back to the user's smartphone, where the user can review it and make any necessary corrections. The corrected text is then sent back to the server as the final version.

[0339] 5. Final saving and publishing:

[0340] The final edited text is stored in a database and published as a review of the food delivery app that can be viewed by other users.

[0341] Specific examples

[0342] Prompt Sentence Examples

[0343] "Transform the original sentence into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right there!"

[0344] This system can be applied not only to food delivery apps but also to other online content generation services, providing an environment where users can post reviews and comments with confidence. It is also expected to improve the user experience by maintaining naturalness and emotional consistency in the writing.

[0345] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0346] Step 1:

[0347] Input: A user uses their smartphone to input a review into a food delivery app.

[0348] Action: For example, type, "The pizza I ordered was so good I ate two right there!"

[0349] Output: This entered review is sent by the device to the server as an HTTP request.

[0350] Step 2:

[0351] Input: Reviews received by the server.

[0352] How it works: The server first passes the review to a text analysis module to analyze its writing style and phrasing, and also uses a sentiment analysis module to analyze the review's sentiment (positive, negative, etc.).

[0353] Output: Stylistic and phrasing characteristics, as well as sentiment information.

[0354] Step 3:

[0355] Input: Stylistic and phrasing characteristics, emotional information.

[0356] What it does: The server generates a prompt based on this information, such as "Transform the original text into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right away!"

[0357] Output: The prompt statement and any parsing information needed to generate it.

[0358] Step 4:

[0359] Input: Prompt statement and parsing information.

[0360] How it works: The server uses a generative AI model to generate new, sentiment-based sentences while preserving anonymity, such as "The pizza I ordered was so delicious, I had two right away!"

[0361] Output: The new sentence generated.

[0362] Step 5:

[0363] Input: The new sentence to be generated.

[0364] What it does: The server resends this generated text to the user's device as an HTTP response.

[0365] Output: The generated text is displayed on the user's smartphone.

[0366] Step 6:

[0367] Input: User-generated text verification.

[0368] How it works: The user reviews the generated sentence and makes corrections if necessary, for example changing "very" to "really" to get "The pizza I ordered was so good, I had two right away!"

[0369] Output: The sentence corrected by the user.

[0370] Step 7:

[0371] Input: Final corrected text.

[0372] How it works: The user's device sends the revised text back to the server, which stores this final text in its database and makes it publicly available for other users to view.

[0373] Output: The final sentence stored in the database.

[0374] By following these steps, users can confidently submit reviews that reflect their feelings while avoiding being identified by their unique writing style or phrasing.

[0375] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0376] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0377] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0378] [Second embodiment]

[0379] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0380] 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.

[0381] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0382] 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.

[0383] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0384] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0385] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0386] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0387] 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.

[0388] 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.

[0389] 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.

[0390] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0391] The present invention is a system that generates texts with enhanced anonymity by removing specific writing styles and phrasing habits in order to provide an environment in which users can post with peace of mind on CGM services. The system of the present invention analyzes texts entered by users and utilizes a generative AI model to remove specific writing styles and phrasing characteristics.

[0392] Below, the program processing of this system is explained in natural language, with specific examples included.

[0393] Program processing flow

[0394] 1. User Input

[0395] Users access the CGM service using their device and enter the text they want to post in the input form. For example, they might enter, "I went to a delicious cafe. The coffee was amazing!"

[0396] 2. Sending text

[0397] The terminal sends the text entered by the user to the server as is. The entered text is sent to the server using an HTTP request.

[0398] 3. Analysis and generation using AI models

[0399] The server passes the received text to a text analysis module. The text analysis module extracts features to identify the style and phrasing of the text. After feature extraction is complete, the analysis results are passed to a generative AI model. The generative AI model generates new text that removes the identified style and phrasing features to increase anonymity. For example, it generates "I went to a new cafe. The coffee was very delicious."

[0400] 4. Sending the generated text

[0401] The server sends the generated text to the terminal, which returns it to the terminal as an HTTP response.

[0402] 5. User Review and Correction

[0403] The terminal displays the generated sentence received from the server to the user. The user can check the generated sentence and further modify the expressions and phrases as necessary. For example, the user can modify "very" to "really" to write "I went to a new cafe. The coffee was really delicious."

[0404] 6. Final Post

[0405] The user then sends the finalized text back to the server, which saves the revised final text in the CGM service database and makes it public. The published final text can then be viewed by other users.

[0406] Specific processing examples

[0407] Case: User B's post

[0408] User Input:

[0409] User B posts on social media, "I went to a new restaurant. The food was delicious."

[0410] Sending text:

[0411] The terminal sends this text to the server.

[0412] AI model analysis and generation:

[0413] The server receives the sentence, analyzes it with a text analysis module to identify stylistic and phrasing habits, and then the generative AI model generates the sentence, "I enjoyed eating at a new restaurant. The food was delicious."

[0414] Sending generated text:

[0415] The server sends the generated text to the terminal.

[0416] User review and correction:

[0417] User B checks the generated sentence and corrects "very" to "very." The final sentence is confirmed as "I enjoyed eating at the new restaurant. The food was delicious."

[0418] Last Post:

[0419] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[0420] This system allows users to post to the CGM service with peace of mind, avoiding identification based on specific writing styles or phrasing, which is expected to increase the number of users and page views of the service.

[0421] The processing flow will be explained below.

[0422] Step 1:

[0423] The user uses a device to access the input form of the CGM service and enters the text they want to post. For example, they might enter, "Today, we went to a new park and had a picnic. It was so much fun!"

[0424] Step 2:

[0425] When a user clicks the "Submit" button on the input form, the device sends the entered text to the server using an HTTP request, with the user's input data being sent to the server in the appropriate format.

[0426] Step 3:

[0427] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify its style and phrasing, specifically by analyzing the sentence structure and extracting specific phrases and expressions.

[0428] Step 4:

[0429] Based on the analysis results, the server passes the data to a generative AI model. The generative AI model takes the analysis results as input, removes stylistic and phrasing habits, and generates new sentences with increased anonymity. This process uses an algorithm that converts the sentences into different expressions while preserving the original meaning. For example, it generates the sentence, "Today we had a picnic at a new park. We had a lot of fun."

[0430] Step 5:

[0431] The server sends the generated new text to the device as an HTTP response, which includes the generated text and is displayed to the user.

[0432] Step 6:

[0433] The user checks the generated sentence on the device. If the user feels that a correction is necessary, the user corrects the generated sentence using the device interface. For example, changing "very" to "really." The final sentence is completed: "Today, we had a picnic at the new park. We really enjoyed it."

[0434] Step 7:

[0435] When the user confirms the final sentence, he clicks the "Confirm" button, and the terminal sends the corrected final sentence to the server again, again using an HTTP request.

[0436] Step 8:

[0437] The server receives the final text and stores it in a database, including any edits made by the user.

[0438] Step 9:

[0439] The server then publishes the final text on the CGM service, where it can be viewed by other users.

[0440] In this way, after going through a series of processes, the user's text is converted into a form that makes it difficult to identify and is then safely posted to the CGM service.

[0441] Example 1

[0442] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0443] In conventional CGM services, there is a risk that users can be identified by their unique writing style and phrasing in the text they post. This makes it difficult for users to post with confidence, which could lead to a decline in service usage. The lack of an appropriate system for increasing anonymity is an issue.

[0444] 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.

[0445] In this invention, the server includes: means for receiving text entered by a user; means for analyzing the received text and identifying writing style and phrasing habits; means for removing writing style and phrasing habits and generating text with increased anonymity based on the analysis results; means for sending the generated text to the user's terminal; means for the user to check and correct the generated text; means for saving and publishing the corrected final text in a database; a text analysis module for extracting features to identify specific writing style and phrasing habits; and generation AI model means for generating text with increased anonymity. This allows users to avoid being identified by their particular writing style or phrasing and to post to CGM services with peace of mind.

[0446] "User" refers to an individual or corporation that uses the CGM Service.

[0447] "Terminal" is a general term for the device used by a User to access and operate the CGM Service.

[0448] "Server" refers to a computer system that manages processing and data related to CGM services.

[0449] "Text" refers to the text data entered by a user to post on the CGM service.

[0450] "Means for receiving" refers to the function by which the server receives text sent by the user.

[0451] "Means for analysis" refers to the function for identifying the style and phrasing of the received text.

[0452] "Means of identification" refers to the function for extracting features in a text based on the analysis results.

[0453] "Means of generation" refers to the function for creating new text with increased anonymity based on the analysis results.

[0454] "Means for sending" refers to the function for returning the generated text to the user's device.

[0455] "Means for reviewing and correcting" refers to a function that allows the user to review the generated text and make corrections if necessary.

[0456] "Means for saving and publishing" refers to the functionality for saving the final revised text in a database and making it available for other users to view.

[0457] "Feature extraction" refers to the automated process of identifying specific stylistic or phrasing patterns in a piece of text.

[0458] "Text analysis module" refers to a software component for analyzing received text and extracting features.

[0459] A "generative AI model" refers to an artificial intelligence system that generates new text by removing identified stylistic and phrasing habits.

[0460] "CGM service" is a general term for online platforms that allow users to post and share content.

[0461] This invention is a system that allows users to generate sentences with enhanced anonymity by removing their specific writing style and phrasing. This system consists of the following main components:

[0462] User Input

[0463] A user accesses a CGM (Consumer Generated Media) service using a device and inputs a sentence into an input form. For example, consider a user who inputs the sentence, "I went to a delicious cafe. The coffee was great!"

[0464] Sending text

[0465] The device sends the text entered by the user to the server as an HTTP POST request, which is sent in JSON format and contains the text data.

[0466] Server analysis

[0467] The server passes the received text to a text analysis module, which uses natural language processing techniques (e.g., Spacy or NLTK) to extract features to identify the text's stylistic and phrasing habits.

[0468] Generation by generative AI model

[0469] The server passes the sentences after feature extraction to a generative AI model (such as OpenAI's GPT-3). The generative AI model generates new sentences that remove the identified stylistic and phrasing habits to enhance anonymity. For example, the sentence "I went to a delicious cafe. The coffee was great!" is converted to "I went to a new cafe. The coffee was really good."

[0470] Sending generated text

[0471] The server returns the generated text to the terminal as an HTTP response, with the appropriate Content-Type set in the response header.

[0472] User review and correction

[0473] The device displays the generated sentence to the user. The user can review the sentence and make corrections as necessary. For example, the user can change "very" to "really" to say, "I went to a new cafe. The coffee was really delicious."

[0474] Last Post

[0475] The user then sends the finalized sentence back to the server. The server saves the final, revised sentence in the CGM service's database and makes it publicly available for other users to view. For example, the published sentence might be, "I went to a new cafe. The coffee was really delicious."

[0476] This system allows users to post to CGM services with confidence, avoiding identification based on specific writing styles or phrasing. Below are some example prompts:

[0477] Example prompt sentence:

[0478] "Please make the following sentence more anonymous by removing certain styles and phrasing: We went to a new restaurant. The food was delicious."

[0479] By operating this system properly, it is expected that the number of service users and page views will increase.

[0480] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0481] Step 1:

[0482] A user accesses the CGM service using a device and enters a sentence into the input form. For example, they might enter the sentence, "I went to a delicious cafe. The coffee was amazing!" This input becomes the starting point for the next process.

[0483] Step 2:

[0484] The device converts the input text into JSON format and sends it to the server as the payload of an HTTP POST request. Specifically, it generates a POST request and sends the input text data to the server. This request becomes the input for the analysis process.

[0485] Step 3:

[0486] The server analyzes the received HTTP POST request and extracts the text data contained therein, which becomes the input for the text analysis module.

[0487] Step 4:

[0488] The server runs a text analysis module and performs feature extraction to identify the style and phrasing of the text. Specifically, it analyzes the text using natural language processing techniques (such as Spacy or NLTK) and generates a feature vector. This feature vector becomes the input for the generative AI model.

[0489] Step 5:

[0490] The server passes the feature vector to a generative AI model, which generates new sentences. The generative AI model (such as OpenAI's GPT-3) removes the identified stylistic and phrasing habits and generates new sentences with increased anonymity. For example, the input "I went to a delicious cafe. The coffee was great!" yields the output "I went to a new cafe. The coffee was delicious." This new sentence becomes the input for the next step.

[0491] Step 6:

[0492] The server converts the generated new text into JSON format and returns it to the terminal as an HTTP response. Specifically, it generates an HTTP response and sends it with the generated text included in the payload. This response data becomes the input for the terminal.

[0493] Step 7:

[0494] The terminal parses the received HTTP response, extracts the generated text, and displays it to the user. The user can check the new text on the screen and make corrections in the text box if necessary. This corrected text becomes the input for the final post.

[0495] Step 8:

[0496] The user then sends the revised text back to the server. Specifically, the device converts the revised text back into JSON format and sends it to the server as an HTTP POST request. This request becomes the input for the final saving process.

[0497] Step 9:

[0498] The server saves the final text after receiving the corrections in a database and makes it available on the CGM service. Specifically, it performs a saving process, stores the text in a database, and makes it available for other users to view. This data is the final output.

[0499] This allows users to avoid being identified by their particular writing style or phrasing, allowing them to post to CGM services with peace of mind.

[0500] (Application example 1)

[0501] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0502] In conventional CGM services and electronic messaging systems, users often face the risk of being identified through their unique writing style and phrasing, and anonymity is often not guaranteed. Furthermore, phishing attacks are on the rise in email and messaging systems, increasing the likelihood that users will be victimized by fraudulent messages. This makes protecting user privacy and improving security even more important.

[0503] 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.

[0504] In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify stylistic and phrasing habits, means for generating text with increased anonymity by removing the stylistic and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, means for converting the stylistic and phrasing characteristics of the electronic message into standard expressions to prevent phishing attacks using received electronic messages, means for recording the converted electronic message in a log file, and means for transmitting the converted electronic message to the user. This not only increases the anonymity of text entered by the user but also makes it possible to prevent phishing attacks using electronic messages.

[0505] "User-entered text" refers to the text content entered by a user through a terminal or system.

[0506] "Means of receiving" refers to the function by which a server or system receives text sent by a user.

[0507] "Means of analyzing and identifying writing style and phrasing habits" refers to a function that uses machine learning models and natural language processing to analyze the writing style and unique expressions of input text and extract features.

[0508] The "means for generating sentences with enhanced anonymity" is a function that removes peculiar writing styles and phrases based on the characteristics obtained through analysis, and generates sentences in a standard writing style.

[0509] "Means for sending the generated text to the user's device" refers to a function that sends the highly anonymous text generated by the server back to the user's device.

[0510] "Means for checking and correcting" is a function that provides an interface for users to view the generated text and correct the content as necessary.

[0511] "Means for saving in a database and publishing" refers to the function of saving the final revised text on a server and publishing it in a form that other users can access.

[0512] "Means for converting stylistic and phrasing characteristics of electronic messages into standard expressions to prevent phishing attacks" refers to a function that converts the stylistic and idiosyncratic expressions of received electronic messages into standard business expressions, thereby preventing attacks before they occur.

[0513] "Means for recording to a log file" means a function that keeps a record of converted electronic messages for auditing and analysis.

[0514] A "means for sending an electronic message to a user" is a function that sends a converted secure electronic message to a user's device.

[0515] This invention is a system for preventing phishing attacks by anonymizing text so that users cannot be identified based on their unique writing style or phrasing habits. To realize this system, a program including the following means is required.

[0516] Program processing overview

[0517] 1. A means of receiving text entered by the user

[0518] Users access the system using a terminal and enter the text they want to post or the text of an email they received into the input form. For example, they could enter, "Please send us your bank details quickly! You'll receive a reward!"

[0519] 2. A means of analyzing received text and identifying stylistic and phrasing habits

[0520] The server receives the text sent by the user and analyzes it using Natural Language Processing (NLP) technology, which identifies stylistic and expressive features.

[0521] 3. A method for generating text with increased anonymity

[0522] The server uses a generative AI model to remove stylistic and phrasing quirks and convert the sentence into standard expressions. The generative AI model uses, for example, the OpenAI API. "Please send us your bank details quickly. We'll give you a reward!" is converted to "Could you please provide us with the following bank details as soon as possible? Thank you."

[0523] 4. A means of sending the generated text to the user's device

[0524] The server then sends the generated standard text back to the user's device using an HTTP response.

[0525] 5. A way for users to review and correct the generated text

[0526] The user can view the generated text received from the server and make further corrections as necessary. The user can freely edit the text displayed on the terminal.

[0527] 6. A means of storing and publishing the final revised text in a database

[0528] Once the user has finalized the text, it is sent back to the server and stored in the database, after which it is made publicly accessible to other users.

[0529] 7. Means of converting stylistic and phrasing features of electronic messages into standard expressions to prevent phishing attacks

[0530] The server converts the style and idioms of received emails into standard business style, particularly to prevent phishing attacks, for example converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[0531] 8. Means for recording converted electronic messages in a log file

[0532] All conversion processes are recorded in log files for later auditing and analysis, facilitating future problem solving and improvement planning.

[0533] 9. Means of sending the converted electronic message to the user

[0534] The server then sends the converted secure email to the user's device, again via an HTTP response.

[0535] Specific examples

[0536] As a specific example of use, if a security department staff member receives a suspicious email such as "Please send your bank information quickly. You will receive a reward!", the system will operate as follows.

[0537] Example prompt for a generative AI model:

[0538] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[0539] Submit your bank details quickly and you'll be rewarded!

[0540] This prompt provides users with standardized text generated by an AI model (e.g., "Could you please provide the following banking information as soon as possible? Thank you."), helping to prevent phishing attacks. The system achieves high levels of anonymity and security through communication between the server and the device, natural language processing, and the use of generative AI models.

[0541] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0542] Step 1:

[0543] The user inputs a sentence into the terminal. The input sentence is something like "Please send your bank information quickly. You will receive a reward!". The terminal sends this sentence to the server via an HTTP request.

[0544] Step 2:

[0545] The server analyzes the text received from the user. To do this, the text is passed to a natural language processing module to identify stylistic and phrasing patterns. This analysis is performed using text feature extraction algorithms.

[0546] Step 3:

[0547] Based on the analysis results, the server uses a generative AI model to generate new sentences that remove stylistic and phrasing habits. Specifically, the server passes the following prompt sentence as input to the generative AI model:

[0548] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[0549] Submit your bank details quickly and you'll be rewarded!

[0550] The output will be a secure sentence such as "Could you please provide the following bank details as soon as possible? Thank you."

[0551] Step 4:

[0552] The generated text is sent from the server to the terminal as an HTTP response, and the terminal displays the generated text to the user.

[0553] Step 5:

[0554] The user checks the generated text and makes any necessary corrections. The corrections are then sent back to the server as an HTTP request from the terminal.

[0555] Step 6:

[0556] The server saves the final corrected sentence in the database. The saved data is the final corrected sentence. For example, "Could you please provide the following bank information as soon as possible? Thank you." is saved in the database.

[0557] Step 7:

[0558] The server analyzes received electronic messages and converts stylistic and phrasing features into standard expressions to prevent phishing attacks, for example, converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[0559] Step 8:

[0560] The converted electronic messages are logged in a log file. The server stores the conversion process in a log file (e.g. "sanitized_emails_log.txt") for audit purposes.

[0561] Step 9:

[0562] Finally, the converted electronic message is sent from the server to the user's terminal and displayed to the user, thereby protecting the user from phishing attacks.

[0563] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0564] This invention relates to a system that generates text with enhanced anonymity by removing idiosyncratic writing styles and phrasing, in order to provide an environment in which users can post content safely on CGM services. It also includes an emotion engine that recognizes the emotions of user input and generated text and incorporates them into the system.

[0565] Below, the program processing of this system is explained in natural language, with specific examples included.

[0566] Program processing flow

[0567] 1. User Input

[0568] Users access the CGM service through their devices and enter the text they want to post in the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[0569] 2. Sending text

[0570] When the user clicks the "Send" button, the device sends the entered text to the server via an HTTP request.

[0571] 3. Analysis and generation using AI models and emotion engines

[0572] The server first passes the received text to a text analysis module, which identifies the text's style and phrasing. The analysis results and the text are then passed to an emotion engine to recognize the text's emotion.

[0573] Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate a more anonymous sentence. This process preserves the emotion of the original sentence while removing the stylistic and phrasing characteristics, resulting in a new sentence. For example, "Today was a really great day! I had a great time with my friends."

[0574] 4. Sending the generated text

[0575] The server sends the generated new text to the user's device as an HTTP response, where it is displayed.

[0576] 5. User Review and Correction

[0577] The user can review the generated sentence and make corrections as needed. The emotion engine recognizes and preserves emotions even as the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the sentence becomes, "Today was a wonderful day! I had a wonderful time with my friends."

[0578] 6. Final Post

[0579] Once the user has confirmed the final sentence, the device sends it back to the server, which then stores the final sentence in a database and makes it available on the CGM service.

[0580] Specific processing examples

[0581] Case: User B's post

[0582] User Input:

[0583] User B posts on social media, "I went to a new cafe. The coffee was delicious and the atmosphere was fun."

[0584] Sending text:

[0585] The terminal sends this text to the server.

[0586] Analysis and generation by AI model and emotion engine:

[0587] The server analyzes the text using a text analysis module to identify characteristics of the writing style and phrasing. Next, an emotion engine recognizes the emotion of the text. For example, positive emotion is recognized. The generative AI model reflects this emotion information and generates a response such as, "I visited a new cafe. The coffee was delicious and the atmosphere was very pleasant."

[0588] Sending generated text:

[0589] The server sends the generated text to the terminal.

[0590] User review and correction:

[0591] User B checks the generated sentence and corrects "very" to "extremely", completing the final sentence: "I visited a new cafe. The coffee was delicious and the atmosphere was very enjoyable."

[0592] Last Post:

[0593] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[0594] This system allows users to avoid being identified by their particular writing style or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of the emotion engine is expected to improve the user experience by maintaining naturalness in writing and consistency in emotions.

[0595] The processing flow will be explained below.

[0596] Step 1:

[0597] The user uses a device to access the input form of the CGM service and enters the sentence they want to post. For example, they might enter, "Today was beautiful weather, so I had a picnic in the park with my family. It was really fun!"

[0598] Step 2:

[0599] When a user clicks the "Submit" button on the input form, the terminal sends the input text to the server using an HTTP request. This request includes the user's input text.

[0600] Step 3:

[0601] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify stylistic and phrasing habits, such as the use of specific punctuation marks or frequently used phrases.

[0602] Step 4:

[0603] The server passes the analysis results to the emotion engine, which recognizes the emotion of the input text. For example, it determines that expressions such as "great weather" and "it was fun" contain positive emotions.

[0604] Step 5:

[0605] The emotion engine returns the recognized emotion information to the server, which then prepares the analysis results and emotion information to be input into the generative AI model.

[0606] Step 6:

[0607] The server inputs the analysis results and emotional information into the generative AI model. Based on this information, the generative AI model removes stylistic and phrasing habits and generates sentences with increased anonymity while preserving the original emotion. For example, it generates a sentence like, "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was a lot of fun."

[0608] Step 7:

[0609] The server sends the generated new text to the device as an HTTP response, which includes the generated text.

[0610] Step 8:

[0611] The device displays the generated text received from the server to the user, who then checks the text to see if it expresses the appropriate emotion.

[0612] Step 9:

[0613] The user can make necessary corrections to the generated sentences, for example, changing "very" to "extremely" to write "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was really fun."

[0614] Step 10:

[0615] When the user completes the edits and clicks the "Confirm" button, the device sends the final edited text back to the server, again using an HTTP request.

[0616] Step 11:

[0617] The server receives the final text and stores it in the CGM service's database, where it can be viewed by other users.

[0618] Step 12:

[0619] The server will then make the final text publicly available for other users to view on the CGM service, which will encourage users to feel more comfortable posting content and increase overall user numbers and engagement for the service.

[0620] In this way, the system of the present invention generates natural sentences that reflect emotions while eliminating particular writing styles and phrasing habits, providing an environment in which users can post to CGM services with peace of mind.

[0621] Example 2

[0622] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0623] In modern CGM services, users are required to maintain anonymity and post with peace of mind, avoiding identification based on specific writing styles or phrasing. However, existing systems have shortcomings in ensuring this anonymity, which can lead to issues with users not feeling secure enough. Furthermore, it is difficult to generate text that accurately reflects the sentiment of the user, which can lead to a poor user experience.

[0624] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text and identifying idiosyncrasies of writing style and phrasing, means for removing idiosyncrasies of writing style and phrasing based on the analysis results and generating text with increased anonymity, means for recognizing the emotion of the generated text and retaining that emotion, means for transmitting the generated text to the user's terminal, means for the user to confirm and correct the generated text, and means for saving the corrected final text in a database and making it public. This makes it possible to generate natural-looking text that reflects emotion while preventing identification based on idiosyncratic writing style and phrasing, allowing users to post their text to the CGM service with confidence.

[0625] "User" means a user who accesses the CGM Service and posts or views content.

[0626] A "terminal" is an electronic device used by a user, such as a computer, smartphone, or tablet.

[0627] A "server" is a computer system that receives a user's request, processes the data in accordance with the request, and returns the data.

[0628] "Text" refers to text data entered or generated by a user.

[0629] "Analysis" is a process for identifying the style and phrasing of the input text.

[0630] "Identifying" means identifying the specific stylistic and phrasing patterns in a piece of writing.

[0631] "Generation" involves removing identified stylistic and phrasing habits to create new sentences.

[0632] "Emotion recognition" means determining emotions such as positive, negative, or neutral from the content of a sentence.

[0633] "Preserving emotion" means maintaining the emotion of the original sentence in the generated new sentence.

[0634] "Send" means to transfer data from the server to the user's terminal.

[0635] "Verification" means that the user checks the generated text and confirms its content.

[0636] "Correction" means that the user makes changes to the generated text.

[0637] "Saving" means recording the final corrected text in the database.

[0638] "Publishing" means making the saved final text available for other users to view.

[0639] "Anonymity" means keeping users difficult to identify.

[0640] A "generative AI model" is an artificial intelligence program that uses natural language processing technology to generate new sentences.

[0641] A "prompt sentence" is text that gives instructions to the generative AI model on the sentence to output.

[0642] This invention is a system that can generate text with enhanced anonymity by removing a user's particular writing style and phrasing habits, and aims to provide an environment in which users can post content safely, particularly in CGM services. By combining a generative AI model and an emotion engine, this system neutralizes the writing style while recognizing and preserving the emotion of the text, ensuring anonymity.

[0643] First, the user accesses the CGM service using a device and enters the text they want to post into the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[0644] Next, when the user clicks the "Send" button, the device sends the input text as an HTTP request to the server, which then passes the received text to a text analysis module to identify stylistic and phrasing patterns. This analysis is performed using a Natural Language Processing (NLP) library (e.g., spaCy or NLTK).

[0645] Based on the analysis results, the server passes the sentence to an emotion engine, which uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to recognize the sentiment of the sentence, such as positive, negative, or neutral. Based on the obtained sentiment information and the identified stylistic features, the server uses a generative AI model (e.g., OpenAI's GPT-3) to generate a sentence with increased anonymity. This generation process uses a prompt such as: "Please neutralize the stylistic features while preserving the sentiment based on the original sentence."

[0646] The generated sentence might be something like, "Today was a really great day! I had a great time with my friends." This generated sentence is sent from the server to the user's device as an HTTP response.

[0647] The user can review the generated sentences on their device and make corrections as necessary. The emotion engine recognizes and retains emotions even when the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the user can write, "Today was a wonderful day! I had a wonderful time with my friends."

[0648] Finally, when the user confirms the final sentence, the device sends this sentence back to the server, which stores the final sentence in a database and makes it available on the CGM service.

[0649] This system allows users to avoid being identified by specific writing styles or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of an emotion engine and generative AI model maintains the naturalness of writing and consistency of emotion, which is expected to improve the user experience.

[0650] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0651] Step 1: The user accesses the CGM service through their device and enters the text they want to post in the input form. Specifically, they enter the text "Today was a great day! I had a great time with my friends" into the text box. The input is text data, and the output is the input text recorded on the device.

[0652] Step 2: When the user clicks the "Submit" button, the terminal sends the entered text to the server via an HTTP request. Specifically, the terminal packages the user's input in JSON format and sends it to the server's API endpoint as an HTTP POST request. The input is the entered text, and the output is the JSON data sent to the server.

[0653] Step 3: The server passes the received text to a text analysis module, which performs processing to identify stylistic and phrasing patterns. Specifically, the server uses a Natural Language Processing (NLP) library (e.g., spaCy or NLTK) to tokenize the text and extract contextual and stylistic features. The input is the JSON data sent in the HTTP request, and the output is a data structure containing the analysis results.

[0654] Step 4: The server passes the analysis results and the text to the emotion engine to recognize the sentiment of the text. Specifically, the server uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to identify the sentiment of the text, such as positive, negative, or neutral. The input is the text analysis result data, and the output is data with sentiment labels.

[0655] Step 5: Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate sentences with enhanced anonymity. Specifically, the server inputs a prompt to the generative AI model: "Please neutralize the style while preserving the emotion based on the original sentence." and obtains the generated sentence. The input is the original sentence data with emotion labels and the prompt sentence, and the output is the generated new sentence.

[0656] Step 6: The server sends the generated new text to the user's device as an HTTP response. Specifically, the generated text is included in the body of the HTTP response in JSON format and sent to the device. The input is the generated new text data, and the output is the JSON data sent to the device.

[0657] Step 7: The user checks the generated sentence on the terminal and corrects it if necessary. Specifically, the user looks at the generated sentence displayed on the terminal and corrects it, for example, from "really" to "very." The input is the generated sentence, and the output is the sentence corrected by the user.

[0658] Step 8: After the user confirms the final text edits, the device resends the text to the server. Specifically, the revised text is repackaged in JSON format and sent to the server's API endpoint as an HTTP POST request. The input is the text edited by the user, and the output is the JSON data resent to the server.

[0659] Step 9: The server saves the revised final sentence in the database and publishes it on the CGM service. Specifically, the server inserts the final sentence into the appropriate table in the database and sets it up for publication so that other users can view it. The input is the revised final sentence data, and the output is the data saved in the database and the sentence published on the CGM service.

[0660] (Application example 2)

[0661] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0662] Current online content generation services pose a risk of identifying individuals through their unique writing style and phrasing. Furthermore, when users want to express their feelings frankly, there is a lack of ways to generate text that maintains their emotions while also increasing anonymity. Therefore, there is a need to provide an environment where users can post reviews and comments with peace of mind.

[0663] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify writing style and phrasing habits, means for generating text with increased anonymity by removing writing style and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, emotion analysis means for analyzing the emotion of the text entered by the user, means for generating a prompt text based on the analyzed emotion information and generating a text that reflects the emotion while maintaining anonymity, means for anonymizing specific words and phrases to increase the anonymity of the generated text, and means for transmitting the final corrected text back to the user's terminal. This allows users to safely post text that reflects their emotions while avoiding identification based on their unique writing style and phrasing.

[0664] "Means for receiving text entered by a user" refers to a process or device that allows the server to receive text data sent from the user's terminal.

[0665] "Means for analyzing received text and identifying writing style and phrasing" refers to a process or program that uses natural language processing technology to analyze input text and extract characteristics of its writing style and expression.

[0666] "Means for generating text with enhanced anonymity by removing stylistic and phrasing characteristics" refers to a technology that uses the analysis results to generate new text with enhanced anonymity by eliminating specific stylistic and phrasing characteristics.

[0667] "Means for transmitting the generated text to the user's terminal" refers to a communication process or device for returning the text generated by the server to the user's terminal.

[0668] The "means for the user to check and correct the generated text" refers to an interface and function that allows the user to check the generated text and correct it if necessary.

[0669] "Means for storing and publishing the final revised text in a database" refers to the process or device by which the final revised text by the user is recorded in a database and published in a form accessible to other users.

[0670] "Sentiment analysis means for analyzing the sentiment of text entered by a user" is a natural language processing technology for analyzing the sentiment of input text and identifying sentiment categories such as positive or negative.

[0671] "A means for generating prompt sentences based on analyzed emotional information and generating sentences that reflect the emotions while maintaining anonymity" is a technology that generates sentences that maintain anonymity while retaining the emotions of the original sentences based on the results of emotional analysis.

[0672] "Means for anonymizing specific words and phrases to increase the anonymity of generated text" refers to a technology that increases anonymity by extracting words and phrases that could identify individuals from text and converting them into other words.

[0673] The "means for retransmitting the final revised text to the user's terminal" refers to a process or device that retransmits the final version of the text that the user has completed the revisions to the user's terminal.

[0674] System Program

[0675] A system for realizing this application example includes the following means.

[0676] Hardware and Software

[0677] Hardware: Smartphones, servers

[0678] Software: Python, Hugging Face Transformers library

[0679] Data processing and calculation

[0680] This system sends text entered by a user on a smartphone to a server, which then analyzes the text and identifies writing style and phrasing habits. It also uses a sentiment analysis tool to analyze the sentiment of the entered text and generates a prompt based on that sentiment information. A generative AI model then generates a sentence that reflects the sentiment while maintaining anonymity, and sends the sentence back to the user's device.

[0681] Processing flow

[0682] 1. User input:

[0683] Users use their smartphones to enter reviews into the food delivery app, which are then sent to the server as HTTP requests.

[0684] 2. Parsing on the server:

[0685] When a review arrives at the server, it is first analyzed by a text analysis module to identify its stylistic and phrasing characteristics, and then by a sentiment analysis module to analyze the review's sentiment (e.g., positive, negative, etc.).

[0686] 3. Sentence generation:

[0687] Based on this analysis, a generative AI model is used to generate prompt sentences that reflect the sentiment while maintaining anonymity. For example, a review with a positive sentiment, "The pizza I ordered was so delicious I ate two right away!", can be converted to "The pizza I ordered was so delicious I ate two right away!"

[0688] 4. User Verification and Correction:

[0689] The generated text is then sent back to the user's smartphone, where the user can review it and make any necessary corrections. The corrected text is then sent back to the server as the final version.

[0690] 5. Final saving and publishing:

[0691] The final edited text is stored in a database and published as a review of the food delivery app that can be viewed by other users.

[0692] Specific examples

[0693] Prompt Sentence Examples

[0694] "Transform the original sentence into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right there!"

[0695] This system can be applied not only to food delivery apps but also to other online content generation services, providing an environment where users can post reviews and comments with confidence. It is also expected to improve the user experience by maintaining naturalness and emotional consistency in the writing.

[0696] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0697] Step 1:

[0698] Input: A user uses their smartphone to input a review into a food delivery app.

[0699] Action: For example, type, "The pizza I ordered was so good I ate two right there!"

[0700] Output: This entered review is sent by the device to the server as an HTTP request.

[0701] Step 2:

[0702] Input: Reviews received by the server.

[0703] How it works: The server first passes the review to a text analysis module to analyze its writing style and phrasing, and also uses a sentiment analysis module to analyze the review's sentiment (positive, negative, etc.).

[0704] Output: Stylistic and phrasing characteristics, as well as sentiment information.

[0705] Step 3:

[0706] Input: Stylistic and phrasing characteristics, emotional information.

[0707] What it does: The server generates a prompt based on this information, such as "Transform the original text into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right away!"

[0708] Output: The prompt statement and any parsing information needed to generate it.

[0709] Step 4:

[0710] Input: Prompt statement and parsing information.

[0711] How it works: The server uses a generative AI model to generate new, sentiment-based sentences while preserving anonymity, such as "The pizza I ordered was so delicious, I had two right away!"

[0712] Output: The new sentence generated.

[0713] Step 5:

[0714] Input: The new sentence to be generated.

[0715] What it does: The server resends this generated text to the user's device as an HTTP response.

[0716] Output: The generated text is displayed on the user's smartphone.

[0717] Step 6:

[0718] Input: User-generated text verification.

[0719] How it works: The user reviews the generated sentence and makes corrections if necessary, for example changing "very" to "really" to get "The pizza I ordered was so good, I had two right away!"

[0720] Output: The sentence corrected by the user.

[0721] Step 7:

[0722] Input: Final corrected text.

[0723] How it works: The user's device sends the revised text back to the server, which stores this final text in its database and makes it publicly available for other users to view.

[0724] Output: The final sentence stored in the database.

[0725] By following these steps, users can confidently submit reviews that reflect their feelings while avoiding being identified by their unique writing style or phrasing.

[0726] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0727] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0728] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0729] [Third embodiment]

[0730] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0731] 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.

[0732] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0733] 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.

[0734] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0735] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0736] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0737] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0738] 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.

[0739] 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.

[0740] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0741] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0742] The present invention is a system that generates texts with enhanced anonymity by removing specific writing styles and phrasing habits in order to provide an environment in which users can post with peace of mind on CGM services. The system of the present invention analyzes texts entered by users and utilizes a generative AI model to remove specific writing styles and phrasing characteristics.

[0743] Below, the program processing of this system is explained in natural language, with specific examples included.

[0744] Program processing flow

[0745] 1. User Input

[0746] Users access the CGM service using their device and enter the text they want to post in the input form. For example, they might enter, "I went to a delicious cafe. The coffee was amazing!"

[0747] 2. Sending text

[0748] The terminal sends the text entered by the user to the server as is. The entered text is sent to the server using an HTTP request.

[0749] 3. Analysis and generation using AI models

[0750] The server passes the received text to a text analysis module. The text analysis module extracts features to identify the style and phrasing of the text. After feature extraction is complete, the analysis results are passed to a generative AI model. The generative AI model generates new text that removes the identified style and phrasing features to increase anonymity. For example, it generates "I went to a new cafe. The coffee was very delicious."

[0751] 4. Sending the generated text

[0752] The server sends the generated text to the terminal, which returns it to the terminal as an HTTP response.

[0753] 5. User Review and Correction

[0754] The terminal displays the generated sentence received from the server to the user. The user can check the generated sentence and further modify the expressions and phrases as necessary. For example, the user can modify "very" to "really" to write "I went to a new cafe. The coffee was really delicious."

[0755] 6. Final Post

[0756] The user then sends the finalized text back to the server, which saves the revised final text in the CGM service database and makes it public. The published final text can then be viewed by other users.

[0757] Specific processing examples

[0758] Case: User B's post

[0759] User Input:

[0760] User B posts on social media, "I went to a new restaurant. The food was delicious."

[0761] Sending text:

[0762] The terminal sends this text to the server.

[0763] AI model analysis and generation:

[0764] The server receives the sentence, analyzes it with a text analysis module to identify stylistic and phrasing habits, and then the generative AI model generates the sentence, "I enjoyed eating at a new restaurant. The food was delicious."

[0765] Sending generated text:

[0766] The server sends the generated text to the terminal.

[0767] User review and correction:

[0768] User B checks the generated sentence and corrects "very" to "very." The final sentence is confirmed as "I enjoyed eating at the new restaurant. The food was delicious."

[0769] Last Post:

[0770] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[0771] This system allows users to post to the CGM service with peace of mind, avoiding identification based on specific writing styles or phrasing, which is expected to increase the number of users and page views of the service.

[0772] The processing flow will be explained below.

[0773] Step 1:

[0774] The user uses a device to access the input form of the CGM service and enters the text they want to post. For example, they might enter, "Today, we went to a new park and had a picnic. It was so much fun!"

[0775] Step 2:

[0776] When a user clicks the "Submit" button on the input form, the device sends the entered text to the server using an HTTP request, with the user's input data being sent to the server in the appropriate format.

[0777] Step 3:

[0778] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify its style and phrasing, specifically by analyzing the sentence structure and extracting specific phrases and expressions.

[0779] Step 4:

[0780] Based on the analysis results, the server passes the data to a generative AI model. The generative AI model takes the analysis results as input, removes stylistic and phrasing habits, and generates new sentences with increased anonymity. This process uses an algorithm that converts the sentences into different expressions while preserving the original meaning. For example, it generates the sentence, "Today we had a picnic at a new park. We had a lot of fun."

[0781] Step 5:

[0782] The server sends the generated new text to the device as an HTTP response, which includes the generated text and is displayed to the user.

[0783] Step 6:

[0784] The user checks the generated sentence on the device. If the user feels that a correction is necessary, the user corrects the generated sentence using the device interface. For example, changing "very" to "really." The final sentence is completed: "Today, we had a picnic at the new park. We really enjoyed it."

[0785] Step 7:

[0786] When the user confirms the final sentence, he clicks the "Confirm" button, and the terminal sends the corrected final sentence to the server again, again using an HTTP request.

[0787] Step 8:

[0788] The server receives the final text and stores it in a database, including any edits made by the user.

[0789] Step 9:

[0790] The server then publishes the final text on the CGM service, where it can be viewed by other users.

[0791] In this way, after going through a series of processes, the user's text is converted into a form that makes it difficult to identify and is then safely posted to the CGM service.

[0792] Example 1

[0793] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0794] In conventional CGM services, there is a risk that users can be identified by their unique writing style and phrasing in the text they post. This makes it difficult for users to post with confidence, which could lead to a decline in service usage. The lack of an appropriate system for increasing anonymity is an issue.

[0795] 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.

[0796] In this invention, the server includes: means for receiving text entered by a user; means for analyzing the received text and identifying writing style and phrasing habits; means for removing writing style and phrasing habits and generating text with increased anonymity based on the analysis results; means for sending the generated text to the user's terminal; means for the user to check and correct the generated text; means for saving and publishing the corrected final text in a database; a text analysis module for extracting features to identify specific writing style and phrasing habits; and generation AI model means for generating text with increased anonymity. This allows users to avoid being identified by their particular writing style or phrasing and to post to CGM services with peace of mind.

[0797] "User" refers to an individual or corporation that uses the CGM Service.

[0798] "Terminal" is a general term for the device used by a User to access and operate the CGM Service.

[0799] "Server" refers to a computer system that manages processing and data related to CGM services.

[0800] "Text" refers to the text data entered by a user to post on the CGM service.

[0801] "Means for receiving" refers to the function by which the server receives text sent by the user.

[0802] "Means for analysis" refers to the function for identifying the style and phrasing of the received text.

[0803] "Means of identification" refers to the function for extracting features in a text based on the analysis results.

[0804] "Means of generation" refers to the function for creating new text with increased anonymity based on the analysis results.

[0805] "Means for sending" refers to the function for returning the generated text to the user's device.

[0806] "Means for reviewing and correcting" refers to a function that allows the user to review the generated text and make corrections if necessary.

[0807] "Means for saving and publishing" refers to the functionality for saving the final revised text in a database and making it available for other users to view.

[0808] "Feature extraction" refers to the automated process of identifying specific stylistic or phrasing patterns in a piece of text.

[0809] "Text analysis module" refers to a software component for analyzing received text and extracting features.

[0810] A "generative AI model" refers to an artificial intelligence system that generates new text by removing identified stylistic and phrasing habits.

[0811] "CGM service" is a general term for online platforms that allow users to post and share content.

[0812] This invention is a system that allows users to generate sentences with enhanced anonymity by removing their specific writing style and phrasing. This system consists of the following main components:

[0813] User Input

[0814] A user accesses a CGM (Consumer Generated Media) service using a device and inputs a sentence into an input form. For example, consider a user who inputs the sentence, "I went to a delicious cafe. The coffee was great!"

[0815] Sending text

[0816] The device sends the text entered by the user to the server as an HTTP POST request, which is sent in JSON format and contains the text data.

[0817] Server analysis

[0818] The server passes the received text to a text analysis module, which uses natural language processing techniques (e.g., Spacy or NLTK) to extract features to identify the text's stylistic and phrasing habits.

[0819] Generation by generative AI model

[0820] The server passes the sentences after feature extraction to a generative AI model (such as OpenAI's GPT-3). The generative AI model generates new sentences that remove the identified stylistic and phrasing habits to enhance anonymity. For example, the sentence "I went to a delicious cafe. The coffee was great!" is converted to "I went to a new cafe. The coffee was really good."

[0821] Sending generated text

[0822] The server returns the generated text to the terminal as an HTTP response, with the appropriate Content-Type set in the response header.

[0823] User review and correction

[0824] The device displays the generated sentence to the user. The user can review the sentence and make corrections as necessary. For example, the user can change "very" to "really" to say, "I went to a new cafe. The coffee was really delicious."

[0825] Last Post

[0826] The user then sends the finalized sentence back to the server. The server saves the final, revised sentence in the CGM service's database and makes it publicly available for other users to view. For example, the published sentence might be, "I went to a new cafe. The coffee was really delicious."

[0827] This system allows users to post to CGM services with confidence, avoiding identification based on specific writing styles or phrasing. Below are some example prompts:

[0828] Example prompt sentence:

[0829] "Please make the following sentence more anonymous by removing certain styles and phrasing: We went to a new restaurant. The food was delicious."

[0830] By operating this system properly, it is expected that the number of service users and page views will increase.

[0831] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0832] Step 1:

[0833] A user accesses the CGM service using a device and enters a sentence into the input form. For example, they might enter the sentence, "I went to a delicious cafe. The coffee was amazing!" This input becomes the starting point for the next process.

[0834] Step 2:

[0835] The device converts the input text into JSON format and sends it to the server as the payload of an HTTP POST request. Specifically, it generates a POST request and sends the input text data to the server. This request becomes the input for the analysis process.

[0836] Step 3:

[0837] The server analyzes the received HTTP POST request and extracts the text data contained therein, which becomes the input for the text analysis module.

[0838] Step 4:

[0839] The server runs a text analysis module and performs feature extraction to identify the style and phrasing of the text. Specifically, it analyzes the text using natural language processing techniques (such as Spacy or NLTK) and generates a feature vector. This feature vector becomes the input for the generative AI model.

[0840] Step 5:

[0841] The server passes the feature vector to a generative AI model, which generates new sentences. The generative AI model (such as OpenAI's GPT-3) removes the identified stylistic and phrasing habits and generates new sentences with increased anonymity. For example, the input "I went to a delicious cafe. The coffee was great!" yields the output "I went to a new cafe. The coffee was delicious." This new sentence becomes the input for the next step.

[0842] Step 6:

[0843] The server converts the generated new text into JSON format and returns it to the terminal as an HTTP response. Specifically, it generates an HTTP response and sends it with the generated text included in the payload. This response data becomes the input for the terminal.

[0844] Step 7:

[0845] The terminal parses the received HTTP response, extracts the generated text, and displays it to the user. The user can check the new text on the screen and make corrections in the text box if necessary. This corrected text becomes the input for the final post.

[0846] Step 8:

[0847] The user then sends the revised text back to the server. Specifically, the device converts the revised text back into JSON format and sends it to the server as an HTTP POST request. This request becomes the input for the final saving process.

[0848] Step 9:

[0849] The server saves the final text after receiving the corrections in a database and makes it available on the CGM service. Specifically, it performs a saving process, stores the text in a database, and makes it available for other users to view. This data is the final output.

[0850] This allows users to avoid being identified by their particular writing style or phrasing, allowing them to post to CGM services with peace of mind.

[0851] (Application example 1)

[0852] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0853] In conventional CGM services and electronic messaging systems, users often face the risk of being identified through their unique writing style and phrasing, and anonymity is often not guaranteed. Furthermore, phishing attacks are on the rise in email and messaging systems, increasing the likelihood that users will be victimized by fraudulent messages. This makes protecting user privacy and improving security even more important.

[0854] 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.

[0855] In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify stylistic and phrasing habits, means for generating text with increased anonymity by removing the stylistic and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, means for converting the stylistic and phrasing characteristics of the received electronic message into standard expressions to prevent phishing attacks using electronic messages, means for recording the converted electronic message in a log file, and means for transmitting the converted electronic message to the user. This not only increases the anonymity of text entered by a user, but also makes it possible to prevent phishing attacks using electronic messages.

[0856] "User-entered text" refers to the text content entered by a user through a terminal or system.

[0857] "Means of receiving" refers to the function by which a server or system receives text sent by a user.

[0858] "Means of analyzing and identifying writing style and phrasing habits" refers to a function that uses machine learning models and natural language processing to analyze the writing style and unique expressions of input text and extract features.

[0859] The "means for generating sentences with enhanced anonymity" is a function that removes peculiar writing styles and phrases based on the characteristics obtained through analysis, and generates sentences in a standard writing style.

[0860] "Means for sending the generated text to the user's device" refers to a function that sends the highly anonymous text generated by the server back to the user's device.

[0861] "Means for checking and correcting" is a function that provides an interface for users to view the generated text and correct the content as necessary.

[0862] "Means for saving in a database and publishing" refers to the function of saving the final revised text on a server and publishing it in a form that other users can access.

[0863] "Means for converting stylistic and phrasing characteristics of electronic messages into standard expressions to prevent phishing attacks" refers to a function that converts the stylistic and idiosyncratic expressions of received electronic messages into standard business expressions, thereby preventing attacks before they occur.

[0864] "Means for recording to a log file" means a function that keeps a record of converted electronic messages for auditing and analysis.

[0865] A "means for sending an electronic message to a user" is a function that sends a converted secure electronic message to a user's device.

[0866] This invention is a system for preventing phishing attacks by anonymizing text so that users cannot be identified based on their unique writing style or phrasing habits. To realize this system, a program including the following means is required.

[0867] Program processing overview

[0868] 1. A means of receiving text entered by the user

[0869] Users access the system using a terminal and enter the text they want to post or the text of an email they received into the input form. For example, they could enter, "Please send us your bank details quickly! You'll receive a reward!"

[0870] 2. A means of analyzing received text and identifying stylistic and phrasing habits

[0871] The server receives the text sent by the user and analyzes it using Natural Language Processing (NLP) technology, which identifies stylistic and expressive features.

[0872] 3. A method for generating text with increased anonymity

[0873] The server uses a generative AI model to remove stylistic and phrasing quirks and convert the sentence into standard expressions. The generative AI model uses, for example, the OpenAI API. "Please send us your bank details quickly. We'll give you a reward!" is converted to "Could you please provide us with the following bank details as soon as possible? Thank you."

[0874] 4. A means of sending the generated text to the user's device

[0875] The server then sends the generated standard text back to the user's device using an HTTP response.

[0876] 5. A way for users to review and correct the generated text

[0877] The user can view the generated text received from the server and make further corrections as necessary. The user can freely edit the text displayed on the terminal.

[0878] 6. A means of storing and publishing the final revised text in a database

[0879] Once the user has finalized the text, it is sent back to the server and stored in the database, after which it is made publicly accessible to other users.

[0880] 7. Means of converting stylistic and phrasing features of electronic messages into standard expressions to prevent phishing attacks

[0881] The server converts the style and idioms of received emails into standard business style, particularly to prevent phishing attacks, for example converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[0882] 8. Means for recording converted electronic messages in a log file

[0883] All conversion processes are recorded in log files for later auditing and analysis, facilitating future problem solving and improvement planning.

[0884] 9. Means of sending the converted electronic message to the user

[0885] The server then sends the converted secure email to the user's device, again via an HTTP response.

[0886] Specific examples

[0887] As a specific example of use, if a security department staff member receives a suspicious email such as "Please send your bank information quickly. You will receive a reward!", the system will operate as follows.

[0888] Example prompt for a generative AI model:

[0889] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[0890] Submit your bank details quickly and you'll be rewarded!

[0891] This prompt provides users with standardized text generated by an AI model (e.g., "Could you please provide the following banking information as soon as possible? Thank you."), helping to prevent phishing attacks. The system achieves high levels of anonymity and security through communication between the server and the device, natural language processing, and the use of generative AI models.

[0892] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0893] Step 1:

[0894] The user inputs a sentence into the terminal. The input sentence is something like "Please send your bank information quickly. You will receive a reward!". The terminal sends this sentence to the server via an HTTP request.

[0895] Step 2:

[0896] The server analyzes the text received from the user. To do this, the text is passed to a natural language processing module to identify stylistic and phrasing patterns. This analysis is performed using text feature extraction algorithms.

[0897] Step 3:

[0898] Based on the analysis results, the server uses a generative AI model to generate new sentences that remove stylistic and phrasing habits. Specifically, the server passes the following prompt sentence as input to the generative AI model:

[0899] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[0900] Submit your bank details quickly and you'll be rewarded!

[0901] The output will be a secure sentence such as "Could you please provide the following bank details as soon as possible? Thank you."

[0902] Step 4:

[0903] The generated text is sent from the server to the terminal as an HTTP response, and the terminal displays the generated text to the user.

[0904] Step 5:

[0905] The user checks the generated text and makes any necessary corrections. The corrections are then sent back to the server as an HTTP request from the terminal.

[0906] Step 6:

[0907] The server saves the final corrected sentence in the database. The saved data is the final corrected sentence. For example, "Could you please provide the following bank information as soon as possible? Thank you." is saved in the database.

[0908] Step 7:

[0909] The server analyzes received electronic messages and converts stylistic and phrasing features into standard expressions to prevent phishing attacks, for example, converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[0910] Step 8:

[0911] The converted electronic messages are logged in a log file. The server stores the conversion process in a log file (e.g. "sanitized_emails_log.txt") for audit purposes.

[0912] Step 9:

[0913] Finally, the converted electronic message is sent from the server to the user's terminal and displayed to the user, thereby protecting the user from phishing attacks.

[0914] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0915] This invention relates to a system that generates text with enhanced anonymity by removing idiosyncratic writing styles and phrasing, in order to provide an environment in which users can post content safely on CGM services. It also includes an emotion engine that recognizes the emotions of user input and generated text and incorporates them into the system.

[0916] Below, the program processing of this system is explained in natural language, with specific examples included.

[0917] Program processing flow

[0918] 1. User Input

[0919] Users access the CGM service through their devices and enter the text they want to post in the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[0920] 2. Sending text

[0921] When the user clicks the "Send" button, the device sends the entered text to the server via an HTTP request.

[0922] 3. Analysis and generation using AI models and emotion engines

[0923] The server first passes the received text to a text analysis module, which identifies the text's style and phrasing. The analysis results and the text are then passed to an emotion engine to recognize the text's emotion.

[0924] Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate a more anonymous sentence. This process preserves the emotion of the original sentence while removing the stylistic and phrasing characteristics, resulting in a new sentence. For example, "Today was a really great day! I had a great time with my friends."

[0925] 4. Sending the generated text

[0926] The server sends the generated new text to the user's device as an HTTP response, where it is displayed.

[0927] 5. User Review and Correction

[0928] The user can review the generated sentence and make corrections as needed. The emotion engine recognizes and preserves emotions even as the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the sentence becomes, "Today was a wonderful day! I had a wonderful time with my friends."

[0929] 6. Final Post

[0930] Once the user has confirmed the final sentence, the device sends it back to the server, which then stores the final sentence in a database and makes it available on the CGM service.

[0931] Specific processing examples

[0932] Case: User B's post

[0933] User Input:

[0934] User B posts on social media, "I went to a new cafe. The coffee was delicious and the atmosphere was fun."

[0935] Sending text:

[0936] The terminal sends this text to the server.

[0937] Analysis and generation by AI model and emotion engine:

[0938] The server analyzes the text using a text analysis module to identify characteristics of the writing style and phrasing. Next, an emotion engine recognizes the emotion of the text. For example, positive emotion is recognized. The generative AI model reflects this emotion information and generates a response such as, "I visited a new cafe. The coffee was delicious and the atmosphere was very pleasant."

[0939] Sending generated text:

[0940] The server sends the generated text to the terminal.

[0941] User review and correction:

[0942] User B checks the generated sentence and corrects "very" to "extremely", completing the final sentence: "I visited a new cafe. The coffee was delicious and the atmosphere was very enjoyable."

[0943] Last Post:

[0944] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[0945] This system allows users to avoid being identified by their particular writing style or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of the emotion engine is expected to improve the user experience by maintaining naturalness in writing and consistency in emotions.

[0946] The processing flow will be explained below.

[0947] Step 1:

[0948] The user uses a device to access the input form of the CGM service and enters the sentence they want to post. For example, they might enter, "Today was beautiful weather, so I had a picnic in the park with my family. It was really fun!"

[0949] Step 2:

[0950] When a user clicks the "Submit" button on the input form, the terminal sends the input text to the server using an HTTP request. This request includes the user's input text.

[0951] Step 3:

[0952] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify stylistic and phrasing habits, such as the use of specific punctuation marks or frequently used phrases.

[0953] Step 4:

[0954] The server passes the analysis results to the emotion engine, which recognizes the emotion of the input text. For example, it determines that expressions such as "great weather" and "it was fun" contain positive emotions.

[0955] Step 5:

[0956] The emotion engine returns the recognized emotion information to the server, which then prepares the analysis results and emotion information to be input into the generative AI model.

[0957] Step 6:

[0958] The server inputs the analysis results and emotional information into the generative AI model. Based on this information, the generative AI model removes stylistic and phrasing habits and generates sentences with increased anonymity while preserving the original emotion. For example, it generates a sentence like, "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was a lot of fun."

[0959] Step 7:

[0960] The server sends the generated new text to the device as an HTTP response, which includes the generated text.

[0961] Step 8:

[0962] The device displays the generated text received from the server to the user, who then checks the text to see if it expresses the appropriate emotion.

[0963] Step 9:

[0964] The user can make necessary corrections to the generated sentences, for example, changing "very" to "extremely" to write "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was really fun."

[0965] Step 10:

[0966] When the user completes the edits and clicks the "Confirm" button, the terminal sends the final edited text back to the server, again using an HTTP request.

[0967] Step 11:

[0968] The server receives the final text and stores it in the CGM service's database, where it can be viewed by other users.

[0969] Step 12:

[0970] The server will then make the final text publicly available for other users to view on the CGM service, which will encourage users to feel more comfortable posting content and increase overall user numbers and engagement for the service.

[0971] In this way, the system of the present invention generates natural sentences that reflect emotions while eliminating particular writing styles and phrasing habits, providing an environment in which users can post to CGM services with peace of mind.

[0972] Example 2

[0973] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0974] In modern CGM services, users are required to maintain anonymity and post with peace of mind, avoiding identification based on specific writing styles or phrasing. However, existing systems have shortcomings in ensuring this anonymity, which can lead to issues with users not feeling secure enough. Furthermore, it is difficult to generate text that accurately reflects the sentiment of the user, which can lead to a poor user experience.

[0975] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text and identifying idiosyncrasies of writing style and phrasing, means for removing idiosyncrasies of writing style and phrasing based on the analysis results and generating text with increased anonymity, means for recognizing the emotion of the generated text and retaining that emotion, means for transmitting the generated text to the user's terminal, means for the user to confirm and correct the generated text, and means for saving the corrected final text in a database and making it public. This makes it possible to generate natural-looking text that reflects emotion while preventing identification based on idiosyncratic writing style and phrasing, allowing users to post their text to the CGM service with confidence.

[0976] "User" means a user who accesses the CGM Service and posts or views content.

[0977] A "terminal" is an electronic device used by a user, such as a computer, smartphone, or tablet.

[0978] A "server" is a computer system that receives a user's request, processes the data in accordance with the request, and returns the data.

[0979] "Text" refers to text data entered or generated by a user.

[0980] "Analysis" is a process for identifying the style and phrasing of the input text.

[0981] "Identifying" means identifying the specific stylistic and phrasing patterns in a piece of writing.

[0982] "Generation" involves removing identified stylistic and phrasing habits to create new sentences.

[0983] "Emotion recognition" means determining emotions such as positive, negative, or neutral from the content of a sentence.

[0984] "Preserving emotion" means maintaining the emotion of the original sentence in the generated new sentence.

[0985] "Send" means to transfer data from the server to the user's terminal.

[0986] "Verification" means that the user checks the generated text and confirms its content.

[0987] "Correction" means that the user makes changes to the generated text.

[0988] "Saving" means recording the final corrected text in the database.

[0989] "Publishing" means making the saved final text available for other users to view.

[0990] "Anonymity" means keeping users difficult to identify.

[0991] A "generative AI model" is an artificial intelligence program that uses natural language processing technology to generate new sentences.

[0992] A "prompt sentence" is text that gives instructions to the generative AI model on the sentence to output.

[0993] This invention is a system that can generate text with enhanced anonymity by removing a user's particular writing style and phrasing habits, and aims to provide an environment in which users can post content safely, particularly in CGM services. By combining a generative AI model and an emotion engine, this system neutralizes the writing style while recognizing and preserving the emotion of the text, ensuring anonymity.

[0994] First, the user accesses the CGM service using a device and enters the text they want to post into the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[0995] Next, when the user clicks the "Send" button, the device sends the input text as an HTTP request to the server, which then passes the received text to a text analysis module to identify stylistic and phrasing patterns. This analysis is performed using a Natural Language Processing (NLP) library (e.g., spaCy or NLTK).

[0996] Based on the analysis results, the server passes the sentence to an emotion engine, which uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to recognize the sentiment of the sentence, such as positive, negative, or neutral. Based on the obtained sentiment information and the identified stylistic features, the server uses a generative AI model (e.g., OpenAI's GPT-3) to generate a sentence with increased anonymity. This generation process uses a prompt such as: "Please neutralize the stylistic features while preserving the sentiment based on the original sentence."

[0997] The generated sentence might be something like, "Today was a really great day! I had a great time with my friends." This generated sentence is sent from the server to the user's device as an HTTP response.

[0998] The user can review the generated sentences on their device and make corrections as necessary. The emotion engine recognizes and retains emotions even when the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the user can write, "Today was a wonderful day! I had a wonderful time with my friends."

[0999] Finally, when the user confirms the final sentence, the device sends this sentence back to the server, which stores the final sentence in a database and makes it available on the CGM service.

[1000] This system allows users to avoid being identified by specific writing styles or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of an emotion engine and generative AI model maintains the naturalness of writing and consistency of emotion, which is expected to improve the user experience.

[1001] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1002] Step 1: The user accesses the CGM service through their device and enters the text they want to post in the input form. Specifically, they enter the text "Today was a great day! I had a great time with my friends" into the text box. The input is text data, and the output is the input text recorded on the device.

[1003] Step 2: When the user clicks the "Submit" button, the terminal sends the entered text to the server via an HTTP request. Specifically, the terminal packages the user's input in JSON format and sends it to the server's API endpoint as an HTTP POST request. The input is the entered text, and the output is the JSON data sent to the server.

[1004] Step 3: The server passes the received text to a text analysis module, which performs processing to identify stylistic and phrasing patterns. Specifically, the server uses a Natural Language Processing (NLP) library (e.g., spaCy or NLTK) to tokenize the text and extract contextual and stylistic features. The input is the JSON data sent in the HTTP request, and the output is a data structure containing the analysis results.

[1005] Step 4: The server passes the analysis results and the text to the emotion engine to recognize the sentiment of the text. Specifically, the server uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to identify the sentiment of the text, such as positive, negative, or neutral. The input is the text analysis result data, and the output is data with sentiment labels.

[1006] Step 5: Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate sentences with enhanced anonymity. Specifically, the server inputs a prompt to the generative AI model: "Please neutralize the style while preserving the emotion based on the original sentence." and obtains the generated sentence. The input is the original sentence data with emotion labels and the prompt sentence, and the output is the generated new sentence.

[1007] Step 6: The server sends the generated new text to the user's device as an HTTP response. Specifically, the generated text is included in the body of the HTTP response in JSON format and sent to the device. The input is the generated new text data, and the output is the JSON data sent to the device.

[1008] Step 7: The user checks the generated sentence on the terminal and corrects it if necessary. Specifically, the user looks at the generated sentence displayed on the terminal and corrects it, for example, from "really" to "very." The input is the generated sentence, and the output is the sentence corrected by the user.

[1009] Step 8: After the user confirms the final text edits, the device resends the text to the server. Specifically, the revised text is repackaged in JSON format and sent to the server's API endpoint as an HTTP POST request. The input is the text edited by the user, and the output is the JSON data resent to the server.

[1010] Step 9: The server saves the revised final sentence in the database and publishes it on the CGM service. Specifically, the server inserts the final sentence into the appropriate table in the database and sets it up for publication so that other users can view it. The input is the revised final sentence data, and the output is the data saved in the database and the sentence published on the CGM service.

[1011] (Application example 2)

[1012] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1013] Current online content generation services pose a risk of identifying individuals through their unique writing style and phrasing. Furthermore, when users want to express their feelings frankly, there is a lack of ways to generate text that maintains their emotions while also increasing anonymity. Therefore, there is a need to provide an environment where users can post reviews and comments with peace of mind.

[1014] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify writing style and phrasing habits, means for generating text with increased anonymity by removing writing style and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, emotion analysis means for analyzing the emotion of the text entered by the user, means for generating a prompt text based on the analyzed emotion information and generating a text that reflects the emotion while maintaining anonymity, means for anonymizing specific words and phrases to increase the anonymity of the generated text, and means for transmitting the final corrected text back to the user's terminal. This allows users to safely post text that reflects their emotions while avoiding identification based on their unique writing style and phrasing.

[1015] "Means for receiving text entered by a user" refers to a process or device that allows the server to receive text data sent from the user's terminal.

[1016] "Means for analyzing received text and identifying writing style and phrasing" refers to a process or program that uses natural language processing technology to analyze input text and extract characteristics of its writing style and expression.

[1017] "Means for generating text with enhanced anonymity by removing stylistic and phrasing characteristics" refers to a technology that uses the analysis results to generate new text with enhanced anonymity by eliminating specific stylistic and phrasing characteristics.

[1018] "Means for transmitting the generated text to the user's terminal" refers to a communication process or device for returning the text generated by the server to the user's terminal.

[1019] The "means for the user to check and correct the generated text" refers to an interface and function that allows the user to check the generated text and correct it if necessary.

[1020] "Means for storing and publishing the final revised text in a database" refers to the process or device by which the final revised text by the user is recorded in a database and published in a form accessible to other users.

[1021] "Sentiment analysis means for analyzing the sentiment of text entered by a user" is a natural language processing technology for analyzing the sentiment of input text and identifying sentiment categories such as positive or negative.

[1022] "A means for generating prompt sentences based on analyzed emotional information and generating sentences that reflect the emotions while maintaining anonymity" is a technology that generates sentences that maintain anonymity while retaining the emotions of the original sentences based on the results of emotional analysis.

[1023] "Means for anonymizing specific words and phrases to increase the anonymity of generated text" refers to a technology that increases anonymity by extracting words and phrases that could identify individuals from text and converting them into other words.

[1024] The "means for retransmitting the final revised text to the user's terminal" refers to a process or device that retransmits the final version of the text that the user has completed the revisions to the user's terminal.

[1025] System Program

[1026] A system for realizing this application example includes the following means.

[1027] Hardware and Software

[1028] Hardware: Smartphones, servers

[1029] Software: Python, Hugging Face Transformers library

[1030] Data processing and calculation

[1031] This system sends text entered by a user on a smartphone to a server, which then analyzes the text and identifies writing style and phrasing habits. It also uses a sentiment analysis tool to analyze the sentiment of the entered text and generates a prompt based on that sentiment information. A generative AI model then generates a sentence that reflects the sentiment while maintaining anonymity, and sends the sentence back to the user's device.

[1032] Processing flow

[1033] 1. User input:

[1034] Users use their smartphones to enter reviews into the food delivery app, which are then sent to the server as HTTP requests.

[1035] 2. Parsing on the server:

[1036] When a review arrives at the server, it is first analyzed by a text analysis module to identify its stylistic and phrasing characteristics, and then by a sentiment analysis module to analyze the review's sentiment (e.g., positive, negative, etc.).

[1037] 3. Sentence generation:

[1038] Based on this analysis, a generative AI model is used to generate prompt sentences that reflect the sentiment while maintaining anonymity. For example, a review with a positive sentiment, "The pizza I ordered was so delicious I ate two right away!", can be converted to "The pizza I ordered was so delicious I ate two right away!"

[1039] 4. User Verification and Correction:

[1040] The generated text is then sent back to the user's smartphone, where the user can review it and make any necessary corrections. The corrected text is then sent back to the server as the final version.

[1041] 5. Final saving and publishing:

[1042] The final edited text is stored in a database and published as a review of the food delivery app that can be viewed by other users.

[1043] Specific examples

[1044] Prompt Sentence Examples

[1045] "Transform the original sentence into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right there!"

[1046] This system can be applied not only to food delivery apps but also to other online content generation services, providing an environment where users can post reviews and comments with confidence. It is also expected to improve the user experience by maintaining naturalness and emotional consistency in the writing.

[1047] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1048] Step 1:

[1049] Input: A user uses their smartphone to input a review into a food delivery app.

[1050] Action: For example, type, "The pizza I ordered was so good I ate two right there!"

[1051] Output: This entered review is sent by the device to the server as an HTTP request.

[1052] Step 2:

[1053] Input: Reviews received by the server.

[1054] How it works: The server first passes the review to a text analysis module to analyze its writing style and phrasing, and also uses a sentiment analysis module to analyze the review's sentiment (positive, negative, etc.).

[1055] Output: Stylistic and phrasing characteristics, as well as sentiment information.

[1056] Step 3:

[1057] Input: Stylistic and phrasing characteristics, emotional information.

[1058] What it does: The server generates a prompt based on this information, such as "Transform the original text into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right away!"

[1059] Output: The prompt statement and any parsing information needed to generate it.

[1060] Step 4:

[1061] Input: Prompt statement and parsing information.

[1062] How it works: The server uses a generative AI model to generate new, sentiment-based sentences while preserving anonymity, such as "The pizza I ordered was so delicious, I had two right away!"

[1063] Output: The new sentence generated.

[1064] Step 5:

[1065] Input: The new sentence to be generated.

[1066] What it does: The server resends this generated text to the user's device as an HTTP response.

[1067] Output: The generated text is displayed on the user's smartphone.

[1068] Step 6:

[1069] Input: User-generated text verification.

[1070] How it works: The user reviews the generated sentence and makes corrections if necessary, for example changing "very" to "really" to get "The pizza I ordered was so good, I had two right away!"

[1071] Output: The sentence corrected by the user.

[1072] Step 7:

[1073] Input: Final corrected text.

[1074] How it works: The user's device sends the revised text back to the server, which stores this final text in its database and makes it publicly available for other users to view.

[1075] Output: The final sentence stored in the database.

[1076] By following these steps, users can confidently submit reviews that reflect their feelings while avoiding being identified by their unique writing style or phrasing.

[1077] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1078] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1079] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1080] [Fourth embodiment]

[1081] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1082] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1083] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1084] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1085] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1086] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1087] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1088] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1089] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1090] 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.

[1091] 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.

[1092] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1093] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1094] The present invention is a system that generates texts with enhanced anonymity by removing specific writing styles and phrasing habits in order to provide an environment in which users can post with peace of mind on CGM services. The system of the present invention analyzes texts entered by users and utilizes a generative AI model to remove specific writing styles and phrasing characteristics.

[1095] Below, the program processing of this system is explained in natural language, with specific examples included.

[1096] Program processing flow

[1097] 1. User Input

[1098] Users access the CGM service using their device and enter the text they want to post in the input form. For example, they might enter, "I went to a delicious cafe. The coffee was amazing!"

[1099] 2. Sending text

[1100] The terminal sends the text entered by the user to the server as is. The entered text is sent to the server using an HTTP request.

[1101] 3. Analysis and generation using AI models

[1102] The server passes the received text to a text analysis module. The text analysis module extracts features to identify the style and phrasing of the text. After feature extraction is complete, the analysis results are passed to a generative AI model. The generative AI model generates new text that removes the identified style and phrasing features to increase anonymity. For example, it generates "I went to a new cafe. The coffee was very delicious."

[1103] 4. Sending the generated text

[1104] The server sends the generated text to the terminal, which returns it to the terminal as an HTTP response.

[1105] 5. User Review and Correction

[1106] The terminal displays the generated sentence received from the server to the user. The user can check the generated sentence and further modify the expressions and phrases as necessary. For example, the user can modify "very" to "really" to write "I went to a new cafe. The coffee was really delicious."

[1107] 6. Final Post

[1108] The user then sends the finalized text back to the server, which saves the revised final text in the CGM service database and makes it public. The published final text can then be viewed by other users.

[1109] Specific processing examples

[1110] Case: User B's post

[1111] User Input:

[1112] User B posts on social media, "I went to a new restaurant. The food was delicious."

[1113] Sending text:

[1114] The terminal sends this text to the server.

[1115] AI model analysis and generation:

[1116] The server receives the sentence, analyzes it with a text analysis module to identify stylistic and phrasing habits, and then the generative AI model generates the sentence, "I enjoyed eating at a new restaurant. The food was delicious."

[1117] Sending generated text:

[1118] The server sends the generated text to the terminal.

[1119] User review and correction:

[1120] User B checks the generated sentence and corrects "very" to "very." The final sentence is confirmed as "I enjoyed eating at the new restaurant. The food was delicious."

[1121] Last Post:

[1122] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[1123] This system allows users to post to the CGM service with peace of mind, avoiding identification based on specific writing styles or phrasing, which is expected to increase the number of users and page views of the service.

[1124] The processing flow will be explained below.

[1125] Step 1:

[1126] The user uses a device to access the input form of the CGM service and enters the text they want to post. For example, they might enter, "Today, we went to a new park and had a picnic. It was so much fun!"

[1127] Step 2:

[1128] When a user clicks the "Submit" button on the input form, the device sends the entered text to the server using an HTTP request, with the user's input data being sent to the server in the appropriate format.

[1129] Step 3:

[1130] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify its style and phrasing, specifically by analyzing the sentence structure and extracting specific phrases and expressions.

[1131] Step 4:

[1132] Based on the analysis results, the server passes the data to a generative AI model. The generative AI model takes the analysis results as input, removes stylistic and phrasing habits, and generates new sentences with increased anonymity. This process uses an algorithm that converts the sentences into different expressions while preserving the original meaning. For example, it generates the sentence, "Today we had a picnic at a new park. We had a lot of fun."

[1133] Step 5:

[1134] The server sends the generated new text to the device as an HTTP response, which includes the generated text and is displayed to the user.

[1135] Step 6:

[1136] The user checks the generated sentence on the device. If the user feels that a correction is necessary, the user corrects the generated sentence using the device interface. For example, changing "very" to "really." The final sentence is completed: "Today, we had a picnic at the new park. We really enjoyed it."

[1137] Step 7:

[1138] When the user confirms the final sentence, he clicks the "Confirm" button, and the terminal sends the corrected final sentence to the server again, again using an HTTP request.

[1139] Step 8:

[1140] The server receives the final text and stores it in a database, including any edits made by the user.

[1141] Step 9:

[1142] The server then publishes the final text on the CGM service, where it can be viewed by other users.

[1143] In this way, after going through a series of processes, the user's text is converted into a form that makes it difficult to identify and is then safely posted to the CGM service.

[1144] Example 1

[1145] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1146] In conventional CGM services, there is a risk that users can be identified by their unique writing style and phrasing in the text they post. This makes it difficult for users to post with confidence, which could lead to a decline in service usage. The lack of an appropriate system for increasing anonymity is an issue.

[1147] 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.

[1148] In this invention, the server includes: means for receiving text entered by a user; means for analyzing the received text and identifying writing style and phrasing habits; means for removing writing style and phrasing habits and generating text with increased anonymity based on the analysis results; means for sending the generated text to the user's terminal; means for the user to check and correct the generated text; means for saving and publishing the corrected final text in a database; a text analysis module for extracting features to identify specific writing style and phrasing habits; and generation AI model means for generating text with increased anonymity. This allows users to avoid being identified by their particular writing style or phrasing and to post to CGM services with peace of mind.

[1149] "User" refers to an individual or corporation that uses the CGM Service.

[1150] "Terminal" is a general term for the device used by a User to access and operate the CGM Service.

[1151] "Server" refers to a computer system that manages processing and data related to CGM services.

[1152] "Text" refers to the text data entered by a user to post on the CGM service.

[1153] "Means for receiving" refers to the function by which the server receives text sent by the user.

[1154] "Means for analysis" refers to the function for identifying the style and phrasing of the received text.

[1155] "Means of identification" refers to the function for extracting features in a text based on the analysis results.

[1156] "Means of generation" refers to the function for creating new text with increased anonymity based on the analysis results.

[1157] "Means for sending" refers to the function for returning the generated text to the user's device.

[1158] "Means for reviewing and correcting" refers to a function that allows the user to review the generated text and make corrections if necessary.

[1159] "Means for saving and publishing" refers to the functionality for saving the final revised text in a database and making it available for other users to view.

[1160] "Feature extraction" refers to the automated process of identifying specific stylistic or phrasing patterns in a piece of text.

[1161] "Text analysis module" refers to a software component for analyzing received text and extracting features.

[1162] A "generative AI model" refers to an artificial intelligence system that generates new text by removing identified stylistic and phrasing habits.

[1163] "CGM service" is a general term for online platforms that allow users to post and share content.

[1164] This invention is a system that allows users to generate sentences with enhanced anonymity by removing their specific writing style and phrasing. This system consists of the following main components:

[1165] User Input

[1166] A user accesses a CGM (Consumer Generated Media) service using a device and inputs a sentence into an input form. For example, consider a user who inputs the sentence, "I went to a delicious cafe. The coffee was great!"

[1167] Sending text

[1168] The device sends the text entered by the user to the server as an HTTP POST request, which is sent in JSON format and contains the text data.

[1169] Server analysis

[1170] The server passes the received text to a text analysis module, which uses natural language processing techniques (e.g., Spacy or NLTK) to extract features to identify the text's stylistic and phrasing habits.

[1171] Generation by generative AI model

[1172] The server passes the sentences after feature extraction to a generative AI model (such as OpenAI's GPT-3). The generative AI model generates new sentences that remove the identified stylistic and phrasing habits to enhance anonymity. For example, the sentence "I went to a delicious cafe. The coffee was great!" is converted to "I went to a new cafe. The coffee was really good."

[1173] Sending generated text

[1174] The server returns the generated text to the terminal as an HTTP response, with the appropriate Content-Type set in the response header.

[1175] User review and correction

[1176] The device displays the generated sentence to the user. The user can review the sentence and make corrections as necessary. For example, the user can change "very" to "really" to say, "I went to a new cafe. The coffee was really delicious."

[1177] Last Post

[1178] The user then sends the finalized sentence back to the server. The server saves the final, revised sentence in the CGM service's database and makes it publicly available for other users to view. For example, the published sentence might be, "I went to a new cafe. The coffee was really delicious."

[1179] This system allows users to post to CGM services with confidence, avoiding identification based on specific writing styles or phrasing. Below are some example prompts:

[1180] Example prompt sentence:

[1181] "Please make the following sentence more anonymous by removing certain styles and phrasing: We went to a new restaurant. The food was delicious."

[1182] By operating this system properly, it is expected that the number of service users and page views will increase.

[1183] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1184] Step 1:

[1185] A user accesses the CGM service using a device and enters a sentence into the input form. For example, they might enter the sentence, "I went to a delicious cafe. The coffee was amazing!" This input becomes the starting point for the next process.

[1186] Step 2:

[1187] The device converts the input text into JSON format and sends it to the server as the payload of an HTTP POST request. Specifically, it generates a POST request and sends the input text data to the server. This request becomes the input for the analysis process.

[1188] Step 3:

[1189] The server analyzes the received HTTP POST request and extracts the text data contained therein, which becomes the input for the text analysis module.

[1190] Step 4:

[1191] The server runs a text analysis module and performs feature extraction to identify the style and phrasing of the text. Specifically, it analyzes the text using natural language processing techniques (such as Spacy or NLTK) and generates a feature vector. This feature vector becomes the input for the generative AI model.

[1192] Step 5:

[1193] The server passes the feature vector to a generative AI model, which generates new sentences. The generative AI model (such as OpenAI's GPT-3) removes the identified stylistic and phrasing habits and generates new sentences with increased anonymity. For example, the input "I went to a delicious cafe. The coffee was great!" yields the output "I went to a new cafe. The coffee was delicious." This new sentence becomes the input for the next step.

[1194] Step 6:

[1195] The server converts the generated new text into JSON format and returns it to the terminal as an HTTP response. Specifically, it generates an HTTP response and sends it with the generated text included in the payload. This response data becomes the input for the terminal.

[1196] Step 7:

[1197] The terminal parses the received HTTP response, extracts the generated text, and displays it to the user. The user can check the new text on the screen and make corrections in the text box if necessary. This corrected text becomes the input for the final post.

[1198] Step 8:

[1199] The user then sends the revised text back to the server. Specifically, the device converts the revised text back into JSON format and sends it to the server as an HTTP POST request. This request becomes the input for the final saving process.

[1200] Step 9:

[1201] The server saves the final text after receiving the corrections in a database and makes it available on the CGM service. Specifically, it performs a saving process, stores the text in a database, and makes it available for other users to view. This data is the final output.

[1202] This allows users to avoid being identified by their particular writing style or phrasing, allowing them to post to CGM services with peace of mind.

[1203] (Application example 1)

[1204] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1205] In conventional CGM services and electronic messaging systems, users often face the risk of being identified through their unique writing style and phrasing, and anonymity is often not guaranteed. Furthermore, phishing attacks are on the rise in email and messaging systems, increasing the likelihood that users will be victimized by fraudulent messages. This makes protecting user privacy and improving security even more important.

[1206] 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.

[1207] In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify stylistic and phrasing habits, means for generating text with increased anonymity by removing the stylistic and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, means for converting the stylistic and phrasing characteristics of the received electronic message into standard expressions to prevent phishing attacks using electronic messages, means for recording the converted electronic message in a log file, and means for transmitting the converted electronic message to the user. This not only increases the anonymity of text entered by a user, but also makes it possible to prevent phishing attacks using electronic messages.

[1208] "User-entered text" refers to the text content entered by a user through a terminal or system.

[1209] "Means of receiving" refers to the function by which a server or system receives text sent by a user.

[1210] "Means of analyzing and identifying writing style and phrasing habits" refers to a function that uses machine learning models and natural language processing to analyze the writing style and unique expressions of input text and extract features.

[1211] The "means for generating sentences with enhanced anonymity" is a function that removes peculiar writing styles and phrases based on the characteristics obtained through analysis, and generates sentences in a standard writing style.

[1212] "Means for sending the generated text to the user's device" refers to a function that sends the highly anonymous text generated by the server back to the user's device.

[1213] "Means for checking and correcting" is a function that provides an interface for users to view the generated text and correct the content as necessary.

[1214] "Means for saving in a database and publishing" refers to the function of saving the final revised text on a server and publishing it in a form that other users can access.

[1215] "Means for converting stylistic and phrasing characteristics of electronic messages into standard expressions to prevent phishing attacks" refers to a function that converts the stylistic and idiosyncratic expressions of received electronic messages into standard business expressions, thereby preventing attacks before they occur.

[1216] "Means for recording to a log file" means a function that keeps a record of converted electronic messages for auditing and analysis.

[1217] A "means for sending an electronic message to a user" is a function that sends a converted secure electronic message to a user's device.

[1218] This invention is a system for preventing phishing attacks by anonymizing text so that users cannot be identified based on their unique writing style or phrasing habits. To realize this system, a program including the following means is required.

[1219] Program processing overview

[1220] 1. A means of receiving text entered by the user

[1221] Users access the system using a terminal and enter the text they want to post or the text of an email they received into the input form. For example, they could enter, "Please send us your bank details quickly! You'll receive a reward!"

[1222] 2. A means of analyzing received text and identifying stylistic and phrasing habits

[1223] The server receives the text sent by the user and analyzes it using Natural Language Processing (NLP) technology, which identifies stylistic and expressive features.

[1224] 3. A method for generating text with increased anonymity

[1225] The server uses a generative AI model to remove stylistic and phrasing quirks and convert the sentence into standard expressions. The generative AI model uses, for example, the OpenAI API. "Please send us your bank details quickly. We'll give you a reward!" is converted to "Could you please provide us with the following bank details as soon as possible? Thank you."

[1226] 4. A means of sending the generated text to the user's device

[1227] The server then sends the generated standard text back to the user's device using an HTTP response.

[1228] 5. A way for users to review and correct the generated text

[1229] The user can view the generated text received from the server and make further corrections as necessary. The user can freely edit the text displayed on the terminal.

[1230] 6. A means of storing and publishing the final revised text in a database

[1231] Once the user has finalized the text, it is sent back to the server and stored in the database, after which it is made publicly accessible to other users.

[1232] 7. Means of converting stylistic and phrasing features of electronic messages into standard expressions to prevent phishing attacks

[1233] The server converts the style and idioms of received emails into standard business style, particularly to prevent phishing attacks, for example converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[1234] 8. Means for recording converted electronic messages in a log file

[1235] All conversion processes are recorded in log files for later auditing and analysis, facilitating future problem solving and improvement planning.

[1236] 9. Means of sending the converted electronic message to the user

[1237] The server then sends the converted secure email to the user's device, again via an HTTP response.

[1238] Specific examples

[1239] As a specific example of use, if a security department staff member receives a suspicious email such as "Please send your bank information quickly. You will receive a reward!", the system will operate as follows.

[1240] Example prompt for a generative AI model:

[1241] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[1242] Submit your bank details quickly and you'll be rewarded!

[1243] This prompt provides users with standardized text generated by an AI model (e.g., "Could you please provide the following banking information as soon as possible? Thank you."), helping to prevent phishing attacks. The system achieves high levels of anonymity and security through communication between the server and the device, natural language processing, and the use of generative AI models.

[1244] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1245] Step 1:

[1246] The user inputs a sentence into the terminal. The input sentence is something like "Please send your bank information quickly. You will receive a reward!". The terminal sends this sentence to the server via an HTTP request.

[1247] Step 2:

[1248] The server analyzes the text received from the user. To do this, the text is passed to a natural language processing module to identify stylistic and phrasing patterns. This analysis is performed using text feature extraction algorithms.

[1249] Step 3:

[1250] Based on the analysis results, the server uses a generative AI model to generate new sentences that remove stylistic and phrasing habits. Specifically, the server passes the following prompt sentence as input to the generative AI model:

[1251] Please remove any stylistic or phrasing features from the following email and convert it to standard business email format:

[1252] Submit your bank details quickly and you'll be rewarded!

[1253] The output will be a secure sentence such as "Could you please provide the following bank details as soon as possible? Thank you."

[1254] Step 4:

[1255] The generated text is sent from the server to the terminal as an HTTP response, and the terminal displays the generated text to the user.

[1256] Step 5:

[1257] The user checks the generated text and makes any necessary corrections. The corrections are then sent back to the server as an HTTP request from the terminal.

[1258] Step 6:

[1259] The server saves the final corrected sentence in the database. The saved data is the final corrected sentence. For example, "Could you please provide the following bank information as soon as possible? Thank you." is saved in the database.

[1260] Step 7:

[1261] The server analyzes received electronic messages and converts stylistic and phrasing features into standard expressions to prevent phishing attacks, for example, converting the message "Your account has been compromised! Click here to fix it now" to "There is a problem with your account that requires immediate attention. Click here."

[1262] Step 8:

[1263] The converted electronic messages are logged in a log file. The server stores the conversion process in a log file (e.g. "sanitized_emails_log.txt") for audit purposes.

[1264] Step 9:

[1265] Finally, the converted electronic message is sent from the server to the user's terminal and displayed to the user, thereby protecting the user from phishing attacks.

[1266] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1267] This invention relates to a system that generates text with enhanced anonymity by removing idiosyncratic writing styles and phrasing, in order to provide an environment in which users can post content safely on CGM services. It also includes an emotion engine that recognizes the emotions of user input and generated text and incorporates them into the system.

[1268] Below, the program processing of this system is explained in natural language, with specific examples included.

[1269] Program processing flow

[1270] 1. User Input

[1271] Users access the CGM service through their devices and enter the text they want to post in the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[1272] 2. Sending text

[1273] When the user clicks the "Send" button, the device sends the entered text to the server via an HTTP request.

[1274] 3. Analysis and generation using AI models and emotion engines

[1275] The server first passes the received text to a text analysis module, which identifies the text's style and phrasing. The analysis results and the text are then passed to an emotion engine to recognize the text's emotion.

[1276] Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate a more anonymous sentence. This process preserves the emotion of the original sentence while removing the stylistic and phrasing characteristics, resulting in a new sentence. For example, "Today was a really great day! I had a great time with my friends."

[1277] 4. Sending the generated text

[1278] The server sends the generated new text to the user's device as an HTTP response, where it is displayed.

[1279] 5. User Review and Correction

[1280] The user can review the generated sentence and make corrections as needed. The emotion engine recognizes and preserves emotions even as the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the sentence becomes, "Today was a wonderful day! I had a wonderful time with my friends."

[1281] 6. Final Post

[1282] Once the user has confirmed the final sentence, the device sends it back to the server, which then stores the final sentence in a database and makes it available on the CGM service.

[1283] Specific processing examples

[1284] Case: User B's post

[1285] User Input:

[1286] User B posts on social media, "I went to a new cafe. The coffee was delicious and the atmosphere was fun."

[1287] Sending text:

[1288] The terminal sends this text to the server.

[1289] Analysis and generation by AI model and emotion engine:

[1290] The server analyzes the text using a text analysis module to identify characteristics of the writing style and phrasing. Next, an emotion engine recognizes the emotion of the text. For example, positive emotion is recognized. The generative AI model reflects this emotion information and generates a response such as, "I visited a new cafe. The coffee was delicious and the atmosphere was very pleasant."

[1291] Sending generated text:

[1292] The server sends the generated text to the terminal.

[1293] User review and correction:

[1294] User B checks the generated sentence and corrects "very" to "extremely", completing the final sentence: "I visited a new cafe. The coffee was delicious and the atmosphere was very enjoyable."

[1295] Last Post:

[1296] The device then sends the revised text back to the server, which stores it in a database and makes it publicly available on the CGM service, where other users can view the final post.

[1297] This system allows users to avoid being identified by their particular writing style or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of the emotion engine is expected to improve the user experience by maintaining naturalness in writing and consistency in emotions.

[1298] The processing flow will be explained below.

[1299] Step 1:

[1300] The user uses a device to access the input form of the CGM service and enters the sentence they want to post. For example, they might enter, "Today was beautiful weather, so I had a picnic in the park with my family. It was really fun!"

[1301] Step 2:

[1302] When a user clicks the "Submit" button on the input form, the terminal sends the input text to the server using an HTTP request. This request includes the user's input text.

[1303] Step 3:

[1304] The server passes the received text to a text analysis module, which uses natural language processing techniques to analyze the text and identify stylistic and phrasing habits, such as the use of specific punctuation marks or frequently used phrases.

[1305] Step 4:

[1306] The server passes the analysis results to the emotion engine, which recognizes the emotion of the input text. For example, it determines that expressions such as "great weather" and "it was fun" contain positive emotions.

[1307] Step 5:

[1308] The emotion engine returns the recognized emotion information to the server, which then prepares the analysis results and emotion information to be input into the generative AI model.

[1309] Step 6:

[1310] The server inputs the analysis results and emotional information into the generative AI model. Based on this information, the generative AI model removes stylistic and phrasing habits and generates sentences with increased anonymity while preserving the original emotion. For example, it generates a sentence like, "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was a lot of fun."

[1311] Step 7:

[1312] The server sends the generated new text to the device as an HTTP response, which includes the generated text.

[1313] Step 8:

[1314] The device displays the generated text received from the server to the user, who then checks the text to see if it expresses the appropriate emotion.

[1315] Step 9:

[1316] The user can make necessary corrections to the generated sentences, for example, changing "very" to "extremely" to write "The weather was very nice today, so I enjoyed a picnic in the park with my family. It was really fun."

[1317] Step 10:

[1318] When the user completes the edits and clicks the "Confirm" button, the terminal sends the final edited text back to the server, again using an HTTP request.

[1319] Step 11:

[1320] The server receives the final text and stores it in the CGM service's database, where it can be viewed by other users.

[1321] Step 12:

[1322] The server will then make the final text publicly available for other users to view on the CGM service, which will encourage users to feel more comfortable posting content and increase overall user numbers and engagement for the service.

[1323] In this way, the system of the present invention generates natural sentences that reflect emotions while eliminating particular writing styles and phrasing habits, providing an environment in which users can post to CGM services with peace of mind.

[1324] Example 2

[1325] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1326] In modern CGM services, users are required to maintain anonymity and post with peace of mind, avoiding identification based on specific writing styles or phrasing. However, existing systems have shortcomings in ensuring this anonymity, which can lead to issues with users not feeling secure enough. Furthermore, it is difficult to generate text that accurately reflects the sentiment of the user, which can lead to a poor user experience.

[1327] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text and identifying idiosyncrasies of writing style and phrasing, means for removing idiosyncrasies of writing style and phrasing based on the analysis results and generating text with increased anonymity, means for recognizing the emotion of the generated text and retaining that emotion, means for transmitting the generated text to the user's terminal, means for the user to confirm and correct the generated text, and means for saving the corrected final text in a database and making it public. This makes it possible to generate natural-looking text that reflects emotion while preventing identification based on idiosyncratic writing style and phrasing, allowing users to post their text to the CGM service with confidence.

[1328] "User" means a user who accesses the CGM Service and posts or views content.

[1329] A "terminal" is an electronic device used by a user, such as a computer, smartphone, or tablet.

[1330] A "server" is a computer system that receives a user's request, processes the data in accordance with the request, and returns the data.

[1331] "Text" refers to text data entered or generated by a user.

[1332] "Analysis" is a process for identifying the style and phrasing of the input text.

[1333] "Identifying" means identifying the specific stylistic and phrasing patterns in a piece of writing.

[1334] "Generation" involves removing identified stylistic and phrasing habits to create new sentences.

[1335] "Emotion recognition" means determining emotions such as positive, negative, or neutral from the content of a sentence.

[1336] "Preserving emotion" means maintaining the emotion of the original sentence in the generated new sentence.

[1337] "Send" means to transfer data from the server to the user's terminal.

[1338] "Verification" means that the user checks the generated text and confirms its content.

[1339] "Correction" means that the user makes changes to the generated text.

[1340] "Saving" means recording the final corrected text in the database.

[1341] "Publishing" means making the saved final text available for other users to view.

[1342] "Anonymity" means keeping users difficult to identify.

[1343] A "generative AI model" is an artificial intelligence program that uses natural language processing technology to generate new sentences.

[1344] A "prompt sentence" is text that gives instructions to the generative AI model on the sentence to output.

[1345] This invention is a system that can generate text with enhanced anonymity by removing a user's particular writing style and phrasing habits, and aims to provide an environment in which users can post content safely, particularly in CGM services. By combining a generative AI model and an emotion engine, this system neutralizes the writing style while recognizing and preserving the emotion of the text, ensuring anonymity.

[1346] First, the user accesses the CGM service using a device and enters the text they want to post into the input form. For example, they might enter, "Today was a great day! I had a great time with my friends."

[1347] Next, when the user clicks the "Send" button, the device sends the input text as an HTTP request to the server, which then passes the received text to a text analysis module to identify stylistic and phrasing patterns. This analysis is performed using a Natural Language Processing (NLP) library (e.g., spaCy or NLTK).

[1348] Based on the analysis results, the server passes the sentence to an emotion engine, which uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to recognize the sentiment of the sentence, such as positive, negative, or neutral. Based on the obtained sentiment information and the identified stylistic features, the server uses a generative AI model (e.g., OpenAI's GPT-3) to generate a sentence with increased anonymity. This generation process uses a prompt such as: "Please neutralize the stylistic features while preserving the sentiment based on the original sentence."

[1349] The generated sentence might be something like, "Today was a really great day! I had a great time with my friends." This generated sentence is sent from the server to the user's device as an HTTP response.

[1350] The user can review the generated sentences on their device and make corrections as necessary. The emotion engine recognizes and retains emotions even when the user makes corrections, allowing for appropriate expressions. For example, by changing "really" to "very," the user can write, "Today was a wonderful day! I had a wonderful time with my friends."

[1351] Finally, when the user confirms the final sentence, the device sends this sentence back to the server, which stores the final sentence in a database and makes it available on the CGM service.

[1352] This system allows users to avoid being identified by specific writing styles or phrasing, and allows them to post to CGM services with confidence, reflecting their emotions. The introduction of an emotion engine and generative AI model maintains the naturalness of writing and consistency of emotion, which is expected to improve the user experience.

[1353] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1354] Step 1: The user accesses the CGM service through their device and enters the text they want to post in the input form. Specifically, they enter the text "Today was a great day! I had a great time with my friends" into the text box. The input is text data, and the output is the input text recorded on the device.

[1355] Step 2: When the user clicks the "Submit" button, the terminal sends the entered text to the server via an HTTP request. Specifically, the terminal packages the user's input in JSON format and sends it to the server's API endpoint as an HTTP POST request. The input is the entered text, and the output is the JSON data sent to the server.

[1356] Step 3: The server passes the received text to a text analysis module, which performs processing to identify stylistic and phrasing patterns. Specifically, the server uses a Natural Language Processing (NLP) library (e.g., spaCy or NLTK) to tokenize the text and extract contextual and stylistic features. The input is the JSON data sent in the HTTP request, and the output is a data structure containing the analysis results.

[1357] Step 4: The server passes the analysis results and the text to the emotion engine to recognize the sentiment of the text. Specifically, the server uses a sentiment analysis model (e.g., Google Cloud Natural Language API or IBM Watson Tone Analyzer) to identify the sentiment of the text, such as positive, negative, or neutral. The input is the text analysis result data, and the output is data with sentiment labels.

[1358] Step 5: Based on the obtained emotional information and the identified stylistic features, the server uses a generative AI model to generate sentences with enhanced anonymity. Specifically, the server inputs a prompt to the generative AI model: "Please neutralize the style while preserving the emotion based on the original sentence." and obtains the generated sentence. The input is the original sentence data with emotion labels and the prompt sentence, and the output is the generated new sentence.

[1359] Step 6: The server sends the generated new text to the user's device as an HTTP response. Specifically, the generated text is included in the body of the HTTP response in JSON format and sent to the device. The input is the generated new text data, and the output is the JSON data sent to the device.

[1360] Step 7: The user checks the generated sentence on the terminal and corrects it if necessary. Specifically, the user looks at the generated sentence displayed on the terminal and corrects it, for example, from "really" to "very." The input is the generated sentence, and the output is the sentence corrected by the user.

[1361] Step 8: After the user confirms the final text edits, the device resends the text to the server. Specifically, the revised text is repackaged in JSON format and sent to the server's API endpoint as an HTTP POST request. The input is the text edited by the user, and the output is the JSON data resent to the server.

[1362] Step 9: The server saves the revised final sentence in the database and publishes it on the CGM service. Specifically, the server inserts the final sentence into the appropriate table in the database and sets it up for publication so that other users can view it. The input is the revised final sentence data, and the output is the data saved in the database and the sentence published on the CGM service.

[1363] (Application example 2)

[1364] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1365] Current online content generation services pose a risk of identifying individuals through their unique writing style and phrasing. Furthermore, when users want to express their feelings frankly, there is a lack of ways to generate text that maintains their emotions while also increasing anonymity. Therefore, there is a need to provide an environment where users can post reviews and comments with peace of mind.

[1366] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving text entered by a user, means for analyzing the received text to identify writing style and phrasing habits, means for generating text with increased anonymity by removing writing style and phrasing habits based on the analysis results, means for transmitting the generated text to the user's terminal, means for the user to check and correct the generated text, means for storing the corrected final text in a database and making it public, emotion analysis means for analyzing the emotion of the text entered by the user, means for generating a prompt text based on the analyzed emotion information and generating a text that reflects the emotion while maintaining anonymity, means for anonymizing specific words and phrases to increase the anonymity of the generated text, and means for transmitting the final corrected text back to the user's terminal. This allows users to safely post text that reflects their emotions while avoiding identification based on their unique writing style and phrasing.

[1367] "Means for receiving text entered by a user" refers to a process or device that allows the server to receive text data sent from the user's terminal.

[1368] "Means for analyzing received text and identifying writing style and phrasing" refers to a process or program that uses natural language processing technology to analyze input text and extract characteristics of its writing style and expression.

[1369] "Means for generating text with enhanced anonymity by removing stylistic and phrasing characteristics" refers to a technology that uses the analysis results to generate new text with enhanced anonymity by eliminating specific stylistic and phrasing characteristics.

[1370] "Means for transmitting the generated text to the user's terminal" refers to a communication process or device for returning the text generated by the server to the user's terminal.

[1371] The "means for the user to check and correct the generated text" refers to an interface and function that allows the user to check the generated text and correct it if necessary.

[1372] "Means for storing and publishing the final revised text in a database" refers to the process or device by which the final revised text by the user is recorded in a database and published in a form accessible to other users.

[1373] "Sentiment analysis means for analyzing the sentiment of text entered by a user" is a natural language processing technology for analyzing the sentiment of input text and identifying sentiment categories such as positive or negative.

[1374] "A means for generating prompt sentences based on analyzed emotional information and generating sentences that reflect the emotions while maintaining anonymity" is a technology that generates sentences that maintain anonymity while retaining the emotions of the original sentences based on the results of emotional analysis.

[1375] "Means for anonymizing specific words and phrases to increase the anonymity of generated text" refers to a technology that increases anonymity by extracting words and phrases that could identify individuals from text and converting them into other words.

[1376] The "means for retransmitting the final revised text to the user's terminal" refers to a process or device that retransmits the final version of the text that the user has completed the revisions to the user's terminal.

[1377] System Program

[1378] A system for realizing this application example includes the following means.

[1379] Hardware and Software

[1380] Hardware: Smartphones, servers

[1381] Software: Python, Hugging Face Transformers library

[1382] Data processing and calculation

[1383] This system sends text entered by a user on a smartphone to a server, which then analyzes the text and identifies writing style and phrasing habits. It also uses a sentiment analysis tool to analyze the sentiment of the entered text and generates a prompt based on that sentiment information. A generative AI model then generates a sentence that reflects the sentiment while maintaining anonymity, and sends the sentence back to the user's device.

[1384] Processing flow

[1385] 1. User input:

[1386] Users use their smartphones to enter reviews into the food delivery app, which are then sent to the server as HTTP requests.

[1387] 2. Parsing on the server:

[1388] When a review arrives at the server, it is first analyzed by a text analysis module to identify its stylistic and phrasing characteristics, and then by a sentiment analysis module to analyze the review's sentiment (e.g., positive, negative, etc.).

[1389] 3. Sentence generation:

[1390] Based on this analysis, a generative AI model is used to generate prompt sentences that reflect the sentiment while maintaining anonymity. For example, a review with a positive sentiment, "The pizza I ordered was so delicious I ate two right away!", can be converted to "The pizza I ordered was so delicious I ate two right away!"

[1391] 4. User Verification and Correction:

[1392] The generated text is then sent back to the user's smartphone, where the user can review it and make any necessary corrections. The corrected text is then sent back to the server as the final version.

[1393] 5. Final saving and publishing:

[1394] The final edited text is stored in a database and published as a review of the food delivery app that can be viewed by other users.

[1395] Specific examples

[1396] Prompt Sentence Examples

[1397] "Transform the original sentence into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right there!"

[1398] This system can be applied not only to food delivery apps but also to other online content generation services, providing an environment where users can post reviews and comments with confidence. It is also expected to improve the user experience by maintaining naturalness and emotional consistency in the writing.

[1399] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1400] Step 1:

[1401] Input: A user uses their smartphone to input a review into a food delivery app.

[1402] Action: For example, type, "The pizza I ordered was so good I ate two right there!"

[1403] Output: This entered review is sent by the device to the server as an HTTP request.

[1404] Step 2:

[1405] Input: Reviews received by the server.

[1406] How it works: The server first passes the review to a text analysis module to analyze its writing style and phrasing, and also uses a sentiment analysis module to analyze the review's sentiment (positive, negative, etc.).

[1407] Output: Stylistic and phrasing characteristics, as well as sentiment information.

[1408] Step 3:

[1409] Input: Stylistic and phrasing characteristics, emotional information.

[1410] What it does: The server generates a prompt based on this information, such as "Transform the original text into a more anonymous form, maintaining the following sentiment: POSITIVE. Sentence: The pizza I ordered was so delicious I ate two right away!"

[1411] Output: The prompt statement and any parsing information needed to generate it.

[1412] Step 4:

[1413] Input: Prompt statement and parsing information.

[1414] How it works: The server uses a generative AI model to generate new, sentiment-based sentences while preserving anonymity, such as "The pizza I ordered was so delicious, I had two right away!"

[1415] Output: The new sentence generated.

[1416] Step 5:

[1417] Input: The new sentence to be generated.

[1418] What it does: The server resends this generated text to the user's device as an HTTP response.

[1419] Output: The generated text is displayed on the user's smartphone.

[1420] Step 6:

[1421] Input: User-generated text verification.

[1422] How it works: The user reviews the generated sentence and makes corrections if necessary, for example changing "very" to "really" to get "The pizza I ordered was so good, I had two right away!"

[1423] Output: The sentence corrected by the user.

[1424] Step 7:

[1425] Input: Final corrected text.

[1426] How it works: The user's device sends the revised text back to the server, which stores this final text in its database and makes it publicly available for other users to view.

[1427] Output: The final sentence stored in the database.

[1428] By following these steps, users can confidently submit reviews that reflect their feelings while avoiding being identified by their unique writing style or phrasing.

[1429] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1430] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1431] 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 robot 414.

[1432] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1433] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1434] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1435] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1436] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1437] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1438] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1439] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1440] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1441] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1442] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1443] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1444] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1445] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1446] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1447] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1448] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1449] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1450] The following is further disclosed regarding the above embodiment.

[1451] (Claim 1)

[1452] means for receiving text entered by a user;

[1453] A means of analyzing received text to identify stylistic and phrasing habits;

[1454] a means for removing stylistic and phrasing habits based on the analysis results and generating sentences with enhanced anonymity;

[1455] A means for transmitting the generated text to a user's terminal;

[1456] A means for the user to review and correct the generated text;

[1457] The system includes a means for storing and publishing the final corrected text in a database.

[1458] (Claim 2)

[1459] 10. The system of claim 1, further comprising means for anonymizing specific words or phrases in the generated text to enhance the anonymity of the generated text.

[1460] (Claim 3)

[1461] 2. The system of claim 1, further comprising means for transmitting the user-modified text back to the server, and for storing and publishing the final text.

[1462] "Example 1"

[1463] (Claim 1)

[1464] means for receiving text entered by a user;

[1465] A means of analyzing received text to identify stylistic and phrasing habits;

[1466] a means for removing stylistic and phrasing habits based on the analysis results and generating sentences with enhanced anonymity;

[1467] A means for transmitting the generated text to a user's terminal;

[1468] A means for the user to review and correct the generated text;

[1469] A means to store and publish the final revised text in a database;

[1470] a text analysis module that performs feature extraction to identify specific writing styles and phrasing;

[1471] A generative AI model means for generating sentences with enhanced anonymity;

[1472] A system including:

[1473] (Claim 2)

[1474] 10. The system of claim 1, further comprising means for anonymizing specific words or phrases in the generated text to enhance the anonymity of the generated text.

[1475] (Claim 3)

[1476] 2. The system of claim 1, further comprising means for transmitting the user-modified text back to the server, and for storing and publishing the final text.

[1477] "Application Example 1"

[1478] (Claim 1)

[1479] means for receiving text entered by a user;

[1480] A means of analyzing received text to identify stylistic and phrasing habits;

[1481] a means for removing stylistic and phrasing habits based on the analysis results and generating sentences with enhanced anonymity;

[1482] A means for transmitting the generated text to a user's terminal;

[1483] A means for the user to review and correct the generated text;

[1484] A means to store and publish the final revised text in a database;

[1485] a means for converting stylistic and phrasing features of received electronic messages into standard expressions to prevent phishing attacks;

[1486] means for recording the converted electronic messages in a log file;

[1487] means for transmitting the converted electronic message to a user;

[1488] A system including:

[1489] (Claim 2)

[1490] 10. The system of claim 1, further comprising means for anonymizing specific words or phrases in the generated text to enhance the anonymity of the generated text.

[1491] (Claim 3)

[1492] 2. The system of claim 1, further comprising means for transmitting the user-modified text back to the server, and for storing and publishing the final text.

[1493] "Example 2: Combining Emotion Engines"

[1494] (Claim 1)

[1495] means for receiving text entered by a user;

[1496] A means of analyzing received text to identify stylistic and phrasing habits;

[1497] a means for removing stylistic and phrasing habits based on the analysis results and generating sentences with enhanced anonymity;

[1498] a means for recognizing and preserving the sentiment of the generated text;

[1499] A means for transmitting the generated text to a user's terminal;

[1500] A means for the user to review and correct the generated text;

[1501] The system includes a means for storing and publishing the final corrected text in a database.

[1502] (Claim 2)

[1503] 10. The system of claim 1, further comprising means for anonymizing specific words or phrases in the generated text to enhance the anonymity of the generated text.

[1504] (Claim 3)

[1505] 2. The system of claim 1, further comprising means for transmitting the user-modified text back to the server, and for storing and publishing the final text.

[1506] "Application example 2 when combining emotion engines"

[1507] (Claim 1)

[1508] means for receiving text entered by a user;

[1509] A means of analyzing received text to identify stylistic and phrasing habits;

[1510] a means for removing stylistic and phrasing habits based on the analysis results and generating sentences with enhanced anonymity;

[1511] A means for transmitting the generated text to a user's terminal;

[1512] A means for the user to review and correct the generated text;

[1513] A means to store and publish the final revised text in a database;

[1514] A sentiment analysis means for analyzing the sentiment of a sentence entered by a user;

[1515] A means for generating prompt sentences based on the analyzed emotional information, and generating sentences that reflect the emotions while maintaining anonymity;

[1516] A means to anonymize specific words or phrases to increase the anonymity of the generated text;

[1517] A means to send the final revised text back to the user's device

[1518] A system including:

[1519] (Claim 2)

[1520] 2. The system according to claim 1, further comprising a sentiment analysis means for analyzing sentiment of a sentence input by a user.

[1521] (Claim 3)

[1522] 2. The system of claim 1, further comprising means for transmitting the final revised text back to the user's terminal, and for saving and publishing the final text. [Explanation of symbols]

[1523] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving text entered by a user; A means of analyzing received text to identify stylistic and phrasing habits; a means for removing stylistic and phrasing habits based on the analysis results and generating sentences with enhanced anonymity; A means for transmitting the generated text to a user's terminal; A means for the user to review and correct the generated text; The system includes a means for storing and publishing the final corrected text in a database.

2. 2. The system of claim 1, further comprising means for anonymizing specific words or phrases in the generated text to enhance the anonymity of the generated text.

3. 2. The system according to claim 1, further comprising means for transmitting the text corrected by the user back to the server, and for storing and publishing the final text.

Images