System

Abstract

A system is provided.SOLUTION: A system comprising: means for analyzing a user's comments in real-time and scoring the degree of aggressiveness; means for suggesting improved rewrite sentences for aggressive comments; and means for providing a gamification component that ranks based on the user's average earned score.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] Comment sections on the Internet are often filled with offensive posts and slander, hindering healthy discussion and information exchange. In particular, those who make these offensive comments are often unaware of their own aggression, and if this situation is left unchecked, the quality of the interactive environment will continue to decline. The problem that this invention aims to solve is to suppress offensive posts in comment sections and create a healthy interactive environment. [Means for solving the problem]

[0005] This invention solves the above-mentioned problems by providing a system that includes a means for analyzing user comments in real time and scoring the degree of offensiveness, a means for suggesting improved rewrites for offensive comments, and a means for providing a gamification element that ranks comments based on the average score earned by users. When a user enters a comment, the content is instantly analyzed and an offensiveness score is assigned. Rewrite suggestions are presented for offensive comments, and the user can choose whether to adopt them. Through this process, users are encouraged to post more appropriate comments. Furthermore, ranking comments based on the user's average score encourages users to learn and improve independently, establishing a healthy dialogue environment.

[0006] "User" refers to any individual or entity that uses the System to enter and post comments.

[0007] "Comment" means a text message that a User enters and intends to post.

[0008] "Real-time" means that processing occurs simultaneously or nearly simultaneously with user input.

[0009] "Analysis" refers to the process by which the system evaluates the content of comments and determines their level of offensiveness.

[0010] "Offensiveness" refers to the degree to which the content of a comment is negative, insulting, or provocative towards others.

[0011] "Scoring" refers to expressing the aggressiveness of a comment in quantitative numbers.

[0012] A "rewrite" is an alternative text that rewrites the original comment to be more appropriate and less offensive.

[0013] "Suggestion" refers to the act of the system providing a rewritten sentence to the user and prompting them to choose whether or not to adopt it.

[0014] "Ranking" refers to the process of determining relative rankings or ratings based on user behavior and posts.

[0015] "Gamification elements" refer to mechanisms that incorporate game elements into a system to encourage users to voluntarily improve their behavior while having fun.

[0016] A "server" refers to a computer or network device that performs back-end processing for a system.

[0017] "Terminal" refers to the device used by a User to enter comments and communicate with the System.

[0018] The "system" refers to a single system consisting of multiple components that execute a series of processes: analyzing user comments, assessing their level of offensiveness, suggesting rewrites, and ranking them. [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 embodiment of the present invention provides a system that analyzes in real time how the content of a comment entered by a user will be perceived by others and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average points earned by users.

[0041] System Overview

[0042] The system consists of the following main components:

[0043] User terminal: A device used to input comments and interact with the system.

[0044] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, and manages rankings.

[0045] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0046] Program processing flow

[0047] User comment input

[0048] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[0049] Sending a comment analysis request

[0050] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0051] AI-powered comment analysis

[0052] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better."

[0053] Returning analysis results

[0054] The server creates a response containing the offensiveness score and the rewritten text and sends it back to the user's device.

[0055] Presenting and selecting rewrite proposals

[0056] The user device receives the response from the server and presents the offensiveness score and rewritten sentence to the user. The user can choose whether to adopt the proposed rewritten sentence. For example, if the user adopts the rewritten sentence, the content is saved as the final comment on the user device.

[0057] Submitting final comments

[0058] The user device sends the final comment back to the server, and this request includes the user ID and the final comment.

[0059] Ranking System Update

[0060] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[0061] Specific examples

[0062] For example, consider a case where a user enters a comment such as, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI. It is determined to have a high offensive score, and a rewrite proposal is generated. The rewrite proposal, "There are many problems with this news. I hope that the people involved will improve their response," is presented to the user. The user adopts the rewrite proposal and sends the final comment to the server. The server saves this comment in a database and updates the ranking.

[0063] This process will help curb offensive posts in the comments section and provide a healthy interactive environment.

[0064] The processing flow will be explained below.

[0065] Specific processing steps of the program

[0066] Step 1:

[0067] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[0068] Terminal: Obtains and temporarily stores comments entered by users.

[0069] Step 2:

[0070] On your device: Create a request to analyze the saved comments. This request includes the user ID and the comment.

[0071] Terminal: Sends the created request to the server.

[0072] Step 3:

[0073] Server: Receives requests from user devices.

[0074] Server: The received comments are passed to the AI ​​analysis model for analysis, where the offensiveness score of the comment is calculated and a rewrite proposal is generated.

[0075] Step 4:

[0076] Server: Create a response containing an offensiveness score and a rewrite suggestion as the analysis result. For example, if the offensiveness score is 80 / 100, the rewrite suggestion is "There are many problems with this news. We hope that the relevant parties will respond better."

[0077] Server: Returns the created response to the user device.

[0078] Step 5:

[0079] Terminal: Receives the response sent back from the server and displays the analysis results to the user.

[0080] Device: Presents the offensiveness score and rewrite suggestions, prompting the user to choose whether or not to adopt the rewrite suggestions.

[0081] User: Checks the rewrite proposal and decides whether to adopt it. For example, adopt the rewrite proposal "There are many problems with this news. I hope that the people involved will respond better."

[0082] Step 6:

[0083] On the device: The user creates a request to send the final comment (in this case, the rewrite proposal) back to the server. This request includes the user ID and the final comment.

[0084] Terminal: Send a request to the server containing the final comment.

[0085] Step 7:

[0086] Server: Receives the final comment from the user terminal.

[0087] Server: Stores received comments in a database.

[0088] Server: Updates the average score of users based on the offensiveness score of their comments and processes the ranking.

[0089] Server: Returns a response to the user device indicating that the ranking update has been completed.

[0090] For example, if a user enters an offensive comment such as "This news is terrible. Aren't all the people involved stupid?", the system will analyze the comment using AI and assign a high offensiveness score (80 / 100). At the same time, the system will generate a rewrite suggestion, "There are many problems with this news. I hope that the people involved will improve their response," and present it to the user. If the user adopts this rewrite suggestion, it will be sent to the server as the final comment and saved in the database. This will also update the user's average score, which will be reflected in the rankings. In this way, users are encouraged to voluntarily post less offensive comments.

[0091] Example 1

[0092] 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."

[0093] In recent years, the number of offensive comments has increased in the comment sections of online platforms, hindering a healthy dialogue environment. There is a need for a system that can detect such offensive comments in real time and allow users to correct them themselves. Gamification elements are also necessary to increase user motivation.

[0094] 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.

[0095] In this invention, the server includes: means for analyzing user comments in real time and scoring the degree of offensiveness; means for proposing improved rewrites for offensive comments; means for providing a gamification element that ranks comments based on the average points earned by users; a terminal for inputting and submitting comments and presenting rewrites; a server equipped with an analytical model for generating offensiveness scores and rewrite suggestions; and a system for displaying rankings based on user comment scores. This makes it possible to provide a healthy interactive environment and increase user motivation by detecting offensive comments in real time and presenting improvement suggestions.

[0096] "User" refers to the end user who enters comments and interacts with the system.

[0097] "Comment" means a text message entered by a User on the Online Platform.

[0098] "Real-time" refers to the fact that user actions and the system's response to them are immediate.

[0099] "Offensiveness" refers to the degree to which the content of a comment is offensive or unpleasant to others.

[0100] "Score" refers to a score used to quantitatively evaluate the offensiveness of a comment.

[0101] "Rewrites" are text messages that convert offensive comments into more gentle and wholesome language.

[0102] "Suggest" means presenting the user with a rewritten statement and giving them the option to correct the original comment.

[0103] "Gamification" refers to the incorporation of game elements to increase user motivation.

[0104] "Average score" refers to the average aggression score a user has earned in the past.

[0105] "Ranking" refers to ranking based on the average score achieved by users.

[0106] "Terminal" refers to the device (smartphone, PC, etc.) through which a user enters comments and interacts with the system.

[0107] "Server" refers to the central computer system used to analyze comments, assign offensiveness scores, generate rewrites, and manage rankings.

[0108] "Analysis model" refers to natural language processing technology used to analyze the content of comments and generate offensiveness scores and rewrite suggestions.

[0109] "Ranking System" refers to a system that displays rankings based on users' comment scores and provides gamification elements.

[0110] A "natural language processing model" refers to an analysis algorithm that uses technology to understand and analyze human language.

[0111] This invention provides a system that analyzes in real time how a user's comment content will be perceived by others when the user enters it, and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average score earned by the user. Specifically, it consists of the following main components:

[0112] User Device

[0113] A user terminal is a device through which users can enter comments on news articles and interact with the system. For example, a smartphone or PC is an example. Users can enter comments through the terminal and check the analysis results in real time.

[0114] server

[0115] The server is a central computer system that receives and analyzes comments submitted by users. The server is equipped with OpenAI's GPT model as a natural language processing (NLP) model, which analyzes comments, assigns offensiveness scores, and generates rewritten sentences.

[0116] Ranking System

[0117] The ranking system displays a ranking based on the average offensiveness score of users and provides a gamification element. The system stores users' comment history and offensiveness scores in a database and updates the user's rank based on the results.

[0118] Explanation of program processing

[0119] When the server receives a user's comment, it analyzes it using an NLP model (e.g., OpenAI's GPT-3). This analysis calculates an offensiveness score for the comment. For offensive comments, it generates a rewritten text with milder language. For example, if a comment such as "This news is terrible. Aren't all the people involved stupid?" is entered, it is determined to have a high offensive score, and a rewrite suggestion such as "There are many problems with this news. I hope that the people involved will respond better." is generated.

[0120] The generated offensiveness score and rewrite proposal are sent back from the server to the user's device. The user can review them on their device and choose whether or not to adopt the proposed rewrite. If the user adopts the rewrite, it is saved on the device as a final comment and sent back to the server. The server saves this final comment in a database and updates the user's average offensiveness score. An updated ranking is generated based on this score and sent back to the user's device.

[0121] Specific examples

[0122] Suppose a user enters a comment such as "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by the AI ​​model. As a result of the analysis, the offensiveness score is determined to be 80 / 100, and a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better." If the user adopts the rewrite suggestion and sends the final comment to the server, the server saves this comment in the database and updates the ranking.

[0123] Prompt Sentence Examples

[0124] "Analyze how an input comment would be perceived by others, score the level of offensiveness, and suggest a milder rewrite that expresses the same meaning. Example: Comment: 'This news is terrible. Are all the people involved stupid?'"

[0125] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0126] Step 1: User comments

[0127] A user uses a device to enter text into a comment section of an online platform, for example, "This news is terrible. Are all the people involved stupid?" This comment is temporarily stored in the device's memory.

[0128] Input: Comments entered by the user

[0129] Output: Saved comments

[0130] Specific operation: The user enters a comment using a keyboard or touch screen and presses the send button.

[0131] Step 2: Submitting a comment analysis request

[0132] The device creates an HTTP request to send an analysis request including the entered comment and user ID to the server. The request includes the user ID and comment content in JSON format. For example, it is sent in the following format: {"user_id": "12345", "comment": "This news is terrible. Aren't all the people involved stupid?"}

[0133] Input: Temporarily saved comment, user ID

[0134] Output: Parse request to server

[0135] Specific operation: The device creates an HTTP request and sends it to the server over the Internet.

[0136] Step 3: Comment analysis by the server

[0137] The server receives the request and analyzes the comment. Using a natural language processing (NLP) model, such as OpenAI's GPT-3 model, the server calculates the comment's offensiveness score and generates a rewrite. If the offensiveness score is 80 / 100, the rewrite reads, "There are many problems with this news. We hope that the relevant parties will improve their response."

[0138] Input: Analysis request (user ID, comment content)

[0139] Output: Analysis results (aggression score, rewritten sentence)

[0140] Specific operation: The server executes the NLP model to analyze the comments and generate rewritten sentences.

[0141] Step 4: Returning the analysis results

[0142] The server creates a response containing the generated aggressiveness score and the rewritten comment, and sends it back to the user's device. The response is also in JSON format, for example, {"aggressiveness_score": 80, "rewritten_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[0143] Input: Analysis results (aggression score, rewritten text)

[0144] Output: Response to user device

[0145] Specific operation: The server creates an HTTP response and sends it to the terminal via the Internet.

[0146] Step 5: Presenting and selecting rewrite proposals

[0147] The user's device receives the response from the server and displays the offensiveness score and the rewritten text to the user. The user can choose to accept the rewritten comment or post the original comment as is. For example, if the user accepts the rewritten comment, the rewritten comment is saved on the device as the final comment. If the user rejects the rewritten comment, the original comment is saved as the final comment.

[0148] Input: Response from the server (aggression score, rewritten text)

[0149] Output: Final comment

[0150] Specific operation: The user's device displays rewrite suggestions, and the user presses the selection button.

[0151] Step 6: Submitting final comments

[0152] The device then sends the final comment and user ID to the server again in JSON format, for example, {"user_id": "12345", "final_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[0153] Input: Last comment, User ID

[0154] Output: Sending the final comment to the server

[0155] Specific operation: The terminal creates an HTTP request including the final comment and sends it to the server.

[0156] Step 7: Update the ranking system

[0157] The server saves the final comment in the database and updates the user's average offensiveness score. This recalculates the user's rank and generates an updated ranking. The server then sends a response back to the user device indicating that the update is complete. For example, the response will be in the format {"status": "success", "new_rank": 5}.

[0158] Input: Final comment

[0159] Output: Updated rank

[0160] Specific operation: The server updates the database, calculates the user's new rank, generates a response and sends it to the device.

[0161] (Application example 1)

[0162] 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."

[0163] In the comment sections of content distribution services, many users often post offensive comments, which can undermine the health of the dialogue. There is a need for an appropriate system to promote communication between users and provide a healthy dialogue environment. However, current technology lacks a mechanism for analyzing comments in real time, scoring their offensiveness, and suggesting rewrites. Furthermore, there is a lack of a system that can promote healthy comment posting by scoring and ranking users' commenting behavior.

[0164] 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.

[0165] In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for offensive comments, and means for updating user scores and rankings. This makes it possible to analyze the aggression of comments in real time and provide a healthy dialogue environment. Furthermore, by suggesting appropriate rewrites to users, the number of offensive comments can be reduced. Furthermore, gamification elements can be used to encourage users to post healthy comments, and a ranking system can increase user participation.

[0166] The "real-time analysis means" is a function that analyzes user comments as they are entered and instantly scores the degree of aggressiveness based on their content.

[0167] The "rewrite suggestion method" is a function that, when an offensive comment is entered, presents the user with rewrite suggestions to revise the content to make it more appropriate and constructive.

[0168] The "gamification element providing means" is a function that introduces game-like elements into user behavior, scores comment posts, and ranks them based on the average score obtained.

[0169] "Smart device display means" is a function that displays the comment's offensiveness score and rewritten text on the screen of the device used by the user, such as a smartphone or tablet.

[0170] "Scoring and ranking update means" is a function that calculates an aggressiveness score based on comments posted by users, records the score in a database, and updates the user's rating and ranking.

[0171] The "server" is a computer system that receives comments sent by users, analyzes them, generates an offensiveness score and a rewritten sentence, and stores the final comments in a database.

[0172] System Overview

[0173] The system analyzes the content of comments entered by users in real time, scores their offensiveness, and suggests rewrites. It also includes a gamification element that scores and ranks users' behavior.

[0174] Hardware Configuration

[0175] The system consists of the following major hardware components:

[0176] 1. User Device: An end-user device, including a smartphone or tablet, that allows for the input of comments and the display of suggested rewrites.

[0177] 2. Server: A central computer system that processes analysis and manages data.

[0178] Software Configuration

[0179] The system consists of the following major software components:

[0180] 1. Comment analysis module: Generative AI model using Python and TensorFlow.

[0181] 2. Rewrite generation module: Processes text data and generates appropriate rewrites for offensive comments.

[0182] 3. Database: Stores user comments, scores, and ranks (e.g. MySQL).

[0183] 4. User interface: An application that runs on a smart device and allows users to enter comments, view analysis results, and propose and adopt rewrite proposals.

[0184] Program processing flow

[0185] User side:

[0186] 1. Comment input: The user enters text into the comment input field of the smartphone app.

[0187] 2. Send: The entered comment is sent to the analysis server along with the user ID as an HTTP POST request.

[0188] Server side:

[0189] 3. Real-time analysis: The server receives the request and analyzes the comment using a generative AI model using Python and TensorFlow. The analysis results in an offensiveness score and generates a rewritten sentence.

[0190] Example: If the comment "This news is terrible. Aren't all the people involved stupid?" is entered, the offensiveness score will be evaluated as 80 / 100, and a rewrite sentence will be generated that reads "There are many problems with this news. I hope that the people involved will respond better."

[0191] 4. Returning the results: The analysis results (offensiveness score and rewritten sentence) are returned from the server to the user's device.

[0192] User side:

[0193] 5. Display and selection: The smart device receives the analysis results sent from the server and presents them to the user. The user can then choose whether to adopt the rewritten sentence.

[0194] 6. Sending the final comment: The final comment that adopts the rewritten sentence is sent to the server again and saved in the database.

[0195] Server side:

[0196] 7. Scoring and Ranking Update: The server updates the average score of the user based on the last comment and recalculates the ranking. The latest ranking information is sent back to the user's device.

[0197] Examples of specific examples and prompts

[0198] As a concrete example, consider the case where a user writes a review of a TV drama. If the user writes, "This drama is not interesting at all. All the actors' acting is terrible," this comment is sent to the analysis server. The generative AI model analyzes the comment on the server side, assigns an offensiveness score of 70 / 100, and suggests a rewrite such as, "This drama did not meet my expectations. I look forward to seeing the actors grow."

[0199] Example prompt sentence:

[0200] If a comment is typed: "This news is terrible. Are all the people involved stupid?"

[0201] Input the text "This news is terrible. Are all the people involved stupid?" into the analytical AI model and generate an offensiveness score and an improved rewrite.

[0202] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0203] Program processing flow

[0204] Step 1:

[0205] Users enter text into the comment input field of the smartphone app, which is temporarily stored on the user's device.

[0206] input:

[0207] The comment text entered by the user.

[0208] output:

[0209] The comment text is temporarily saved on the user's device.

[0210] Step 2:

[0211] The user device sends an HTTP POST request to the server to analyze the comment, which includes the user ID and the comment text.

[0212] input:

[0213] User ID, comment text.

[0214] output:

[0215] An HTTP POST request containing the user ID and comment text is sent to the server.

[0216] Step 3:

[0217] The server receives the request and analyzes the comments using a generative AI model powered by Python and TensorFlow.

[0218] input:

[0219] Request data (user ID, comment text).

[0220] output:

[0221] Aggression scores and rewritten statements.

[0222] Data processing and calculation:

[0223] The server launches a generative AI model that generates an offensiveness score and rewrite sentences using the input comment text as prompts.

[0224] Step 4:

[0225] The server sends the generated offensive score and the rewritten sentence back to the user's device.

[0226] input:

[0227] Aggression score, rewritten sentence.

[0228] output:

[0229] A response containing the offensiveness score and the rewritten sentence is sent to the user's device.

[0230] Step 5:

[0231] The user device receives the offensiveness score and rewritten sentences sent from the server and presents them to the user, who can then review the rewritten sentences and choose whether or not to adopt them.

[0232] input:

[0233] Response from the server (aggression score, rewritten text).

[0234] output:

[0235] The offensiveness score and rewritten sentence presented to the user.

[0236] Specific behavior:

[0237] The user's device displays the received offensiveness score and rewritten text on the screen, and the user confirms them.

[0238] Step 6:

[0239] If the user accepts the rewritten comment, it is sent to the server again as a final comment. This request includes the user ID and the final comment.

[0240] input:

[0241] User ID, last comment (rewritten text).

[0242] output:

[0243] An HTTP POST request containing the user ID and the last comment is sent to the server.

[0244] Step 7:

[0245] The server receives the final comments and stores them in the database, while calculating the average score of the users and updating the ranking.

[0246] input:

[0247] Final comment.

[0248] output:

[0249] Updated user ranking information.

[0250] Data processing and calculation:

[0251] The server stores the final comment received in a database, compares it with the user's past comment scores, calculates an average score, and updates the user's ranking based on the result to generate the latest ranking information.

[0252] 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.

[0253] The present invention provides a system that analyzes in real time how a user's comments are perceived by others and scores the degree of offensiveness when the user enters a comment. The system also combines a gamification element that suggests improved rewrites for offensive comments and ranks them based on the user's average score, with an emotion engine that recognizes the user's emotions.

[0254] System Overview

[0255] The system consists of the following main components:

[0256] User terminal: A device used to input comments and interact with the system.

[0257] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[0258] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0259] Program processing flow

[0260] User comment input

[0261] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[0262] Sending a comment analysis request

[0263] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0264] AI-powered comment analysis

[0265] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite, and uses an emotion engine to recognize the user's emotional state. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better." The server also recognizes the user's emotion as anger.

[0266] Returning analysis results

[0267] The server creates a response that includes an offensiveness score, suggested rewrites, and the emotion recognized by the emotion engine.

[0268] Example: Aggression score 80, rewrite suggestion: "There are many problems with this news. I hope that the people involved will respond better.", user emotional state: anger.

[0269] Presenting and selecting rewrite proposals

[0270] The user terminal receives the response sent back from the server and displays the analysis results to the user.

[0271] The user device presents the offensiveness score and the rewrite suggestion, and prompts the user to choose whether to adopt the rewrite suggestion. The user decides whether to adopt the rewrite suggestion. For example, if the user adopts the rewrite suggestion, the content is saved as the final comment on the user device.

[0272] Submitting final comments

[0273] The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. The final comment is sent to the server.

[0274] Ranking System Update

[0275] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[0276] Specific examples

[0277] For example, imagine a user comments, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI and an emotion engine. It is determined to have a high offensive score, and a rewrite suggestion is generated. The user's emotion is also recognized as "anger," and a rewrite suggestion is presented based on this result. If the user adopts this rewrite suggestion, it is sent to the server as the final comment and saved in the database. This updates the user's average score, which is also reflected in the rankings. In this way, users are encouraged to voluntarily post less aggressive comments. This process discourages aggressive comments in the comment section, providing a healthy dialogue environment.

[0278] The processing flow will be explained below.

[0279] Specific processing steps of the program

[0280] Step 1:

[0281] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[0282] Terminal: Obtains and temporarily stores comments entered by users.

[0283] Step 2:

[0284] Device: Create a request to analyze the saved comments. This request contains the user ID, the comment, and the user's device session information.

[0285] Terminal: Sends the created request to the server.

[0286] Step 3:

[0287] Server: Receives requests from user devices.

[0288] Server: The received comments are passed to an AI analysis model for analysis, where an offensiveness score is calculated and a rewrite proposal is generated.

[0289] Step 4:

[0290] Server: Based on the analysis of the comments, an emotion engine is used to identify the user's emotional state. For example, the user's comment is recognized as being based on anger.

[0291] Server: Adjusts rewrite suggestions based on the user's emotional state. In the future, especially in urgent cases, the server will determine the appropriate timing to present rewrite suggestions to the user according to the emotion recognition results.

[0292] Step 5:

[0293] Server: Create a response that includes the offensiveness score, emotional state, and rewrite suggestion. For example, the offensiveness score is 80 / 100, and the rewrite suggestion is "There are many problems with this news. I hope that the people involved will respond better."

[0294] Server: Returns the created response to the user device.

[0295] Step 6:

[0296] Terminal: Receives the response sent back from the server and displays the analysis results (aggression score, emotional state, rewrite suggestions) to the user.

[0297] Device: Ask the user whether they would like to adopt the proposed rewrite: "There are many problems with this news. We hope that those involved will respond better."

[0298] User: Review the rewrite suggestion and decide whether to adopt it. For example, if the user adopts the rewrite suggestion.

[0299] Step 7:

[0300] Device: Create a request to send the user's final comment (in this case, a rewrite proposal) back to the server. This request includes the user ID, the final comment, and the emotional state.

[0301] Terminal: Send a request to the server containing the final comment.

[0302] Step 8:

[0303] Server: Receives the final comment from the user terminal.

[0304] Server: Stores received comments in a database, including the comment, its emotional state, and its aggression score.

[0305] Step 9:

[0306] Server: Update the average score of the user based on the offensiveness score of the last comment. Update the rank based on the average score of the user.

[0307] Server: Returns a response to the user device indicating that the rank update has been completed.

[0308] Specific examples

[0309] For example, if a user comments, "This news is terrible. Aren't all the people involved stupid?", the system sends the comment to the server. The server uses an AI model to calculate an aggression score and simultaneously recognizes the user's emotional state using an emotion engine. As a result, the aggression score is determined to be 80 / 100, and the user's emotion is recognized as "anger." The server then generates a rewrite suggestion, "There are many problems with this news. I hope the people involved will improve their response," and presents it to the user. If the user adopts the rewrite suggestion, the comment is sent to the server and stored in the database. The user's average aggression score is also updated, along with their rank. This process encourages users to voluntarily refrain from making aggressive comments, creating a healthy dialogue environment.

[0310] Example 2

[0311] 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."

[0312] There are many offensive comments in online comment sections and message boards, which is a problem as it damages a healthy dialogue environment. There are also cases where users post offensive comments without realizing it, which calls for improvement. Furthermore, there is a lack of ways for users to learn what words are offensive and to communicate better.

[0313] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a comment, a means for transmitting the comment to the server, a means for the server to analyze the comment in real time and score the degree of aggression, a means for generating a rewritten sentence based on the aggression score, a means for presenting the rewritten sentence to the user and prompting the user to select, a means for providing a gamification element that ranks the rewritten sentences based on the average score obtained by the users, and a means for storing the comments and the analysis results in a database. This encourages users to voluntarily post less aggressive comments, and provides a healthy dialogue environment and enables users to learn.

[0314] "User" refers to the entity that accesses the system and enters comments.

[0315] A "comment" is text information entered by a user that accompanies content such as a news article or message board.

[0316] "Server" refers to the central processing unit that receives and analyzes comments.

[0317] "Real-time" refers to a situation where there is almost no delay between the time a comment is entered and the time the analysis results are returned.

[0318] "Level of aggression" refers to a numerical score that indicates how aggressive the content of a comment is.

[0319] A "rewrite" refers to more appropriately worded text that is generated to make the original comment less offensive.

[0320] "Analysis results" refers to information such as scores, rewritten sentences, and emotional state obtained after the server analyzes the comments.

[0321] "Gamification elements" are elements that make user behavior fun and motivating, like a game, and specifically include ranking and reward systems.

[0322] "Database" refers to a collection of information for storing comments and analysis results.

[0323] An "emotion engine" refers to a mechanism that analyzes and detects emotional states (such as anger or joy) from user comments.

[0324] The embodiment of the present invention is a system that, when a user inputs a comment, analyzes in real time how the content of the comment will be received by others and scores the degree of offensiveness. A detailed description of this system is provided below.

[0325] System Overview

[0326] The system consists of the following main components:

[0327] User terminal: A device used to input comments and interact with the system. User terminals include personal computers and smartphones.

[0328] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[0329] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0330] Hardware and software used

[0331] User device: A device that can connect to the Internet (e.g., smartphone, PC)

[0332] Server: A high-performance computer using cloud services

[0333] Software: generative AI models (e.g., open-source natural language processing models), database management systems, emotion engines

[0334] Data processing and calculation

[0335] 1. Comment input: The user inputs a comment, and the user terminal sends this data to the server.

[0336] 2. Data analysis: The server receives the comments and uses a generative AI model to score the degree of offensiveness, and an emotion engine to analyze the user's emotional state.

[0337] 3. Generating rewritten sentences: The generative AI model generates rewritten sentences for comments that are judged to be highly offensive.

[0338] 4. Presentation of results: The analysis results and rewritten sentences are returned from the server and displayed on the user's terminal.

[0339] 5. Accepting the comment: The user selects whether to accept the rewritten sentence and sends the final comment to the server.

[0340] 6. Update ranking: The server saves the final comments to the database and updates the ranking based on the average score of the users.

[0341] Specific examples

[0342] For example, if a user types a comment like "This news is terrible. Aren't all the people involved stupid?", this comment is sent to the server, which then inputs the following prompt sentence into the generative AI model for analysis:

[0343] Example prompt sentence:

[0344] Comment Analysis:

[0345] Typed comment: "This news is terrible. Are all the people involved stupid?"

[0346] Aggression Score: 80

[0347] Suggested rewrite: "There are many problems with this news. I hope that those involved will improve their response."

[0348] Emotional state: Anger

[0349] After the analysis is complete, the server sends back the results, including the offensiveness score, rewrite suggestions, and emotional state, to the user's device. The user's device displays these results on the screen and prompts the user to decide whether to accept the rewrite suggestions. If the user accepts the rewrite suggestions, the rewritten comment is resubmitted to the server as the final comment. Finally, the final comment is saved in the database, and the ranking is updated based on the user's average score.

[0350] Through this process, users are encouraged to voluntarily post less offensive comments, creating a healthier interactive environment.

[0351] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0352] Step 1: Enter a comment

[0353] A user types a comment into the comment section of a news article or message board.

[0354] What happens: A user types "This news is terrible. Are all the people involved stupid?" This comment is stored as temporary data on the user's device.

[0355] Input: User comment

[0356] Output: Input comments as temporary data

[0357] Step 2: Send a comment analysis request

[0358] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0359] Specific operation: The user's device packages the user ID and comments in JSON format and sends an HTTP request to the server's API endpoint.

[0360] Input: User ID, Comment

[0361] Output: HTTP request to the server

[0362] Step 3: Receiving and analyzing comments

[0363] The server receives the request and analyzes the comment using a generative AI model, which generates an offensiveness score and a rewrite, and also uses an emotion engine to recognize the user's emotional state.

[0364] Specific operation: The server analyzes the request and creates a prompt sentence to input the comment part into the AI ​​model. The AI ​​model analyzes it and generates an offensiveness score, rewrite sentence, and emotional state.

[0365] Input: User ID, Comment

[0366] Output: Analysis results including aggression score, rewritten sentences, and emotional state

[0367] Step 4: Returning the analysis results

[0368] The server creates a response containing the offensiveness score, suggested rewrites, and the perceived emotion, and sends it back to the user device.

[0369] Specific operation: The server packages the analysis results in JSON format and sends them to the user device as an HTTP response.

[0370] Input: Aggression score, rewritten sentence, emotional state

[0371] Output: HTTP response to the user's device

[0372] Step 5: Presenting and selecting rewrite proposals

[0373] The user device receives the response and displays the analysis results to the user, who can then decide whether to adopt the proposed rewrite.

[0374] Specific operation: The user's device parses the response, displays it on the screen, and presents rewrite suggestions. The user can choose whether to adopt the rewrite suggestions.

[0375] Input: Analysis results (aggression score, rewritten sentence, emotional state)

[0376] Output: User's choice (whether to accept the rewrite proposal or not)

[0377] Step 6: Submitting final comments

[0378] The user device creates a request to resend the last comment to the server, including the user ID and the last comment.

[0379] Specific operation: The user's device packages the user ID and the last comment in JSON format and sends an HTTP request to the server's API endpoint.

[0380] Input: User ID, last comment

[0381] Output: HTTP request to the server

[0382] Step 7: Update the ranking system

[0383] The server saves the final comments to the database and updates the user's average score, which in turn updates the user's rank to generate the latest ranking.

[0384] Specific operation: The server saves the final comment in the database, calculates a new average score based on the user's previous comment scores, updates the ranking database, and sends an update completion message to the user's device.

[0385] Input: Last comment, user's past score

[0386] Output: Updated ranking, update complete message

[0387] (Application example 2)

[0388] 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."

[0389] Current content distribution services often have users posting offensive comments, making it difficult to maintain a healthy dialogue environment. Furthermore, because users often post offensive comments without being aware of their own feelings, a mechanism to curb this is needed. Furthermore, there is a lack of gamification elements to encourage users to voluntarily adopt positive behavior.

[0390] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for aggressive comments, means for providing a gamification element that ranks comments based on the average points earned by users, means for recognizing users' emotional states in real time and providing that information, and means for applying it to the comment section of the content distribution service. This enables users to voluntarily post less aggressive comments and maintain a healthy dialogue environment.

[0391] "User" refers to a user who posts comments within the content distribution service.

[0392] "Comment" refers to a text message that a user enters within the Content Distribution Service and shares with others.

[0393] "Real-time analysis" refers to the process of instantly analyzing the content of a comment the moment it is entered and providing the results.

[0394] "Offensiveness" refers to a rating that indicates how offensive or harmful a comment may be to others.

[0395] A "rewritten sentence" refers to a sentence in which an offensive comment has been revised to use more neutral or constructive language.

[0396] "Gamification elements" refer to game-like elements that assign scores and ranks to user behavior to increase user motivation.

[0397] "Emotional state" refers to the state of a user's emotions when they are entering a comment, and includes anger, joy, sadness, etc.

[0398] "Content distribution service" refers to an online service that provides users with various content such as video, audio, and text, and allows them to share and comment on it.

[0399] "Server" refers to the computer system that analyzes comments, calculates offensiveness scores, generates rewritten text, recognizes the user's emotional state, and stores and manages data.

[0400] As an embodiment of the present invention, a specific description of the system is given below. This system is composed of the main components of a user terminal, a server, and a ranking system.

[0401] User Device

[0402] User devices are expected to include smartphones, tablets, personal computers, etc. Users enter comments into the comment section of the content distribution service through a comment form. While users are entering their comments, a system is in place to temporarily save them in real time.

[0403] server

[0404] The server analyzes comments, calculates offensiveness scores, generates rewritten sentences, and recognizes the user's emotional state. Specifically, it uses the following software and hardware:

[0405] software:

[0406] Analysis model: We use the Hugging Face text classification model to evaluate the offensiveness of comments.

[0407] Emotion Engine: Recognizes the user's emotional state using the Hugging Face emotion recognition model.

[0408] Hardware:

[0409] High-performance servers: Cloud-based high-performance servers (e.g., Amazon Web Services, Google Cloud Platform) for rapid data processing.

[0410] The server then sends the analyzed comments back to the user's device along with an offensiveness score and a rewritten version of the comment, along with the user's emotional state.

[0411] Ranking System

[0412] The ranking system provides a gamification element by ranking users based on their average score, which is based on a database managed by the server. By posting less offensive comments, users can improve their scores and move up the rankings.

[0413] Specific examples

[0414] For example, if a user types a comment like "This news is terrible. Are all the people involved stupid?", the following process will occur:

[0415] 1. Real-time analysis:

[0416] The comment is immediately sent to a server, where an analysis model calculates an offensiveness score.

[0417] The emotion engine recognizes the user's emotional state as "anger."

[0418] 2. Generating and presenting rewritten sentences:

[0419] Along with the offensiveness score, a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better."

[0420] It is sent back to the user's device in real time.

[0421] 3. User Choice:

[0422] Users can choose whether to adopt the rewrite proposal.

[0423] If adopted, the rewritten comment is resubmitted to the server as the final comment.

[0424] 4. Ranking System Update:

[0425] The server saves the final comments in the database and updates the user's average score.

[0426] The rankings are updated and feedback is provided to users.

[0427] Prompt Sentence Examples

[0428] "Please analyze comments entered by users in real time and calculate an offensiveness score. For example, let's analyze the following comment.

[0429] Comment: "This news is terrible. Aren't all the people involved stupid?"

[0430] The above embodiment makes it possible to realize a mechanism that suppresses offensive comments in the comment section and provides a healthy dialogue environment.

[0431] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0432] Step 1:

[0433] The user enters a comment. The user uses a smartphone or personal computer to enter text into the comment field of the content distribution service. This comment is temporarily saved on the user's device. Input: Text data of the comment. Output: Text data temporarily saved on the device.

[0434] Step 2:

[0435] Prepare to send the comment to the server. The user's device generates a request to send the entered comment to the server in real time. This request includes the user ID and comment. Input: User ID and comment text data. Output: Request data to be sent to the server.

[0436] Step 3:

[0437] The server receives the comment. The server receives a request from the user's device and obtains the comment text data and user ID. Input: Request data. Output: Comment text data and user ID.

[0438] Step 4:

[0439] Analyze the comments. The server uses an analysis model (Hugging Face's text classification model) to calculate the comment's aggressiveness score. It also uses an emotion engine (Hugging Face's emotion recognition model) to recognize the user's emotional state. Input: Comment text data. Output: Aggression score and emotional state.

[0440] Step 5:

[0441] Generate a rewritten sentence. The server generates a rewritten sentence that changes the comment to a more neutral or constructive expression based on the offensiveness score. Input: Comment text data and offensiveness score. Output: Rewritten sentence.

[0442] Step 6:

[0443] The analysis results are returned to the user device. The server creates a response including the offensiveness score, rewrite suggestion, and emotional state, and returns it to the user device. Input: offensiveness score, rewrite sentence, emotional state. Output: response data.

[0444] Step 7:

[0445] The analysis results are presented. The user device receives the response from the server and displays the offensiveness score, rewrite suggestions, and emotional state to the user. Input: Response data. Output: Display data on the device.

[0446] Step 8:

[0447] The user makes a choice. The user chooses whether to adopt the proposed rewrite. If the user adopts the rewrite, the content is saved as the final comment. Input: User's choice. Output: Text data of the final comment.

[0448] Step 9:

[0449] Send the final comment to the server. The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. Input: User ID and text data of the final comment. Output: Request data to be sent to the server.

[0450] Step 10:

[0451] Update the ranking system. The server saves the final comment to the database and updates the user's average score. Based on that score, the user's rank is updated and the latest ranking is generated. Input: Text data of the final comment. Output: Updated ranking data.

[0452] Through the above steps, a system is realized that analyzes users' comments and encourages them to post less offensive comments.

[0453] 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.

[0454] 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.

[0455] 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.

[0456] [Second embodiment]

[0457] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0458] 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.

[0459] 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).

[0460] 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.

[0461] 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.

[0462] 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).

[0463] 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.

[0464] 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.

[0465] 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.

[0466] 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.

[0467] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0468] 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."

[0469] The embodiment of the present invention provides a system that analyzes in real time how the content of a comment entered by a user will be perceived by others and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average points earned by users.

[0470] System Overview

[0471] The system consists of the following main components:

[0472] User terminal: A device used to input comments and interact with the system.

[0473] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, and manages rankings.

[0474] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0475] Program processing flow

[0476] User comment input

[0477] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[0478] Sending a comment analysis request

[0479] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0480] AI-powered comment analysis

[0481] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better."

[0482] Returning analysis results

[0483] The server creates a response containing the offensiveness score and the rewritten text and sends it back to the user's device.

[0484] Presenting and selecting rewrite proposals

[0485] The user device receives the response from the server and presents the offensiveness score and rewritten sentence to the user. The user can choose whether to adopt the proposed rewritten sentence. For example, if the user adopts the rewritten sentence, the content is saved as the final comment on the user device.

[0486] Submitting final comments

[0487] The user device sends the final comment back to the server, and this request includes the user ID and the final comment.

[0488] Ranking System Update

[0489] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[0490] Specific examples

[0491] For example, consider a case where a user enters a comment such as, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI. It is determined to have a high offensive score, and a rewrite proposal is generated. The rewrite proposal, "There are many problems with this news. I hope that the people involved will improve their response," is presented to the user. The user adopts the rewrite proposal and sends the final comment to the server. The server saves this comment in a database and updates the ranking.

[0492] This process will help curb offensive posts in the comments section and provide a healthy interactive environment.

[0493] The processing flow will be explained below.

[0494] Specific processing steps of the program

[0495] Step 1:

[0496] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[0497] Terminal: Obtains and temporarily stores comments entered by users.

[0498] Step 2:

[0499] On your device: Create a request to analyze the saved comments. This request includes the user ID and the comment.

[0500] Terminal: Sends the created request to the server.

[0501] Step 3:

[0502] Server: Receives requests from user devices.

[0503] Server: The received comments are passed to the AI ​​analysis model for analysis, where the offensiveness score of the comment is calculated and a rewrite proposal is generated.

[0504] Step 4:

[0505] Server: Create a response containing an offensiveness score and a rewrite suggestion as the analysis result. For example, if the offensiveness score is 80 / 100, the rewrite suggestion is "There are many problems with this news. We hope that the relevant parties will respond better."

[0506] Server: Returns the created response to the user device.

[0507] Step 5:

[0508] Terminal: Receives the response sent back from the server and displays the analysis results to the user.

[0509] Device: Presents the offensiveness score and rewrite suggestions, prompting the user to choose whether or not to adopt the rewrite suggestions.

[0510] User: Checks the rewrite proposal and decides whether to adopt it. For example, adopt the rewrite proposal "There are many problems with this news. I hope that the people involved will respond better."

[0511] Step 6:

[0512] On the device: The user creates a request to send the final comment (in this case, the rewrite proposal) back to the server. This request includes the user ID and the final comment.

[0513] Terminal: Send a request to the server containing the final comment.

[0514] Step 7:

[0515] Server: Receives the final comment from the user terminal.

[0516] Server: Stores received comments in a database.

[0517] Server: Updates the average score of users based on the offensiveness score of their comments and processes the ranking.

[0518] Server: Returns a response to the user device indicating that the ranking update has been completed.

[0519] For example, if a user enters an offensive comment such as "This news is terrible. Aren't all the people involved stupid?", the system will analyze the comment using AI and assign a high offensiveness score (80 / 100). At the same time, the system will generate a rewrite suggestion, "There are many problems with this news. I hope that the people involved will improve their response," and present it to the user. If the user adopts this rewrite suggestion, it will be sent to the server as the final comment and saved in the database. This will also update the user's average score, which will be reflected in the rankings. In this way, users are encouraged to voluntarily post less offensive comments.

[0520] Example 1

[0521] 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."

[0522] In recent years, the number of offensive comments has increased in the comment sections of online platforms, hindering a healthy dialogue environment. There is a need for a system that can detect such offensive comments in real time and allow users to correct them themselves. Gamification elements are also necessary to increase user motivation.

[0523] 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.

[0524] In this invention, the server includes: means for analyzing user comments in real time and scoring the degree of offensiveness; means for proposing improved rewrites for offensive comments; means for providing a gamification element that ranks comments based on the average points earned by users; a terminal for inputting and submitting comments and presenting rewrites; a server equipped with an analytical model for generating offensiveness scores and rewrite suggestions; and a system for displaying rankings based on user comment scores. This makes it possible to provide a healthy interactive environment and increase user motivation by detecting offensive comments in real time and presenting improvement suggestions.

[0525] "User" refers to the end user who enters comments and interacts with the system.

[0526] "Comment" means a text message entered by a User on the Online Platform.

[0527] "Real-time" refers to the fact that user actions and the system's response to them are immediate.

[0528] "Offensiveness" refers to the degree to which the content of a comment is offensive or unpleasant to others.

[0529] "Score" refers to a score used to quantitatively evaluate the offensiveness of a comment.

[0530] "Rewrites" are text messages that convert offensive comments into more gentle and wholesome language.

[0531] "Suggest" means presenting the user with a rewritten statement and giving them the option to correct the original comment.

[0532] "Gamification" refers to the incorporation of game elements to increase user motivation.

[0533] "Average score" refers to the average aggression score a user has earned in the past.

[0534] "Ranking" refers to ranking based on the average score achieved by users.

[0535] "Terminal" refers to the device (smartphone, PC, etc.) through which a user enters comments and interacts with the system.

[0536] "Server" refers to the central computer system used to analyze comments, assign offensiveness scores, generate rewrites, and manage rankings.

[0537] "Analysis model" refers to natural language processing technology used to analyze the content of comments and generate offensiveness scores and rewrite suggestions.

[0538] "Ranking System" refers to a system that displays rankings based on users' comment scores and provides gamification elements.

[0539] A "natural language processing model" refers to an analysis algorithm that uses technology to understand and analyze human language.

[0540] This invention provides a system that analyzes in real time how a user's comment content will be perceived by others when the user enters it, and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average score earned by the user. Specifically, it consists of the following main components:

[0541] User Device

[0542] A user terminal is a device through which users can enter comments on news articles and interact with the system. For example, a smartphone or PC is an example. Users can enter comments through the terminal and check the analysis results in real time.

[0543] server

[0544] The server is a central computer system that receives and analyzes comments submitted by users. The server is equipped with OpenAI's GPT model as a natural language processing (NLP) model, which analyzes comments, assigns offensiveness scores, and generates rewritten sentences.

[0545] Ranking System

[0546] The ranking system displays a ranking based on the average offensiveness score of users and provides a gamification element. The system stores users' comment history and offensiveness scores in a database and updates the user's rank based on the results.

[0547] Explanation of program processing

[0548] When the server receives a user's comment, it analyzes it using an NLP model (e.g., OpenAI's GPT-3). This analysis calculates an offensiveness score for the comment. For offensive comments, it generates a rewritten text with milder language. For example, if a comment such as "This news is terrible. Aren't all the people involved stupid?" is entered, it is determined to have a high offensive score, and a rewrite suggestion such as "There are many problems with this news. I hope that the people involved will respond better." is generated.

[0549] The generated offensiveness score and rewrite proposal are sent back from the server to the user's device. The user can review them on their device and choose whether or not to adopt the proposed rewrite. If the user adopts the rewrite, it is saved on the device as a final comment and sent back to the server. The server saves this final comment in a database and updates the user's average offensiveness score. An updated ranking is generated based on this score and sent back to the user's device.

[0550] Specific examples

[0551] Suppose a user enters a comment such as "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by the AI ​​model. As a result of the analysis, the offensiveness score is determined to be 80 / 100, and a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better." If the user adopts the rewrite suggestion and sends the final comment to the server, the server saves this comment in the database and updates the ranking.

[0552] Prompt Sentence Examples

[0553] "Analyze how an input comment would be perceived by others, score the level of offensiveness, and suggest a milder rewrite that expresses the same meaning. Example: Comment: 'This news is terrible. Are all the people involved stupid?'"

[0554] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0555] Step 1: User comments

[0556] A user uses a device to enter text into a comment section of an online platform, for example, "This news is terrible. Are all the people involved stupid?" This comment is temporarily stored in the device's memory.

[0557] Input: Comments entered by the user

[0558] Output: Saved comments

[0559] Specific operation: The user enters a comment using a keyboard or touch screen and presses the send button.

[0560] Step 2: Submitting a comment analysis request

[0561] The device creates an HTTP request to send an analysis request including the entered comment and user ID to the server. The request includes the user ID and comment content in JSON format. For example, it is sent in the following format: {"user_id": "12345", "comment": "This news is terrible. Aren't all the people involved stupid?"}

[0562] Input: Temporarily saved comment, user ID

[0563] Output: Parse request to server

[0564] Specific operation: The device creates an HTTP request and sends it to the server over the Internet.

[0565] Step 3: Comment analysis by the server

[0566] The server receives the request and analyzes the comment. Using a natural language processing (NLP) model, such as OpenAI's GPT-3 model, the server calculates the comment's offensiveness score and generates a rewrite. If the offensiveness score is 80 / 100, the rewrite reads, "There are many problems with this news. We hope that the relevant parties will improve their response."

[0567] Input: Analysis request (user ID, comment content)

[0568] Output: Analysis results (aggression score, rewritten sentence)

[0569] Specific operation: The server executes the NLP model to analyze the comments and generate rewritten sentences.

[0570] Step 4: Returning the analysis results

[0571] The server creates a response containing the generated aggressiveness score and the rewritten comment, and sends it back to the user's device. The response is also in JSON format, for example, {"aggressiveness_score": 80, "rewritten_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[0572] Input: Analysis results (aggression score, rewritten text)

[0573] Output: Response to user device

[0574] Specific operation: The server creates an HTTP response and sends it to the terminal via the Internet.

[0575] Step 5: Presenting and selecting rewrite proposals

[0576] The user's device receives the response from the server and displays the offensiveness score and the rewritten text to the user. The user can choose to accept the rewritten comment or post the original comment as is. For example, if the user accepts the rewritten comment, the rewritten comment is saved on the device as the final comment. If the user rejects the rewritten comment, the original comment is saved as the final comment.

[0577] Input: Response from the server (aggression score, rewritten text)

[0578] Output: Final comment

[0579] Specific operation: The user's device displays rewrite suggestions, and the user presses the selection button.

[0580] Step 6: Submitting final comments

[0581] The device then sends the final comment and user ID to the server again in JSON format, for example, {"user_id": "12345", "final_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[0582] Input: Last comment, User ID

[0583] Output: Sending the final comment to the server

[0584] Specific operation: The terminal creates an HTTP request including the final comment and sends it to the server.

[0585] Step 7: Update the ranking system

[0586] The server saves the final comment in the database and updates the user's average offensiveness score. This recalculates the user's rank and generates an updated ranking. The server then sends a response back to the user device indicating that the update is complete. For example, the response will be in the format {"status": "success", "new_rank": 5}.

[0587] Input: Final comment

[0588] Output: Updated rank

[0589] Specific operation: The server updates the database, calculates the user's new rank, generates a response and sends it to the device.

[0590] (Application example 1)

[0591] 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."

[0592] In the comment sections of content distribution services, many users often post offensive comments, which can undermine the health of the dialogue. There is a need for an appropriate system to promote communication between users and provide a healthy dialogue environment. However, current technology lacks a mechanism for analyzing comments in real time, scoring their offensiveness, and suggesting rewrites. Furthermore, there is a lack of a system that can promote healthy comment posting by scoring and ranking users' commenting behavior.

[0593] 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.

[0594] In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for offensive comments, and means for updating user scores and rankings. This makes it possible to analyze the aggression of comments in real time and provide a healthy dialogue environment. Furthermore, by suggesting appropriate rewrites to users, the number of offensive comments can be reduced. Furthermore, gamification elements can be used to encourage users to post healthy comments, and a ranking system can increase user participation.

[0595] The "real-time analysis means" is a function that analyzes user comments as they are entered and instantly scores the degree of aggressiveness based on their content.

[0596] The "rewrite suggestion method" is a function that, when an offensive comment is entered, presents the user with rewrite suggestions to revise the content to make it more appropriate and constructive.

[0597] The "gamification element providing means" is a function that introduces game-like elements into user behavior, scores comment posts, and ranks them based on the average score obtained.

[0598] "Smart device display means" is a function that displays the comment's offensiveness score and rewritten text on the screen of the device used by the user, such as a smartphone or tablet.

[0599] "Scoring and ranking update means" is a function that calculates an aggressiveness score based on comments posted by users, records the score in a database, and updates the user's rating and ranking.

[0600] The "server" is a computer system that receives comments sent by users, analyzes them, generates an offensiveness score and a rewritten sentence, and stores the final comments in a database.

[0601] System Overview

[0602] The system analyzes the content of comments entered by users in real time, scores their offensiveness, and suggests rewrites. It also includes a gamification element that scores and ranks users' behavior.

[0603] Hardware Configuration

[0604] The system consists of the following major hardware components:

[0605] 1. User Device: An end-user device, including a smartphone or tablet, that allows for the input of comments and the display of suggested rewrites.

[0606] 2. Server: A central computer system that processes analysis and manages data.

[0607] Software Configuration

[0608] The system consists of the following major software components:

[0609] 1. Comment analysis module: Generative AI model using Python and TensorFlow.

[0610] 2. Rewrite generation module: Processes text data and generates appropriate rewrites for offensive comments.

[0611] 3. Database: Stores user comments, scores, and ranks (e.g. MySQL).

[0612] 4. User interface: An application that runs on a smart device and allows users to enter comments, view analysis results, and propose and adopt rewrite proposals.

[0613] Program processing flow

[0614] User side:

[0615] 1. Comment input: The user enters text into the comment input field of the smartphone app.

[0616] 2. Send: The entered comment is sent to the analysis server along with the user ID as an HTTP POST request.

[0617] Server side:

[0618] 3. Real-time analysis: The server receives the request and analyzes the comment using a generative AI model using Python and TensorFlow. The analysis results in an offensiveness score and generates a rewritten sentence.

[0619] Example: If the comment "This news is terrible. Aren't all the people involved stupid?" is entered, the offensiveness score will be evaluated as 80 / 100, and a rewrite sentence will be generated that reads "There are many problems with this news. I hope that the people involved will respond better."

[0620] 4. Returning the results: The analysis results (offensiveness score and rewritten sentence) are returned from the server to the user's device.

[0621] User side:

[0622] 5. Display and selection: The smart device receives the analysis results sent from the server and presents them to the user. The user can then choose whether to adopt the rewritten sentence.

[0623] 6. Sending the final comment: The final comment that adopts the rewritten sentence is sent to the server again and saved in the database.

[0624] Server side:

[0625] 7. Scoring and Ranking Update: The server updates the average score of the user based on the last comment and recalculates the ranking. The latest ranking information is sent back to the user's device.

[0626] Examples of specific examples and prompts

[0627] As a concrete example, consider the case where a user writes a review of a TV drama. If the user writes, "This drama is not interesting at all. All the actors' acting is terrible," this comment is sent to the analysis server. The generative AI model analyzes the comment on the server side, assigns an offensiveness score of 70 / 100, and suggests a rewrite such as, "This drama did not meet my expectations. I look forward to seeing the actors grow."

[0628] Example prompt sentence:

[0629] If a comment is typed: "This news is terrible. Are all the people involved stupid?"

[0630] Input the text "This news is terrible. Are all the people involved stupid?" into the analytical AI model and generate an offensiveness score and an improved rewrite.

[0631] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0632] Program processing flow

[0633] Step 1:

[0634] Users enter text into the comment input field of the smartphone app, which is temporarily stored on the user's device.

[0635] input:

[0636] The comment text entered by the user.

[0637] output:

[0638] The comment text is temporarily saved on the user's device.

[0639] Step 2:

[0640] The user device sends an HTTP POST request to the server to analyze the comment, which includes the user ID and the comment text.

[0641] input:

[0642] User ID, comment text.

[0643] output:

[0644] An HTTP POST request containing the user ID and comment text is sent to the server.

[0645] Step 3:

[0646] The server receives the request and analyzes the comments using a generative AI model powered by Python and TensorFlow.

[0647] input:

[0648] Request data (user ID, comment text).

[0649] output:

[0650] Aggression scores and rewritten statements.

[0651] Data processing and calculation:

[0652] The server launches a generative AI model that generates an offensiveness score and rewrite sentences using the input comment text as prompts.

[0653] Step 4:

[0654] The server sends the generated offensive score and the rewritten sentence back to the user's device.

[0655] input:

[0656] Aggression score, rewritten sentence.

[0657] output:

[0658] A response containing the offensiveness score and the rewritten sentence is sent to the user's device.

[0659] Step 5:

[0660] The user device receives the offensiveness score and rewritten sentences sent from the server and presents them to the user, who can then review the rewritten sentences and choose whether or not to adopt them.

[0661] input:

[0662] Response from the server (aggression score, rewritten text).

[0663] output:

[0664] The offensiveness score and rewritten sentence presented to the user.

[0665] Specific behavior:

[0666] The user's device displays the received offensiveness score and rewritten text on the screen, and the user confirms them.

[0667] Step 6:

[0668] If the user accepts the rewritten comment, it is sent to the server again as a final comment. This request includes the user ID and the final comment.

[0669] input:

[0670] User ID, last comment (rewritten text).

[0671] output:

[0672] An HTTP POST request containing the user ID and the last comment is sent to the server.

[0673] Step 7:

[0674] The server receives the final comments and stores them in the database, while calculating the average score of the users and updating the ranking.

[0675] input:

[0676] Final comment.

[0677] output:

[0678] Updated user ranking information.

[0679] Data processing and calculation:

[0680] The server stores the final comment received in a database, compares it with the user's past comment scores, calculates an average score, and updates the user's ranking based on the result to generate the latest ranking information.

[0681] 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.

[0682] The present invention provides a system that analyzes in real time how a user's comments are perceived by others and scores the degree of offensiveness when the user enters a comment. The system also combines a gamification element that suggests improved rewrites for offensive comments and ranks them based on the user's average score, with an emotion engine that recognizes the user's emotions.

[0683] System Overview

[0684] The system consists of the following main components:

[0685] User terminal: A device used to input comments and interact with the system.

[0686] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[0687] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0688] Program processing flow

[0689] User comment input

[0690] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[0691] Sending a comment analysis request

[0692] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0693] AI-powered comment analysis

[0694] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite, and uses an emotion engine to recognize the user's emotional state. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better." The server also recognizes the user's emotion as anger.

[0695] Returning analysis results

[0696] The server creates a response that includes an offensiveness score, suggested rewrites, and the emotion recognized by the emotion engine.

[0697] Example: Aggression score 80, rewrite suggestion: "There are many problems with this news. I hope that the people involved will respond better.", user emotional state: anger.

[0698] Presenting and selecting rewrite proposals

[0699] The user terminal receives the response sent back from the server and displays the analysis results to the user.

[0700] The user device presents the offensiveness score and the rewrite suggestion, and prompts the user to choose whether to adopt the rewrite suggestion. The user decides whether to adopt the rewrite suggestion. For example, if the user adopts the rewrite suggestion, the content is saved as the final comment on the user device.

[0701] Submitting final comments

[0702] The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. The final comment is sent to the server.

[0703] Ranking System Update

[0704] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[0705] Specific examples

[0706] For example, imagine a user comments, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI and an emotion engine. It is determined to have a high offensive score, and a rewrite suggestion is generated. The user's emotion is also recognized as "anger," and a rewrite suggestion is presented based on this result. If the user adopts this rewrite suggestion, it is sent to the server as the final comment and saved in the database. This updates the user's average score, which is also reflected in the rankings. In this way, users are encouraged to voluntarily post less aggressive comments. This process discourages aggressive comments in the comment section, providing a healthy dialogue environment.

[0707] The processing flow will be explained below.

[0708] Specific processing steps of the program

[0709] Step 1:

[0710] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[0711] Terminal: Obtains and temporarily stores comments entered by users.

[0712] Step 2:

[0713] Device: Create a request to analyze the saved comments. This request contains the user ID, the comment, and the user's device session information.

[0714] Terminal: Sends the created request to the server.

[0715] Step 3:

[0716] Server: Receives requests from user devices.

[0717] Server: The received comments are passed to an AI analysis model for analysis, where an offensiveness score is calculated and a rewrite proposal is generated.

[0718] Step 4:

[0719] Server: Based on the analysis of the comments, an emotion engine is used to identify the user's emotional state. For example, the user's comment is recognized as being based on anger.

[0720] Server: Adjusts rewrite suggestions based on the user's emotional state. In the future, especially in urgent cases, the server will determine the appropriate timing to present rewrite suggestions to the user according to the emotion recognition results.

[0721] Step 5:

[0722] Server: Create a response that includes the offensiveness score, emotional state, and rewrite suggestion. For example, the offensiveness score is 80 / 100, and the rewrite suggestion is "There are many problems with this news. I hope that the people involved will respond better."

[0723] Server: Returns the created response to the user device.

[0724] Step 6:

[0725] Terminal: Receives the response sent back from the server and displays the analysis results (aggression score, emotional state, rewrite suggestions) to the user.

[0726] Device: Ask the user whether they would like to adopt the proposed rewrite: "There are many problems with this news. We hope that those involved will respond better."

[0727] User: Review the rewrite suggestion and decide whether to adopt it. For example, if the user adopts the rewrite suggestion.

[0728] Step 7:

[0729] Device: Create a request to send the user's final comment (in this case, a rewrite proposal) back to the server. This request includes the user ID, the final comment, and the emotional state.

[0730] Terminal: Send a request to the server containing the final comment.

[0731] Step 8:

[0732] Server: Receives the final comment from the user terminal.

[0733] Server: Stores received comments in a database, including the comment, its emotional state, and its aggression score.

[0734] Step 9:

[0735] Server: Update the average score of the user based on the offensiveness score of the last comment. Update the rank based on the average score of the user.

[0736] Server: Returns a response to the user device indicating that the rank update has been completed.

[0737] Specific examples

[0738] For example, if a user comments, "This news is terrible. Aren't all the people involved stupid?", the system sends the comment to the server. The server uses an AI model to calculate an aggression score and simultaneously recognizes the user's emotional state using an emotion engine. As a result, the aggression score is determined to be 80 / 100, and the user's emotion is recognized as "anger." The server then generates a rewrite suggestion, "There are many problems with this news. I hope the people involved will improve their response," and presents it to the user. If the user adopts the rewrite suggestion, the comment is sent to the server and stored in the database. The user's average aggression score is also updated, along with their rank. This process encourages users to voluntarily refrain from making aggressive comments, creating a healthy dialogue environment.

[0739] Example 2

[0740] 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."

[0741] There are many offensive comments in online comment sections and message boards, which is a problem as it damages a healthy dialogue environment. There are also cases where users post offensive comments without realizing it, which calls for improvement. Furthermore, there is a lack of ways for users to learn what words are offensive and to communicate better.

[0742] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a comment, a means for transmitting the comment to the server, a means for the server to analyze the comment in real time and score the degree of aggression, a means for generating a rewritten sentence based on the aggression score, a means for presenting the rewritten sentence to the user and prompting the user to select, a means for providing a gamification element that ranks the rewritten sentences based on the average score obtained by the users, and a means for storing the comments and the analysis results in a database. This encourages users to voluntarily post less aggressive comments, and provides a healthy dialogue environment and enables users to learn.

[0743] "User" refers to the entity that accesses the system and enters comments.

[0744] A "comment" is text information entered by a user that accompanies content such as a news article or message board.

[0745] "Server" refers to the central processing unit that receives and analyzes comments.

[0746] "Real-time" refers to a situation where there is almost no delay between the time a comment is entered and the time the analysis results are returned.

[0747] "Level of aggression" refers to a numerical score that indicates how aggressive the content of a comment is.

[0748] A "rewrite" refers to more appropriately worded text that is generated to make the original comment less offensive.

[0749] "Analysis results" refers to information such as scores, rewritten sentences, and emotional state obtained after the server analyzes the comments.

[0750] "Gamification elements" are elements that make user behavior fun and motivating, like a game, and specifically include ranking and reward systems.

[0751] "Database" refers to a collection of information for storing comments and analysis results.

[0752] An "emotion engine" refers to a mechanism that analyzes and detects emotional states (such as anger or joy) from user comments.

[0753] The embodiment of the present invention is a system that, when a user inputs a comment, analyzes in real time how the content of the comment will be received by others and scores the degree of offensiveness. A detailed description of this system is provided below.

[0754] System Overview

[0755] The system consists of the following main components:

[0756] User terminal: A device used to input comments and interact with the system. User terminals include personal computers and smartphones.

[0757] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[0758] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0759] Hardware and software used

[0760] User device: A device that can connect to the Internet (e.g., smartphone, PC)

[0761] Server: A high-performance computer using cloud services

[0762] Software: generative AI models (e.g., open-source natural language processing models), database management systems, emotion engines

[0763] Data processing and calculation

[0764] 1. Comment input: The user inputs a comment, and the user terminal sends this data to the server.

[0765] 2. Data analysis: The server receives the comments and uses a generative AI model to score the degree of offensiveness, and an emotion engine to analyze the user's emotional state.

[0766] 3. Generating rewritten sentences: The generative AI model generates rewritten sentences for comments that are judged to be highly offensive.

[0767] 4. Presentation of results: The analysis results and rewritten sentences are returned from the server and displayed on the user's terminal.

[0768] 5. Accepting the comment: The user selects whether to accept the rewritten sentence and sends the final comment to the server.

[0769] 6. Update ranking: The server saves the final comments to the database and updates the ranking based on the average score of the users.

[0770] Specific examples

[0771] For example, if a user types a comment like "This news is terrible. Aren't all the people involved stupid?", this comment is sent to the server, which then inputs the following prompt sentence into the generative AI model for analysis:

[0772] Example prompt sentence:

[0773] Comment Analysis:

[0774] Typed comment: "This news is terrible. Are all the people involved stupid?"

[0775] Aggression Score: 80

[0776] Suggested rewrite: "There are many problems with this news. I hope that those involved will improve their response."

[0777] Emotional state: Anger

[0778] After the analysis is complete, the server sends back the results, including the offensiveness score, rewrite suggestions, and emotional state, to the user's device. The user's device displays these results on the screen and prompts the user to decide whether to accept the rewrite suggestions. If the user accepts the rewrite suggestions, the rewritten comment is resubmitted to the server as the final comment. Finally, the final comment is saved in the database, and the ranking is updated based on the user's average score.

[0779] Through this process, users are encouraged to voluntarily post less offensive comments, creating a healthier interactive environment.

[0780] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0781] Step 1: Enter a comment

[0782] A user types a comment into the comment section of a news article or message board.

[0783] What happens: A user types "This news is terrible. Are all the people involved stupid?" This comment is stored as temporary data on the user's device.

[0784] Input: User comment

[0785] Output: Input comments as temporary data

[0786] Step 2: Send a comment analysis request

[0787] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0788] Specific operation: The user's device packages the user ID and comments in JSON format and sends an HTTP request to the server's API endpoint.

[0789] Input: User ID, Comment

[0790] Output: HTTP request to the server

[0791] Step 3: Receiving and analyzing comments

[0792] The server receives the request and analyzes the comment using a generative AI model, which generates an offensiveness score and a rewrite, and also uses an emotion engine to recognize the user's emotional state.

[0793] Specific operation: The server analyzes the request and creates a prompt sentence to input the comment part into the AI ​​model. The AI ​​model analyzes it and generates an offensiveness score, rewrite sentence, and emotional state.

[0794] Input: User ID, Comment

[0795] Output: Analysis results including aggression score, rewritten sentences, and emotional state

[0796] Step 4: Returning the analysis results

[0797] The server creates a response containing the offensiveness score, suggested rewrites, and the perceived emotion, and sends it back to the user device.

[0798] Specific operation: The server packages the analysis results in JSON format and sends them to the user device as an HTTP response.

[0799] Input: Aggression score, rewritten sentence, emotional state

[0800] Output: HTTP response to the user's device

[0801] Step 5: Presenting and selecting rewrite proposals

[0802] The user device receives the response and displays the analysis results to the user, who can then decide whether to adopt the proposed rewrite.

[0803] Specific operation: The user's device parses the response, displays it on the screen, and presents rewrite suggestions. The user can choose whether to adopt the rewrite suggestions.

[0804] Input: Analysis results (aggression score, rewritten sentence, emotional state)

[0805] Output: User's choice (whether to accept the rewrite proposal or not)

[0806] Step 6: Submitting final comments

[0807] The user device creates a request to resend the last comment to the server, including the user ID and the last comment.

[0808] Specific operation: The user's device packages the user ID and the last comment in JSON format and sends an HTTP request to the server's API endpoint.

[0809] Input: User ID, last comment

[0810] Output: HTTP request to the server

[0811] Step 7: Update the ranking system

[0812] The server saves the final comments to the database and updates the user's average score, which in turn updates the user's rank to generate the latest ranking.

[0813] Specific operation: The server saves the final comment in the database, calculates a new average score based on the user's previous comment scores, updates the ranking database, and sends an update completion message to the user's device.

[0814] Input: Last comment, user's past score

[0815] Output: Updated ranking, update complete message

[0816] (Application example 2)

[0817] 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."

[0818] Current content distribution services often have users posting offensive comments, making it difficult to maintain a healthy dialogue environment. Furthermore, because users often post offensive comments without being aware of their own feelings, a mechanism to curb this is needed. Furthermore, there is a lack of gamification elements to encourage users to voluntarily adopt positive behavior.

[0819] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for aggressive comments, means for providing a gamification element that ranks comments based on the average points earned by users, means for recognizing users' emotional states in real time and providing that information, and means for applying it to the comment section of the content distribution service. This enables users to voluntarily post less aggressive comments and maintain a healthy dialogue environment.

[0820] "User" refers to a user who posts comments within the content distribution service.

[0821] "Comment" refers to a text message that a user enters within the Content Distribution Service and shares with others.

[0822] "Real-time analysis" refers to the process of instantly analyzing the content of a comment the moment it is entered and providing the results.

[0823] "Offensiveness" refers to a rating that indicates how offensive or harmful a comment may be to others.

[0824] A "rewritten sentence" refers to a sentence in which an offensive comment has been revised to use more neutral or constructive language.

[0825] "Gamification elements" refer to game-like elements that assign scores and ranks to user behavior to increase user motivation.

[0826] "Emotional state" refers to the state of a user's emotions when they are entering a comment, and includes anger, joy, sadness, etc.

[0827] "Content distribution service" refers to an online service that provides users with various content such as video, audio, and text, and allows them to share and comment on it.

[0828] "Server" refers to the computer system that analyzes comments, calculates offensiveness scores, generates rewritten text, recognizes the user's emotional state, and stores and manages data.

[0829] As an embodiment of the present invention, a specific description of the system is given below. This system is composed of the main components of a user terminal, a server, and a ranking system.

[0830] User Device

[0831] User devices are expected to include smartphones, tablets, personal computers, etc. Users enter comments into the comment section of the content distribution service through a comment form. While users are entering their comments, a system is in place to temporarily save them in real time.

[0832] server

[0833] The server analyzes comments, calculates offensiveness scores, generates rewritten sentences, and recognizes the user's emotional state. Specifically, it uses the following software and hardware:

[0834] software:

[0835] Analysis model: We use the Hugging Face text classification model to evaluate the offensiveness of comments.

[0836] Emotion Engine: Recognizes the user's emotional state using the Hugging Face emotion recognition model.

[0837] Hardware:

[0838] High-performance servers: Cloud-based high-performance servers (e.g., Amazon Web Services, Google Cloud Platform) for rapid data processing.

[0839] The server then sends the analyzed comments back to the user's device along with an offensiveness score and a rewritten version of the comment, along with the user's emotional state.

[0840] Ranking System

[0841] The ranking system provides a gamification element by ranking users based on their average score, which is based on a database managed by the server. By posting less offensive comments, users can improve their scores and move up the rankings.

[0842] Specific examples

[0843] For example, if a user types a comment like "This news is terrible. Are all the people involved stupid?", the following process will occur:

[0844] 1. Real-time analysis:

[0845] The comment is immediately sent to a server, where an analysis model calculates an offensiveness score.

[0846] The emotion engine recognizes the user's emotional state as "anger."

[0847] 2. Generating and presenting rewritten sentences:

[0848] Along with the offensiveness score, a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better."

[0849] It is sent back to the user's device in real time.

[0850] 3. User Choice:

[0851] Users can choose whether to adopt the rewrite proposal.

[0852] If adopted, the rewritten comment is resubmitted to the server as the final comment.

[0853] 4. Ranking System Update:

[0854] The server saves the final comments in the database and updates the user's average score.

[0855] The rankings are updated and feedback is provided to users.

[0856] Prompt Sentence Examples

[0857] "Please analyze comments entered by users in real time and calculate an offensiveness score. For example, let's analyze the following comment.

[0858] Comment: "This news is terrible. Aren't all the people involved stupid?"

[0859] The above embodiment makes it possible to realize a mechanism that suppresses offensive comments in the comment section and provides a healthy dialogue environment.

[0860] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0861] Step 1:

[0862] The user enters a comment. The user uses a smartphone or personal computer to enter text into the comment field of the content distribution service. This comment is temporarily saved on the user's device. Input: Text data of the comment. Output: Text data temporarily saved on the device.

[0863] Step 2:

[0864] Prepare to send the comment to the server. The user's device generates a request to send the entered comment to the server in real time. This request includes the user ID and comment. Input: User ID and comment text data. Output: Request data to be sent to the server.

[0865] Step 3:

[0866] The server receives the comment. The server receives a request from the user's device and obtains the comment text data and user ID. Input: Request data. Output: Comment text data and user ID.

[0867] Step 4:

[0868] Analyze the comments. The server uses an analysis model (Hugging Face's text classification model) to calculate the comment's aggressiveness score. It also uses an emotion engine (Hugging Face's emotion recognition model) to recognize the user's emotional state. Input: Comment text data. Output: Aggression score and emotional state.

[0869] Step 5:

[0870] Generate a rewritten sentence. The server generates a rewritten sentence that changes the comment to a more neutral or constructive expression based on the offensiveness score. Input: Comment text data and offensiveness score. Output: Rewritten sentence.

[0871] Step 6:

[0872] The analysis results are returned to the user device. The server creates a response including the offensiveness score, rewrite suggestion, and emotional state, and returns it to the user device. Input: offensiveness score, rewrite sentence, emotional state. Output: response data.

[0873] Step 7:

[0874] The analysis results are presented. The user device receives the response from the server and displays the offensiveness score, rewrite suggestions, and emotional state to the user. Input: Response data. Output: Display data on the device.

[0875] Step 8:

[0876] The user makes a choice. The user chooses whether to adopt the proposed rewrite. If the user adopts the rewrite, the content is saved as the final comment. Input: User's choice. Output: Text data of the final comment.

[0877] Step 9:

[0878] Send the final comment to the server. The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. Input: User ID and text data of the final comment. Output: Request data to be sent to the server.

[0879] Step 10:

[0880] Update the ranking system. The server saves the final comment to the database and updates the user's average score. Based on that score, the user's rank is updated and the latest ranking is generated. Input: Text data of the final comment. Output: Updated ranking data.

[0881] Through the above steps, a system is realized that analyzes users' comments and encourages them to post less offensive comments.

[0882] 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.

[0883] 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.

[0884] 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.

[0885] [Third embodiment]

[0886] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0887] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0888] 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).

[0889] 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.

[0890] 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.

[0891] 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).

[0892] 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.

[0893] 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.

[0894] 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.

[0895] 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.

[0896] 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.

[0897] 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."

[0898] The embodiment of the present invention provides a system that analyzes in real time how the content of a comment entered by a user will be perceived by others and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average points earned by users.

[0899] System Overview

[0900] The system consists of the following main components:

[0901] User terminal: A device used to input comments and interact with the system.

[0902] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, and manages rankings.

[0903] Ranking system: Ranking based on users' average scores, providing a gamification element.

[0904] Program processing flow

[0905] User comment input

[0906] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[0907] Sending a comment analysis request

[0908] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[0909] AI-powered comment analysis

[0910] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better."

[0911] Returning analysis results

[0912] The server creates a response containing the offensiveness score and the rewritten text and sends it back to the user's device.

[0913] Presenting and selecting rewrite proposals

[0914] The user device receives the response from the server and presents the offensiveness score and rewritten sentence to the user. The user can choose whether to adopt the proposed rewritten sentence. For example, if the user adopts the rewritten sentence, the content is saved as the final comment on the user device.

[0915] Submitting final comments

[0916] The user device sends the final comment back to the server, and this request includes the user ID and the final comment.

[0917] Ranking System Update

[0918] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[0919] Specific examples

[0920] For example, consider a case where a user enters a comment such as, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI. It is determined to have a high offensive score, and a rewrite proposal is generated. The rewrite proposal, "There are many problems with this news. I hope that the people involved will improve their response," is presented to the user. The user adopts the rewrite proposal and sends the final comment to the server. The server saves this comment in a database and updates the ranking.

[0921] This process will help curb offensive posts in the comments section and provide a healthy interactive environment.

[0922] The processing flow will be explained below.

[0923] Specific processing steps of the program

[0924] Step 1:

[0925] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[0926] Terminal: Obtains and temporarily stores comments entered by users.

[0927] Step 2:

[0928] On your device: Create a request to analyze the saved comments. This request includes the user ID and the comment.

[0929] Terminal: Sends the created request to the server.

[0930] Step 3:

[0931] Server: Receives requests from user devices.

[0932] Server: The received comments are passed to the AI ​​analysis model for analysis, where the offensiveness score of the comment is calculated and a rewrite proposal is generated.

[0933] Step 4:

[0934] Server: Create a response containing an offensiveness score and a rewrite suggestion as the analysis result. For example, if the offensiveness score is 80 / 100, the rewrite suggestion is "There are many problems with this news. We hope that the relevant parties will respond better."

[0935] Server: Returns the created response to the user device.

[0936] Step 5:

[0937] Terminal: Receives the response sent back from the server and displays the analysis results to the user.

[0938] Device: Presents the offensiveness score and rewrite suggestions, prompting the user to choose whether or not to adopt the rewrite suggestions.

[0939] User: Checks the rewrite proposal and decides whether to adopt it. For example, adopt the rewrite proposal "There are many problems with this news. I hope that the people involved will respond better."

[0940] Step 6:

[0941] On the device: The user creates a request to send the final comment (in this case, the rewrite proposal) back to the server. This request includes the user ID and the final comment.

[0942] Terminal: Send a request to the server containing the final comment.

[0943] Step 7:

[0944] Server: Receives the final comment from the user terminal.

[0945] Server: Stores received comments in a database.

[0946] Server: Updates the average score of users based on the offensiveness score of their comments and processes the ranking.

[0947] Server: Returns a response to the user device indicating that the ranking update has been completed.

[0948] For example, if a user enters an offensive comment such as "This news is terrible. Aren't all the people involved stupid?", the system will analyze the comment using AI and assign a high offensiveness score (80 / 100). At the same time, the system will generate a rewrite suggestion, "There are many problems with this news. I hope that the people involved will improve their response," and present it to the user. If the user adopts this rewrite suggestion, it will be sent to the server as the final comment and saved in the database. This will also update the user's average score, which will be reflected in the rankings. In this way, users are encouraged to voluntarily post less offensive comments.

[0949] Example 1

[0950] 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."

[0951] In recent years, the number of offensive comments has increased in the comment sections of online platforms, hindering a healthy dialogue environment. There is a need for a system that can detect such offensive comments in real time and allow users to correct them themselves. Gamification elements are also necessary to increase user motivation.

[0952] 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.

[0953] In this invention, the server includes: means for analyzing user comments in real time and scoring the degree of offensiveness; means for proposing improved rewrites for offensive comments; means for providing a gamification element that ranks comments based on the average points earned by users; a terminal for inputting and submitting comments and presenting rewrites; a server equipped with an analytical model for generating offensiveness scores and rewrite suggestions; and a system for displaying rankings based on user comment scores. This makes it possible to provide a healthy interactive environment and increase user motivation by detecting offensive comments in real time and presenting improvement suggestions.

[0954] "User" refers to the end user who enters comments and interacts with the system.

[0955] "Comment" means a text message entered by a User on the Online Platform.

[0956] "Real-time" refers to the fact that user actions and the system's response to them are immediate.

[0957] "Offensiveness" refers to the degree to which the content of a comment is offensive or unpleasant to others.

[0958] "Score" refers to a score used to quantitatively evaluate the offensiveness of a comment.

[0959] "Rewrites" are text messages that convert offensive comments into more gentle and wholesome language.

[0960] "Suggest" means presenting the user with a rewritten statement and giving them the option to correct the original comment.

[0961] "Gamification" refers to the incorporation of game elements to increase user motivation.

[0962] "Average score" refers to the average aggression score a user has earned in the past.

[0963] "Ranking" refers to ranking based on the average score achieved by users.

[0964] "Terminal" refers to the device (smartphone, PC, etc.) through which a user enters comments and interacts with the system.

[0965] "Server" refers to the central computer system used to analyze comments, assign offensiveness scores, generate rewrites, and manage rankings.

[0966] "Analysis model" refers to natural language processing technology used to analyze the content of comments and generate offensiveness scores and rewrite suggestions.

[0967] "Ranking System" refers to a system that displays rankings based on users' comment scores and provides gamification elements.

[0968] A "natural language processing model" refers to an analysis algorithm that uses technology to understand and analyze human language.

[0969] This invention provides a system that analyzes in real time how a user's comment content will be perceived by others when the user enters it, and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average score earned by the user. Specifically, it consists of the following main components:

[0970] User Device

[0971] A user terminal is a device through which users can enter comments on news articles and interact with the system. For example, a smartphone or PC is an example. Users can enter comments through the terminal and check the analysis results in real time.

[0972] server

[0973] The server is a central computer system that receives and analyzes comments submitted by users. The server is equipped with OpenAI's GPT model as a natural language processing (NLP) model, which analyzes comments, assigns offensiveness scores, and generates rewritten sentences.

[0974] Ranking System

[0975] The ranking system displays a ranking based on the average offensiveness score of users and provides a gamification element. The system stores users' comment history and offensiveness scores in a database and updates the user's rank based on the results.

[0976] Explanation of program processing

[0977] When the server receives a user's comment, it analyzes it using an NLP model (e.g., OpenAI's GPT-3). This analysis calculates an offensiveness score for the comment. For offensive comments, it generates a rewritten text with milder language. For example, if a comment such as "This news is terrible. Aren't all the people involved stupid?" is entered, it is determined to have a high offensive score, and a rewrite suggestion such as "There are many problems with this news. I hope that the people involved will respond better." is generated.

[0978] The generated offensiveness score and rewrite proposal are sent back from the server to the user's device. The user can review them on their device and choose whether or not to adopt the proposed rewrite. If the user adopts the rewrite, it is saved on the device as a final comment and sent back to the server. The server saves this final comment in a database and updates the user's average offensiveness score. An updated ranking is generated based on this score and sent back to the user's device.

[0979] Specific examples

[0980] Suppose a user enters a comment such as "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by the AI ​​model. As a result of the analysis, the offensiveness score is determined to be 80 / 100, and a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better." If the user adopts the rewrite suggestion and sends the final comment to the server, the server saves this comment in the database and updates the ranking.

[0981] Prompt Sentence Examples

[0982] "Analyze how an input comment would be perceived by others, score the level of offensiveness, and suggest a milder rewrite that expresses the same meaning. Example: Comment: 'This news is terrible. Are all the people involved stupid?'"

[0983] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0984] Step 1: User comments

[0985] A user uses a device to enter text into a comment section of an online platform, for example, "This news is terrible. Are all the people involved stupid?" This comment is temporarily stored in the device's memory.

[0986] Input: Comments entered by the user

[0987] Output: Saved comments

[0988] Specific operation: The user enters a comment using a keyboard or touch screen and presses the send button.

[0989] Step 2: Submitting a comment analysis request

[0990] The device creates an HTTP request to send an analysis request including the entered comment and user ID to the server. The request includes the user ID and comment content in JSON format. For example, it is sent in the following format: {"user_id": "12345", "comment": "This news is terrible. Aren't all the people involved stupid?"}

[0991] Input: Temporarily saved comment, user ID

[0992] Output: Parse request to server

[0993] Specific operation: The device creates an HTTP request and sends it to the server over the Internet.

[0994] Step 3: Comment analysis by the server

[0995] The server receives the request and analyzes the comment. Using a natural language processing (NLP) model, such as OpenAI's GPT-3 model, the server calculates the comment's offensiveness score and generates a rewrite. If the offensiveness score is 80 / 100, the rewrite reads, "There are many problems with this news. We hope that the relevant parties will improve their response."

[0996] Input: Analysis request (user ID, comment content)

[0997] Output: Analysis results (aggression score, rewritten sentence)

[0998] Specific operation: The server executes the NLP model to analyze the comments and generate rewritten sentences.

[0999] Step 4: Returning the analysis results

[1000] The server creates a response containing the generated aggressiveness score and the rewritten comment, and sends it back to the user's device. The response is also in JSON format, for example, {"aggressiveness_score": 80, "rewritten_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[1001] Input: Analysis results (aggression score, rewritten text)

[1002] Output: Response to user device

[1003] Specific operation: The server creates an HTTP response and sends it to the terminal via the Internet.

[1004] Step 5: Presenting and selecting rewrite proposals

[1005] The user's device receives the response from the server and displays the offensiveness score and the rewritten text to the user. The user can choose to accept the rewritten comment or post the original comment as is. For example, if the user accepts the rewritten comment, the rewritten comment is saved on the device as the final comment. If the user rejects the rewritten comment, the original comment is saved as the final comment.

[1006] Input: Response from the server (aggression score, rewritten text)

[1007] Output: Final comment

[1008] Specific operation: The user's device displays rewrite suggestions, and the user presses the selection button.

[1009] Step 6: Submitting final comments

[1010] The device then sends the final comment and user ID to the server again in JSON format, for example, {"user_id": "12345", "final_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[1011] Input: Last comment, User ID

[1012] Output: Sending the final comment to the server

[1013] Specific operation: The terminal creates an HTTP request including the final comment and sends it to the server.

[1014] Step 7: Update the ranking system

[1015] The server saves the final comment in the database and updates the user's average offensiveness score. This recalculates the user's rank and generates an updated ranking. The server then sends a response back to the user device indicating that the update is complete. For example, the response will be in the format {"status": "success", "new_rank": 5}.

[1016] Input: Final comment

[1017] Output: Updated rank

[1018] Specific operation: The server updates the database, calculates the user's new rank, generates a response and sends it to the device.

[1019] (Application example 1)

[1020] 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."

[1021] In the comment sections of content distribution services, many users often post offensive comments, which can undermine the health of the dialogue. There is a need for an appropriate system to promote communication between users and provide a healthy dialogue environment. However, current technology lacks a mechanism for analyzing comments in real time, scoring their offensiveness, and suggesting rewrites. Furthermore, there is a lack of a system that can promote healthy comment posting by scoring and ranking users' commenting behavior.

[1022] 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.

[1023] In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for offensive comments, and means for updating user scores and rankings. This makes it possible to analyze the aggression of comments in real time and provide a healthy dialogue environment. Furthermore, by suggesting appropriate rewrites to users, the number of offensive comments can be reduced. Furthermore, gamification elements can be used to encourage users to post healthy comments, and a ranking system can increase user participation.

[1024] The "real-time analysis means" is a function that analyzes user comments as they are entered and instantly scores the degree of aggressiveness based on their content.

[1025] The "rewrite suggestion method" is a function that, when an offensive comment is entered, presents the user with rewrite suggestions to revise the content to make it more appropriate and constructive.

[1026] The "gamification element providing means" is a function that introduces game-like elements into user behavior, scores comment posts, and ranks them based on the average score obtained.

[1027] "Smart device display means" is a function that displays the comment's offensiveness score and rewritten text on the screen of the device used by the user, such as a smartphone or tablet.

[1028] "Scoring and ranking update means" is a function that calculates an aggressiveness score based on comments posted by users, records the score in a database, and updates the user's rating and ranking.

[1029] The "server" is a computer system that receives comments sent by users, analyzes them, generates an offensiveness score and a rewritten sentence, and stores the final comments in a database.

[1030] System Overview

[1031] The system analyzes the content of comments entered by users in real time, scores their offensiveness, and suggests rewrites. It also includes a gamification element that scores and ranks users' behavior.

[1032] Hardware Configuration

[1033] The system consists of the following major hardware components:

[1034] 1. User Device: An end-user device, including a smartphone or tablet, that allows for the input of comments and the display of suggested rewrites.

[1035] 2. Server: A central computer system that processes analysis and manages data.

[1036] Software Configuration

[1037] The system consists of the following major software components:

[1038] 1. Comment analysis module: Generative AI model using Python and TensorFlow.

[1039] 2. Rewrite generation module: Processes text data and generates appropriate rewrites for offensive comments.

[1040] 3. Database: Stores user comments, scores, and ranks (e.g. MySQL).

[1041] 4. User interface: An application that runs on a smart device and allows users to enter comments, view analysis results, and propose and adopt rewrite proposals.

[1042] Program processing flow

[1043] User side:

[1044] 1. Comment input: The user enters text into the comment input field of the smartphone app.

[1045] 2. Send: The entered comment is sent to the analysis server along with the user ID as an HTTP POST request.

[1046] Server side:

[1047] 3. Real-time analysis: The server receives the request and analyzes the comment using a generative AI model using Python and TensorFlow. The analysis results in an offensiveness score and generates a rewritten sentence.

[1048] Example: If the comment "This news is terrible. Aren't all the people involved stupid?" is entered, the offensiveness score will be evaluated as 80 / 100, and a rewrite sentence will be generated that reads "There are many problems with this news. I hope that the people involved will respond better."

[1049] 4. Returning the results: The analysis results (offensiveness score and rewritten sentence) are returned from the server to the user's device.

[1050] User side:

[1051] 5. Display and selection: The smart device receives the analysis results sent from the server and presents them to the user. The user can then choose whether to adopt the rewritten sentence.

[1052] 6. Sending the final comment: The final comment that adopts the rewritten sentence is sent to the server again and saved in the database.

[1053] Server side:

[1054] 7. Scoring and Ranking Update: The server updates the average score of the user based on the last comment and recalculates the ranking. The latest ranking information is sent back to the user's device.

[1055] Examples of specific examples and prompts

[1056] As a concrete example, consider the case where a user writes a review of a TV drama. If the user writes, "This drama is not interesting at all. All the actors' acting is terrible," this comment is sent to the analysis server. The generative AI model analyzes the comment on the server side, assigns an offensiveness score of 70 / 100, and suggests a rewrite such as, "This drama did not meet my expectations. I look forward to seeing the actors grow."

[1057] Example prompt sentence:

[1058] If a comment is typed: "This news is terrible. Are all the people involved stupid?"

[1059] Input the text "This news is terrible. Are all the people involved stupid?" into the analytical AI model and generate an offensiveness score and an improved rewrite.

[1060] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1061] Program processing flow

[1062] Step 1:

[1063] Users enter text into the comment input field of the smartphone app, which is temporarily stored on the user's device.

[1064] input:

[1065] The comment text entered by the user.

[1066] output:

[1067] The comment text is temporarily saved on the user's device.

[1068] Step 2:

[1069] The user device sends an HTTP POST request to the server to analyze the comment, which includes the user ID and the comment text.

[1070] input:

[1071] User ID, comment text.

[1072] output:

[1073] An HTTP POST request containing the user ID and comment text is sent to the server.

[1074] Step 3:

[1075] The server receives the request and analyzes the comments using a generative AI model powered by Python and TensorFlow.

[1076] input:

[1077] Request data (user ID, comment text).

[1078] output:

[1079] Aggression scores and rewritten statements.

[1080] Data processing and calculation:

[1081] The server launches a generative AI model that generates an offensiveness score and rewrite sentences using the input comment text as prompts.

[1082] Step 4:

[1083] The server sends the generated offensive score and the rewritten sentence back to the user's device.

[1084] input:

[1085] Aggression score, rewritten sentence.

[1086] output:

[1087] A response containing the offensiveness score and the rewritten sentence is sent to the user's device.

[1088] Step 5:

[1089] The user device receives the offensiveness score and rewritten sentences sent from the server and presents them to the user, who can then review the rewritten sentences and choose whether or not to adopt them.

[1090] input:

[1091] Response from the server (aggression score, rewritten text).

[1092] output:

[1093] The offensiveness score and rewritten sentence presented to the user.

[1094] Specific behavior:

[1095] The user's device displays the received offensiveness score and rewritten text on the screen, and the user confirms them.

[1096] Step 6:

[1097] If the user accepts the rewritten comment, it is sent to the server again as a final comment. This request includes the user ID and the final comment.

[1098] input:

[1099] User ID, last comment (rewritten text).

[1100] output:

[1101] An HTTP POST request containing the user ID and the last comment is sent to the server.

[1102] Step 7:

[1103] The server receives the final comments and stores them in the database, while calculating the average score of the users and updating the ranking.

[1104] input:

[1105] Final comment.

[1106] output:

[1107] Updated user ranking information.

[1108] Data processing and calculation:

[1109] The server stores the final comment received in a database, compares it with the user's past comment scores, calculates an average score, and updates the user's ranking based on the result to generate the latest ranking information.

[1110] 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.

[1111] The present invention provides a system that analyzes in real time how a user's comments are perceived by others and scores the degree of offensiveness when the user enters a comment. The system also combines a gamification element that suggests improved rewrites for offensive comments and ranks them based on the user's average score, with an emotion engine that recognizes the user's emotions.

[1112] System Overview

[1113] The system consists of the following main components:

[1114] User terminal: A device used to input comments and interact with the system.

[1115] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[1116] Ranking system: Ranking based on users' average scores, providing a gamification element.

[1117] Program processing flow

[1118] User comment input

[1119] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[1120] Sending a comment analysis request

[1121] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[1122] AI-powered comment analysis

[1123] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite, and uses an emotion engine to recognize the user's emotional state. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better." The server also recognizes the user's emotion as anger.

[1124] Returning analysis results

[1125] The server creates a response that includes an offensiveness score, suggested rewrites, and the emotion recognized by the emotion engine.

[1126] Example: Aggression score 80, rewrite suggestion: "There are many problems with this news. I hope that the people involved will respond better.", user emotional state: anger.

[1127] Presenting and selecting rewrite proposals

[1128] The user terminal receives the response sent back from the server and displays the analysis results to the user.

[1129] The user device presents the offensiveness score and the rewrite suggestion, and prompts the user to choose whether to adopt the rewrite suggestion. The user decides whether to adopt the rewrite suggestion. For example, if the user adopts the rewrite suggestion, the content is saved as the final comment on the user device.

[1130] Submitting final comments

[1131] The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. The final comment is sent to the server.

[1132] Ranking System Update

[1133] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[1134] Specific examples

[1135] For example, imagine a user comments, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI and an emotion engine. It is determined to have a high offensive score, and a rewrite suggestion is generated. The user's emotion is also recognized as "anger," and a rewrite suggestion is presented based on this result. If the user adopts this rewrite suggestion, it is sent to the server as the final comment and saved in the database. This updates the user's average score, which is also reflected in the rankings. In this way, users are encouraged to voluntarily post less aggressive comments. This process discourages aggressive comments in the comment section, providing a healthy dialogue environment.

[1136] The processing flow will be explained below.

[1137] Specific processing steps of the program

[1138] Step 1:

[1139] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[1140] Terminal: Obtains and temporarily stores comments entered by users.

[1141] Step 2:

[1142] Device: Create a request to analyze the saved comments. This request contains the user ID, the comment, and the user's device session information.

[1143] Terminal: Sends the created request to the server.

[1144] Step 3:

[1145] Server: Receives requests from user devices.

[1146] Server: The received comments are passed to an AI analysis model for analysis, where an offensiveness score is calculated and a rewrite proposal is generated.

[1147] Step 4:

[1148] Server: Based on the analysis of the comments, an emotion engine is used to identify the user's emotional state. For example, the user's comment is recognized as being based on anger.

[1149] Server: Adjusts rewrite suggestions based on the user's emotional state. In the future, especially in urgent cases, the server will determine the appropriate timing to present rewrite suggestions to the user according to the emotion recognition results.

[1150] Step 5:

[1151] Server: Create a response that includes the offensiveness score, emotional state, and rewrite suggestion. For example, the offensiveness score is 80 / 100, and the rewrite suggestion is "There are many problems with this news. I hope that the people involved will respond better."

[1152] Server: Returns the created response to the user device.

[1153] Step 6:

[1154] Terminal: Receives the response sent back from the server and displays the analysis results (aggression score, emotional state, rewrite suggestions) to the user.

[1155] Device: Ask the user whether they would like to adopt the proposed rewrite: "There are many problems with this news. We hope that those involved will respond better."

[1156] User: Review the rewrite suggestion and decide whether to adopt it. For example, if the user adopts the rewrite suggestion.

[1157] Step 7:

[1158] Device: Create a request to send the user's final comment (in this case, a rewrite proposal) back to the server. This request includes the user ID, the final comment, and the emotional state.

[1159] Terminal: Send a request to the server containing the final comment.

[1160] Step 8:

[1161] Server: Receives the final comment from the user terminal.

[1162] Server: Stores received comments in a database, including the comment, its emotional state, and its aggression score.

[1163] Step 9:

[1164] Server: Update the average score of the user based on the offensiveness score of the last comment. Update the rank based on the average score of the user.

[1165] Server: Returns a response to the user device indicating that the rank update has been completed.

[1166] Specific examples

[1167] For example, if a user comments, "This news is terrible. Aren't all the people involved stupid?", the system sends the comment to the server. The server uses an AI model to calculate an aggression score and simultaneously recognizes the user's emotional state using an emotion engine. As a result, the aggression score is determined to be 80 / 100, and the user's emotion is recognized as "anger." The server then generates a rewrite suggestion, "There are many problems with this news. I hope the people involved will improve their response," and presents it to the user. If the user adopts the rewrite suggestion, the comment is sent to the server and stored in the database. The user's average aggression score is also updated, along with their rank. This process encourages users to voluntarily refrain from making aggressive comments, creating a healthy dialogue environment.

[1168] Example 2

[1169] 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."

[1170] There are many offensive comments in online comment sections and message boards, which is a problem as it damages a healthy dialogue environment. There are also cases where users post offensive comments without realizing it, which calls for improvement. Furthermore, there is a lack of ways for users to learn what words are offensive and to communicate better.

[1171] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a comment, a means for transmitting the comment to the server, a means for the server to analyze the comment in real time and score the degree of aggression, a means for generating a rewritten sentence based on the aggression score, a means for presenting the rewritten sentence to the user and prompting the user to select, a means for providing a gamification element that ranks the rewritten sentences based on the average score obtained by the users, and a means for storing the comments and the analysis results in a database. This encourages users to voluntarily post less aggressive comments, and provides a healthy dialogue environment and enables users to learn.

[1172] "User" refers to the entity that accesses the system and enters comments.

[1173] A "comment" is text information entered by a user that accompanies content such as a news article or message board.

[1174] "Server" refers to the central processing unit that receives and analyzes comments.

[1175] "Real-time" refers to a situation where there is almost no delay between the time a comment is entered and the time the analysis results are returned.

[1176] "Level of aggression" refers to a numerical score that indicates how aggressive the content of a comment is.

[1177] A "rewrite" refers to more appropriately worded text that is generated to make the original comment less offensive.

[1178] "Analysis results" refers to information such as scores, rewritten sentences, and emotional state obtained after the server analyzes the comments.

[1179] "Gamification elements" are elements that make user behavior fun and motivating, like a game, and specifically include ranking and reward systems.

[1180] "Database" refers to a collection of information for storing comments and analysis results.

[1181] An "emotion engine" refers to a mechanism that analyzes and detects emotional states (such as anger or joy) from user comments.

[1182] The embodiment of the present invention is a system that, when a user inputs a comment, analyzes in real time how the content of the comment will be received by others and scores the degree of offensiveness. A detailed description of this system is provided below.

[1183] System Overview

[1184] The system consists of the following main components:

[1185] User terminal: A device used to input comments and interact with the system. User terminals include personal computers and smartphones.

[1186] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[1187] Ranking system: Ranking based on users' average scores, providing a gamification element.

[1188] Hardware and software used

[1189] User device: A device that can connect to the Internet (e.g., smartphone, PC)

[1190] Server: A high-performance computer using cloud services

[1191] Software: generative AI models (e.g., open-source natural language processing models), database management systems, emotion engines

[1192] Data processing and calculation

[1193] 1. Comment input: The user inputs a comment, and the user terminal sends this data to the server.

[1194] 2. Data analysis: The server receives the comments and uses a generative AI model to score the degree of offensiveness, and an emotion engine to analyze the user's emotional state.

[1195] 3. Generating rewritten sentences: The generative AI model generates rewritten sentences for comments that are judged to be highly offensive.

[1196] 4. Presentation of results: The analysis results and rewritten sentences are returned from the server and displayed on the user's terminal.

[1197] 5. Accepting the comment: The user selects whether to accept the rewritten sentence and sends the final comment to the server.

[1198] 6. Update ranking: The server saves the final comments to the database and updates the ranking based on the average score of the users.

[1199] Specific examples

[1200] For example, if a user types a comment like "This news is terrible. Aren't all the people involved stupid?", this comment is sent to the server, which then inputs the following prompt sentence into the generative AI model for analysis:

[1201] Example prompt sentence:

[1202] Comment Analysis:

[1203] Typed comment: "This news is terrible. Are all the people involved stupid?"

[1204] Aggression Score: 80

[1205] Suggested rewrite: "There are many problems with this news. I hope that those involved will improve their response."

[1206] Emotional state: Anger

[1207] After the analysis is complete, the server sends back the results, including the offensiveness score, rewrite suggestions, and emotional state, to the user's device. The user's device displays these results on the screen and prompts the user to decide whether to accept the rewrite suggestions. If the user accepts the rewrite suggestions, the rewritten comment is resubmitted to the server as the final comment. Finally, the final comment is saved in the database, and the ranking is updated based on the user's average score.

[1208] Through this process, users are encouraged to voluntarily post less offensive comments, creating a healthier interactive environment.

[1209] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1210] Step 1: Enter a comment

[1211] A user types a comment into the comment section of a news article or message board.

[1212] What happens: A user types "This news is terrible. Are all the people involved stupid?" This comment is stored as temporary data on the user's device.

[1213] Input: User comment

[1214] Output: Input comments as temporary data

[1215] Step 2: Send a comment analysis request

[1216] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[1217] Specific operation: The user's device packages the user ID and comments in JSON format and sends an HTTP request to the server's API endpoint.

[1218] Input: User ID, Comment

[1219] Output: HTTP request to the server

[1220] Step 3: Receiving and analyzing comments

[1221] The server receives the request and analyzes the comment using a generative AI model, which generates an offensiveness score and a rewrite, and also uses an emotion engine to recognize the user's emotional state.

[1222] Specific operation: The server analyzes the request and creates a prompt sentence to input the comment part into the AI ​​model. The AI ​​model analyzes it and generates an offensiveness score, rewrite sentence, and emotional state.

[1223] Input: User ID, Comment

[1224] Output: Analysis results including aggression score, rewritten sentences, and emotional state

[1225] Step 4: Returning the analysis results

[1226] The server creates a response containing the offensiveness score, suggested rewrites, and the perceived emotion, and sends it back to the user device.

[1227] Specific operation: The server packages the analysis results in JSON format and sends them to the user device as an HTTP response.

[1228] Input: Aggression score, rewritten sentence, emotional state

[1229] Output: HTTP response to the user's device

[1230] Step 5: Presenting and selecting rewrite proposals

[1231] The user device receives the response and displays the analysis results to the user, who can then decide whether to adopt the proposed rewrite.

[1232] Specific operation: The user's device parses the response, displays it on the screen, and presents rewrite suggestions. The user can choose whether to adopt the rewrite suggestions.

[1233] Input: Analysis results (aggression score, rewritten sentence, emotional state)

[1234] Output: User's choice (whether to accept the rewrite proposal or not)

[1235] Step 6: Submitting final comments

[1236] The user device creates a request to resend the last comment to the server, including the user ID and the last comment.

[1237] Specific operation: The user's device packages the user ID and the last comment in JSON format and sends an HTTP request to the server's API endpoint.

[1238] Input: User ID, last comment

[1239] Output: HTTP request to the server

[1240] Step 7: Update the ranking system

[1241] The server saves the final comments to the database and updates the user's average score, which in turn updates the user's rank to generate the latest ranking.

[1242] Specific operation: The server saves the final comment in the database, calculates a new average score based on the user's previous comment scores, updates the ranking database, and sends an update completion message to the user's device.

[1243] Input: Last comment, user's past score

[1244] Output: Updated ranking, update complete message

[1245] (Application example 2)

[1246] 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."

[1247] Current content distribution services often have users posting offensive comments, making it difficult to maintain a healthy dialogue environment. Furthermore, because users often post offensive comments without being aware of their own feelings, a mechanism to curb this is needed. Furthermore, there is a lack of gamification elements to encourage users to voluntarily adopt positive behavior.

[1248] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for aggressive comments, means for providing a gamification element that ranks comments based on the average points earned by users, means for recognizing users' emotional states in real time and providing that information, and means for applying it to the comment section of the content distribution service. This enables users to voluntarily post less aggressive comments and maintain a healthy dialogue environment.

[1249] "User" refers to a user who posts comments within the content distribution service.

[1250] "Comment" refers to a text message that a user enters within the Content Distribution Service and shares with others.

[1251] "Real-time analysis" refers to the process of instantly analyzing the content of a comment the moment it is entered and providing the results.

[1252] "Offensiveness" refers to a rating that indicates how offensive or harmful a comment may be to others.

[1253] A "rewritten sentence" refers to a sentence in which an offensive comment has been revised to use more neutral or constructive language.

[1254] "Gamification elements" refer to game-like elements that assign scores and ranks to user behavior to increase user motivation.

[1255] "Emotional state" refers to the state of a user's emotions when they are entering a comment, and includes anger, joy, sadness, etc.

[1256] "Content distribution service" refers to an online service that provides users with various content such as video, audio, and text, and allows them to share and comment on it.

[1257] "Server" refers to the computer system that analyzes comments, calculates offensiveness scores, generates rewritten text, recognizes the user's emotional state, and stores and manages data.

[1258] As an embodiment of the present invention, a specific description of the system is given below. This system is composed of the main components of a user terminal, a server, and a ranking system.

[1259] User Device

[1260] User devices are expected to include smartphones, tablets, personal computers, etc. Users enter comments into the comment section of the content distribution service through a comment form. While users are entering their comments, a system is in place to temporarily save them in real time.

[1261] server

[1262] The server analyzes comments, calculates offensiveness scores, generates rewritten sentences, and recognizes the user's emotional state. Specifically, it uses the following software and hardware:

[1263] software:

[1264] Analysis model: We use the Hugging Face text classification model to evaluate the offensiveness of comments.

[1265] Emotion Engine: Recognizes the user's emotional state using the Hugging Face emotion recognition model.

[1266] Hardware:

[1267] High-performance servers: Cloud-based high-performance servers (e.g., Amazon Web Services, Google Cloud Platform) for rapid data processing.

[1268] The server then sends the analyzed comments back to the user's device along with an offensiveness score and a rewritten version of the comment, along with the user's emotional state.

[1269] Ranking System

[1270] The ranking system provides a gamification element by ranking users based on their average score, which is based on a database managed by the server. By posting less offensive comments, users can improve their scores and move up the rankings.

[1271] Specific examples

[1272] For example, if a user types a comment like "This news is terrible. Are all the people involved stupid?", the following process will occur:

[1273] 1. Real-time analysis:

[1274] The comment is immediately sent to a server, where an analysis model calculates an offensiveness score.

[1275] The emotion engine recognizes the user's emotional state as "anger."

[1276] 2. Generating and presenting rewritten sentences:

[1277] Along with the offensiveness score, a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better."

[1278] It is sent back to the user's device in real time.

[1279] 3. User Choice:

[1280] Users can choose whether to adopt the rewrite proposal.

[1281] If adopted, the rewritten comment is resubmitted to the server as the final comment.

[1282] 4. Ranking System Update:

[1283] The server saves the final comments in the database and updates the user's average score.

[1284] The rankings are updated and feedback is provided to users.

[1285] Prompt Sentence Examples

[1286] "Please analyze comments entered by users in real time and calculate an offensiveness score. For example, let's analyze the following comment.

[1287] Comment: "This news is terrible. Aren't all the people involved stupid?"

[1288] The above embodiment makes it possible to realize a mechanism that suppresses offensive comments in the comment section and provides a healthy dialogue environment.

[1289] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1290] Step 1:

[1291] The user enters a comment. The user uses a smartphone or personal computer to enter text into the comment field of the content distribution service. This comment is temporarily saved on the user's device. Input: Text data of the comment. Output: Text data temporarily saved on the device.

[1292] Step 2:

[1293] Prepare to send the comment to the server. The user's device generates a request to send the entered comment to the server in real time. This request includes the user ID and comment. Input: User ID and comment text data. Output: Request data to be sent to the server.

[1294] Step 3:

[1295] The server receives the comment. The server receives a request from the user's device and obtains the comment text data and user ID. Input: Request data. Output: Comment text data and user ID.

[1296] Step 4:

[1297] Analyze the comments. The server uses an analysis model (Hugging Face's text classification model) to calculate the comment's aggressiveness score. It also uses an emotion engine (Hugging Face's emotion recognition model) to recognize the user's emotional state. Input: Comment text data. Output: Aggression score and emotional state.

[1298] Step 5:

[1299] Generate a rewritten sentence. The server generates a rewritten sentence that changes the comment to a more neutral or constructive expression based on the offensiveness score. Input: Comment text data and offensiveness score. Output: Rewritten sentence.

[1300] Step 6:

[1301] The analysis results are returned to the user device. The server creates a response including the offensiveness score, rewrite suggestion, and emotional state, and returns it to the user device. Input: offensiveness score, rewrite sentence, emotional state. Output: response data.

[1302] Step 7:

[1303] The analysis results are presented. The user device receives the response from the server and displays the offensiveness score, rewrite suggestions, and emotional state to the user. Input: Response data. Output: Display data on the device.

[1304] Step 8:

[1305] The user makes a choice. The user chooses whether to adopt the proposed rewrite. If the user adopts the rewrite, the content is saved as the final comment. Input: User's choice. Output: Text data of the final comment.

[1306] Step 9:

[1307] Send the final comment to the server. The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. Input: User ID and text data of the final comment. Output: Request data to be sent to the server.

[1308] Step 10:

[1309] Update the ranking system. The server saves the final comment to the database and updates the user's average score. Based on that score, the user's rank is updated and the latest ranking is generated. Input: Text data of the final comment. Output: Updated ranking data.

[1310] Through the above steps, a system is realized that analyzes users' comments and encourages them to post less offensive comments.

[1311] 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.

[1312] 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.

[1313] 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.

[1314] [Fourth embodiment]

[1315] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1316] 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.

[1317] 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).

[1318] 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.

[1319] 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.

[1320] 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).

[1321] 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.

[1322] 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.

[1323] 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.

[1324] 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.

[1325] 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.

[1326] 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.

[1327] 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."

[1328] The embodiment of the present invention provides a system that analyzes in real time how the content of a comment entered by a user will be perceived by others and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average points earned by users.

[1329] System Overview

[1330] The system consists of the following main components:

[1331] User terminal: A device used to input comments and interact with the system.

[1332] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, and manages rankings.

[1333] Ranking system: Ranking based on users' average scores, providing a gamification element.

[1334] Program processing flow

[1335] User comment input

[1336] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[1337] Sending a comment analysis request

[1338] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[1339] AI-powered comment analysis

[1340] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better."

[1341] Returning analysis results

[1342] The server creates a response containing the offensiveness score and the rewritten text and sends it back to the user's device.

[1343] Presenting and selecting rewrite proposals

[1344] The user device receives the response from the server and presents the offensiveness score and rewritten sentence to the user. The user can choose whether to adopt the proposed rewritten sentence. For example, if the user adopts the rewritten sentence, the content is saved as the final comment on the user device.

[1345] Submitting final comments

[1346] The user device sends the final comment back to the server, and this request includes the user ID and the final comment.

[1347] Ranking System Update

[1348] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[1349] Specific examples

[1350] For example, consider a case where a user enters a comment such as, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI. It is determined to have a high offensive score, and a rewrite proposal is generated. The rewrite proposal, "There are many problems with this news. I hope that the people involved will improve their response," is presented to the user. The user adopts the rewrite proposal and sends the final comment to the server. The server saves this comment in a database and updates the ranking.

[1351] This process will help curb offensive posts in the comments section and provide a healthy interactive environment.

[1352] The processing flow will be explained below.

[1353] Specific processing steps of the program

[1354] Step 1:

[1355] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[1356] Terminal: Obtains and temporarily stores comments entered by users.

[1357] Step 2:

[1358] On your device: Create a request to analyze the saved comments. This request includes the user ID and the comment.

[1359] Terminal: Sends the created request to the server.

[1360] Step 3:

[1361] Server: Receives requests from user devices.

[1362] Server: The received comments are passed to the AI ​​analysis model for analysis, where the offensiveness score of the comment is calculated and a rewrite proposal is generated.

[1363] Step 4:

[1364] Server: Create a response containing an offensiveness score and a rewrite suggestion as the analysis result. For example, if the offensiveness score is 80 / 100, the rewrite suggestion is "There are many problems with this news. We hope that the relevant parties will respond better."

[1365] Server: Returns the created response to the user device.

[1366] Step 5:

[1367] Terminal: Receives the response sent back from the server and displays the analysis results to the user.

[1368] Device: Presents the offensiveness score and rewrite suggestions, prompting the user to choose whether or not to adopt the rewrite suggestions.

[1369] User: Checks the rewrite proposal and decides whether to adopt it. For example, adopt the rewrite proposal "There are many problems with this news. I hope that the people involved will respond better."

[1370] Step 6:

[1371] On the device: The user creates a request to send the final comment (in this case, the rewrite proposal) back to the server. This request includes the user ID and the final comment.

[1372] Terminal: Send a request to the server containing the final comment.

[1373] Step 7:

[1374] Server: Receives the final comment from the user terminal.

[1375] Server: Stores received comments in a database.

[1376] Server: Updates the average score of users based on the offensiveness score of their comments and processes the ranking.

[1377] Server: Returns a response to the user device indicating that the ranking update has been completed.

[1378] For example, if a user enters an offensive comment such as "This news is terrible. Aren't all the people involved stupid?", the system will analyze the comment using AI and assign a high offensiveness score (80 / 100). At the same time, the system will generate a rewrite suggestion, "There are many problems with this news. I hope that the people involved will improve their response," and present it to the user. If the user adopts this rewrite suggestion, it will be sent to the server as the final comment and saved in the database. This will also update the user's average score, which will be reflected in the rankings. In this way, users are encouraged to voluntarily post less offensive comments.

[1379] Example 1

[1380] 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."

[1381] In recent years, the number of offensive comments has increased in the comment sections of online platforms, hindering a healthy dialogue environment. There is a need for a system that can detect such offensive comments in real time and allow users to correct them themselves. Gamification elements are also necessary to increase user motivation.

[1382] 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.

[1383] In this invention, the server includes: means for analyzing user comments in real time and scoring the degree of offensiveness; means for proposing improved rewrites for offensive comments; means for providing a gamification element that ranks comments based on the average points earned by users; a terminal for inputting and submitting comments and presenting rewrites; a server equipped with an analytical model for generating offensiveness scores and rewrite suggestions; and a system for displaying rankings based on user comment scores. This makes it possible to provide a healthy interactive environment and increase user motivation by detecting offensive comments in real time and presenting improvement suggestions.

[1384] "User" refers to the end user who enters comments and interacts with the system.

[1385] "Comment" means a text message entered by a User on the Online Platform.

[1386] "Real-time" refers to the fact that user actions and the system's response to them are immediate.

[1387] "Offensiveness" refers to the degree to which the content of a comment is offensive or unpleasant to others.

[1388] "Score" refers to a score used to quantitatively evaluate the offensiveness of a comment.

[1389] "Rewrites" are text messages that convert offensive comments into more gentle and wholesome language.

[1390] "Suggest" means presenting the user with a rewritten statement and giving them the option to correct the original comment.

[1391] "Gamification" refers to the incorporation of game elements to increase user motivation.

[1392] "Average score" refers to the average aggression score a user has earned in the past.

[1393] "Ranking" refers to ranking based on the average score achieved by users.

[1394] "Terminal" refers to the device (smartphone, PC, etc.) through which a user enters comments and interacts with the system.

[1395] "Server" refers to the central computer system used to analyze comments, assign offensiveness scores, generate rewrites, and manage rankings.

[1396] "Analysis model" refers to natural language processing technology used to analyze the content of comments and generate offensiveness scores and rewrite suggestions.

[1397] "Ranking System" refers to a system that displays rankings based on users' comment scores and provides gamification elements.

[1398] A "natural language processing model" refers to an analysis algorithm that uses technology to understand and analyze human language.

[1399] This invention provides a system that analyzes in real time how a user's comment content will be perceived by others when the user enters it, and scores the degree of offensiveness. It also includes a gamification element that suggests improved rewrites for offensive comments and ranks them based on the average score earned by the user. Specifically, it consists of the following main components:

[1400] User Device

[1401] A user terminal is a device through which users can enter comments on news articles and interact with the system. For example, a smartphone or PC is an example. Users can enter comments through the terminal and check the analysis results in real time.

[1402] server

[1403] The server is a central computer system that receives and analyzes comments submitted by users. The server is equipped with OpenAI's GPT model as a natural language processing (NLP) model, which analyzes comments, assigns offensiveness scores, and generates rewritten sentences.

[1404] Ranking System

[1405] The ranking system displays a ranking based on the average offensiveness score of users and provides a gamification element. The system stores users' comment history and offensiveness scores in a database and updates the user's rank based on the results.

[1406] Explanation of program processing

[1407] When the server receives a user's comment, it analyzes it using an NLP model (e.g., OpenAI's GPT-3). This analysis calculates an offensiveness score for the comment. For offensive comments, it generates a rewritten text with milder language. For example, if a comment such as "This news is terrible. Aren't all the people involved stupid?" is entered, it is determined to have a high offensive score, and a rewrite suggestion such as "There are many problems with this news. I hope that the people involved will respond better." is generated.

[1408] The generated offensiveness score and rewrite proposal are sent back from the server to the user's device. The user can review them on their device and choose whether or not to adopt the proposed rewrite. If the user adopts the rewrite, it is saved on the device as a final comment and sent back to the server. The server saves this final comment in a database and updates the user's average offensiveness score. An updated ranking is generated based on this score and sent back to the user's device.

[1409] Specific examples

[1410] Suppose a user enters a comment such as "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by the AI ​​model. As a result of the analysis, the offensiveness score is determined to be 80 / 100, and a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better." If the user adopts the rewrite suggestion and sends the final comment to the server, the server saves this comment in the database and updates the ranking.

[1411] Prompt Sentence Examples

[1412] "Analyze how an input comment would be perceived by others, score the level of offensiveness, and suggest a milder rewrite that expresses the same meaning. Example: Comment: 'This news is terrible. Are all the people involved stupid?'"

[1413] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1414] Step 1: User comments

[1415] A user uses a device to enter text into a comment section of an online platform, for example, "This news is terrible. Are all the people involved stupid?" This comment is temporarily stored in the device's memory.

[1416] Input: Comments entered by the user

[1417] Output: Saved comments

[1418] Specific operation: The user enters a comment using a keyboard or touch screen and presses the send button.

[1419] Step 2: Submitting a comment analysis request

[1420] The device creates an HTTP request to send an analysis request including the entered comment and user ID to the server. The request includes the user ID and comment content in JSON format. For example, it is sent in the following format: {"user_id": "12345", "comment": "This news is terrible. Aren't all the people involved stupid?"}

[1421] Input: Temporarily saved comment, user ID

[1422] Output: Parse request to server

[1423] Specific operation: The device creates an HTTP request and sends it to the server over the Internet.

[1424] Step 3: Comment analysis by the server

[1425] The server receives the request and analyzes the comment. Using a natural language processing (NLP) model, such as OpenAI's GPT-3 model, the server calculates the comment's offensiveness score and generates a rewrite. If the offensiveness score is 80 / 100, the rewrite reads, "There are many problems with this news. We hope that the relevant parties will improve their response."

[1426] Input: Analysis request (user ID, comment content)

[1427] Output: Analysis results (aggression score, rewritten sentence)

[1428] Specific operation: The server executes the NLP model to analyze the comments and generate rewritten sentences.

[1429] Step 4: Returning the analysis results

[1430] The server creates a response containing the generated aggressiveness score and the rewritten comment, and sends it back to the user's device. The response is also in JSON format, for example, {"aggressiveness_score": 80, "rewritten_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[1431] Input: Analysis results (aggression score, rewritten text)

[1432] Output: Response to user device

[1433] Specific operation: The server creates an HTTP response and sends it to the terminal via the Internet.

[1434] Step 5: Presenting and selecting rewrite proposals

[1435] The user's device receives the response from the server and displays the offensiveness score and the rewritten text to the user. The user can choose to accept the rewritten comment or post the original comment as is. For example, if the user accepts the rewritten comment, the rewritten comment is saved on the device as the final comment. If the user rejects the rewritten comment, the original comment is saved as the final comment.

[1436] Input: Response from the server (aggression score, rewritten text)

[1437] Output: Final comment

[1438] Specific operation: The user's device displays rewrite suggestions, and the user presses the selection button.

[1439] Step 6: Submitting final comments

[1440] The device then sends the final comment and user ID to the server again in JSON format, for example, {"user_id": "12345", "final_comment": "There are many problems with this news. I hope that the people involved will improve their response."}

[1441] Input: Last comment, User ID

[1442] Output: Sending the final comment to the server

[1443] Specific operation: The terminal creates an HTTP request including the final comment and sends it to the server.

[1444] Step 7: Update the ranking system

[1445] The server saves the final comment in the database and updates the user's average offensiveness score. This recalculates the user's rank and generates an updated ranking. The server then sends a response back to the user device indicating that the update is complete. For example, the response will be in the format {"status": "success", "new_rank": 5}.

[1446] Input: Final comment

[1447] Output: Updated rank

[1448] Specific operation: The server updates the database, calculates the user's new rank, generates a response and sends it to the device.

[1449] (Application example 1)

[1450] 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."

[1451] In the comment sections of content distribution services, many users often post offensive comments, which can undermine the health of the dialogue. There is a need for an appropriate system to promote communication between users and provide a healthy dialogue environment. However, current technology lacks a mechanism for analyzing comments in real time, scoring their offensiveness, and suggesting rewrites. Furthermore, there is a lack of a system that can promote healthy comment posting by scoring and ranking users' commenting behavior.

[1452] 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.

[1453] In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for offensive comments, and means for updating user scores and rankings. This makes it possible to analyze the aggression of comments in real time and provide a healthy dialogue environment. Furthermore, by suggesting appropriate rewrites to users, the number of offensive comments can be reduced. Furthermore, gamification elements can be used to encourage users to post healthy comments, and a ranking system can increase user participation.

[1454] The "real-time analysis means" is a function that analyzes user comments as they are entered and instantly scores the degree of aggressiveness based on their content.

[1455] The "rewrite suggestion method" is a function that, when an offensive comment is entered, presents the user with rewrite suggestions to revise the content to make it more appropriate and constructive.

[1456] The "gamification element providing means" is a function that introduces game-like elements into user behavior, scores comment posts, and ranks them based on the average score obtained.

[1457] "Smart device display means" is a function that displays the comment's offensiveness score and rewritten text on the screen of the device used by the user, such as a smartphone or tablet.

[1458] "Scoring and ranking update means" is a function that calculates an aggressiveness score based on comments posted by users, records the score in a database, and updates the user's rating and ranking.

[1459] The "server" is a computer system that receives comments sent by users, analyzes them, generates an offensiveness score and a rewritten sentence, and stores the final comments in a database.

[1460] System Overview

[1461] The system analyzes the content of comments entered by users in real time, scores their offensiveness, and suggests rewrites. It also includes a gamification element that scores and ranks users' behavior.

[1462] Hardware Configuration

[1463] The system consists of the following major hardware components:

[1464] 1. User Device: An end-user device, including a smartphone or tablet, that allows for the input of comments and the display of suggested rewrites.

[1465] 2. Server: A central computer system that processes analysis and manages data.

[1466] Software Configuration

[1467] The system consists of the following major software components:

[1468] 1. Comment analysis module: Generative AI model using Python and TensorFlow.

[1469] 2. Rewrite generation module: Processes text data and generates appropriate rewrites for offensive comments.

[1470] 3. Database: Stores user comments, scores, and ranks (e.g. MySQL).

[1471] 4. User interface: An application that runs on a smart device and allows users to enter comments, view analysis results, and propose and adopt rewrite proposals.

[1472] Program processing flow

[1473] User side:

[1474] 1. Comment input: The user enters text into the comment input field of the smartphone app.

[1475] 2. Send: The entered comment is sent to the analysis server along with the user ID as an HTTP POST request.

[1476] Server side:

[1477] 3. Real-time analysis: The server receives the request and analyzes the comment using a generative AI model using Python and TensorFlow. The analysis results in an offensiveness score and generates a rewritten sentence.

[1478] Example: If the comment "This news is terrible. Aren't all the people involved stupid?" is entered, the offensiveness score will be evaluated as 80 / 100, and a rewrite sentence will be generated that reads "There are many problems with this news. I hope that the people involved will respond better."

[1479] 4. Returning the results: The analysis results (offensiveness score and rewritten sentence) are returned from the server to the user's device.

[1480] User side:

[1481] 5. Display and selection: The smart device receives the analysis results sent from the server and presents them to the user. The user can then choose whether to adopt the rewritten sentence.

[1482] 6. Sending the final comment: The final comment that adopts the rewritten sentence is sent to the server again and saved in the database.

[1483] Server side:

[1484] 7. Scoring and Ranking Update: The server updates the average score of the user based on the last comment and recalculates the ranking. The latest ranking information is sent back to the user's device.

[1485] Examples of specific examples and prompts

[1486] As a concrete example, consider the case where a user writes a review of a TV drama. If the user writes, "This drama is not interesting at all. All the actors' acting is terrible," this comment is sent to the analysis server. The generative AI model analyzes the comment on the server side, assigns an offensiveness score of 70 / 100, and suggests a rewrite such as, "This drama did not meet my expectations. I look forward to seeing the actors grow."

[1487] Example prompt sentence:

[1488] If a comment is typed: "This news is terrible. Are all the people involved stupid?"

[1489] Input the text "This news is terrible. Are all the people involved stupid?" into the analytical AI model and generate an offensiveness score and an improved rewrite.

[1490] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1491] Program processing flow

[1492] Step 1:

[1493] Users enter text into the comment input field of the smartphone app, which is temporarily stored on the user's device.

[1494] input:

[1495] The comment text entered by the user.

[1496] output:

[1497] The comment text is temporarily saved on the user's device.

[1498] Step 2:

[1499] The user device sends an HTTP POST request to the server to analyze the comment, which includes the user ID and the comment text.

[1500] input:

[1501] User ID, comment text.

[1502] output:

[1503] An HTTP POST request containing the user ID and comment text is sent to the server.

[1504] Step 3:

[1505] The server receives the request and analyzes the comments using a generative AI model powered by Python and TensorFlow.

[1506] input:

[1507] Request data (user ID, comment text).

[1508] output:

[1509] Aggression scores and rewritten statements.

[1510] Data processing and calculation:

[1511] The server launches a generative AI model that generates an offensiveness score and rewrite sentences using the input comment text as prompts.

[1512] Step 4:

[1513] The server sends the generated offensive score and the rewritten sentence back to the user's device.

[1514] input:

[1515] Aggression score, rewritten sentence.

[1516] output:

[1517] A response containing the offensiveness score and the rewritten sentence is sent to the user's device.

[1518] Step 5:

[1519] The user device receives the offensiveness score and rewritten sentences sent from the server and presents them to the user, who can then review the rewritten sentences and choose whether or not to adopt them.

[1520] input:

[1521] Response from the server (aggression score, rewritten text).

[1522] output:

[1523] The offensiveness score and rewritten sentence presented to the user.

[1524] Specific behavior:

[1525] The user's device displays the received offensiveness score and rewritten text on the screen, and the user confirms them.

[1526] Step 6:

[1527] If the user accepts the rewritten comment, it is sent to the server again as a final comment. This request includes the user ID and the final comment.

[1528] input:

[1529] User ID, last comment (rewritten text).

[1530] output:

[1531] An HTTP POST request containing the user ID and the last comment is sent to the server.

[1532] Step 7:

[1533] The server receives the final comments and stores them in the database, while calculating the average score of the users and updating the ranking.

[1534] input:

[1535] Final comment.

[1536] output:

[1537] Updated user ranking information.

[1538] Data processing and calculation:

[1539] The server stores the final comment received in a database, compares it with the user's past comment scores, calculates an average score, and updates the user's ranking based on the result to generate the latest ranking information.

[1540] 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.

[1541] The present invention provides a system that analyzes in real time how a user's comments are perceived by others and scores the degree of offensiveness when the user enters a comment. The system also combines a gamification element that suggests improved rewrites for offensive comments and ranks them based on the user's average score, with an emotion engine that recognizes the user's emotions.

[1542] System Overview

[1543] The system consists of the following main components:

[1544] User terminal: A device used to input comments and interact with the system.

[1545] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[1546] Ranking system: Ranking based on users' average scores, providing a gamification element.

[1547] Program processing flow

[1548] User comment input

[1549] A user types text into a comment section of a news article, for example, "This news is terrible. Are all the people involved stupid?" This input is temporarily stored by the user's device.

[1550] Sending a comment analysis request

[1551] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[1552] AI-powered comment analysis

[1553] The server receives the request and analyzes the comment. It uses an analysis model to generate an offensiveness score and a rewrite, and uses an emotion engine to recognize the user's emotional state. For example, if the offensiveness score is 80 / 100, the rewrite suggestion would be "There are many problems with this news. I hope that the people involved will respond better." The server also recognizes the user's emotion as anger.

[1554] Returning analysis results

[1555] The server creates a response that includes an offensiveness score, suggested rewrites, and the emotion recognized by the emotion engine.

[1556] Example: Aggression score 80, rewrite suggestion: "There are many problems with this news. I hope that the people involved will respond better.", user emotional state: anger.

[1557] Presenting and selecting rewrite proposals

[1558] The user terminal receives the response sent back from the server and displays the analysis results to the user.

[1559] The user device presents the offensiveness score and the rewrite suggestion, and prompts the user to choose whether to adopt the rewrite suggestion. The user decides whether to adopt the rewrite suggestion. For example, if the user adopts the rewrite suggestion, the content is saved as the final comment on the user device.

[1560] Submitting final comments

[1561] The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. The final comment is sent to the server.

[1562] Ranking System Update

[1563] The server saves the final comment in the database, updates the user's average score, updates the user's rank based on that score, and generates the latest ranking. It then sends a response back to the user's device indicating that the update is complete.

[1564] Specific examples

[1565] For example, imagine a user comments, "This news is terrible. Aren't all the people involved stupid?" This comment is sent to the server and analyzed by AI and an emotion engine. It is determined to have a high offensive score, and a rewrite suggestion is generated. The user's emotion is also recognized as "anger," and a rewrite suggestion is presented based on this result. If the user adopts this rewrite suggestion, it is sent to the server as the final comment and saved in the database. This updates the user's average score, which is also reflected in the rankings. In this way, users are encouraged to voluntarily post less aggressive comments. This process discourages aggressive comments in the comment section, providing a healthy dialogue environment.

[1566] The processing flow will be explained below.

[1567] Specific processing steps of the program

[1568] Step 1:

[1569] User: In the comments section of a news article page, they type a comment like, "This news is terrible. Are all the people involved stupid?"

[1570] Terminal: Obtains and temporarily stores comments entered by users.

[1571] Step 2:

[1572] Device: Create a request to analyze the saved comments. This request contains the user ID, the comment, and the user's device session information.

[1573] Terminal: Sends the created request to the server.

[1574] Step 3:

[1575] Server: Receives requests from user devices.

[1576] Server: The received comments are passed to an AI analysis model for analysis, where an offensiveness score is calculated and a rewrite proposal is generated.

[1577] Step 4:

[1578] Server: Based on the analysis of the comments, an emotion engine is used to identify the user's emotional state. For example, the user's comment is recognized as being based on anger.

[1579] Server: Adjusts rewrite suggestions based on the user's emotional state. In the future, especially in urgent cases, the server will determine the appropriate timing to present rewrite suggestions to the user according to the emotion recognition results.

[1580] Step 5:

[1581] Server: Create a response that includes the offensiveness score, emotional state, and rewrite suggestion. For example, the offensiveness score is 80 / 100, and the rewrite suggestion is "There are many problems with this news. I hope that the people involved will respond better."

[1582] Server: Returns the created response to the user device.

[1583] Step 6:

[1584] Terminal: Receives the response sent back from the server and displays the analysis results (aggression score, emotional state, rewrite suggestions) to the user.

[1585] Device: Ask the user whether they would like to adopt the proposed rewrite: "There are many problems with this news. We hope that those involved will respond better."

[1586] User: Review the rewrite suggestion and decide whether to adopt it. For example, if the user adopts the rewrite suggestion.

[1587] Step 7:

[1588] Device: Create a request to send the user's final comment (in this case, a rewrite proposal) back to the server. This request includes the user ID, the final comment, and the emotional state.

[1589] Terminal: Send a request to the server containing the final comment.

[1590] Step 8:

[1591] Server: Receives the final comment from the user terminal.

[1592] Server: Stores received comments in a database, including the comment, its emotional state, and its aggression score.

[1593] Step 9:

[1594] Server: Update the average score of the user based on the offensiveness score of the last comment. Update the rank based on the average score of the user.

[1595] Server: Returns a response to the user device indicating that the rank update has been completed.

[1596] Specific examples

[1597] For example, if a user comments, "This news is terrible. Aren't all the people involved stupid?", the system sends the comment to the server. The server uses an AI model to calculate an aggression score and simultaneously recognizes the user's emotional state using an emotion engine. As a result, the aggression score is determined to be 80 / 100, and the user's emotion is recognized as "anger." The server then generates a rewrite suggestion, "There are many problems with this news. I hope the people involved will improve their response," and presents it to the user. If the user adopts the rewrite suggestion, the comment is sent to the server and stored in the database. The user's average aggression score is also updated, along with their rank. This process encourages users to voluntarily refrain from making aggressive comments, creating a healthy dialogue environment.

[1598] Example 2

[1599] 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."

[1600] There are many offensive comments in online comment sections and message boards, which is a problem as it damages a healthy dialogue environment. There are also cases where users post offensive comments without realizing it, which calls for improvement. Furthermore, there is a lack of ways for users to learn what words are offensive and to communicate better.

[1601] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a comment, a means for transmitting the comment to the server, a means for the server to analyze the comment in real time and score the degree of aggression, a means for generating a rewritten sentence based on the aggression score, a means for presenting the rewritten sentence to the user and prompting the user to select, a means for providing a gamification element that ranks the rewritten sentences based on the average score obtained by the users, and a means for storing the comments and the analysis results in a database. This encourages users to voluntarily post less aggressive comments, and provides a healthy dialogue environment and enables users to learn.

[1602] "User" refers to the entity that accesses the system and enters comments.

[1603] A "comment" is text information entered by a user that accompanies content such as a news article or message board.

[1604] "Server" refers to the central processing unit that receives and analyzes comments.

[1605] "Real-time" refers to a situation where there is almost no delay between the time a comment is entered and the time the analysis results are returned.

[1606] "Level of aggression" refers to a numerical score that indicates how aggressive the content of a comment is.

[1607] A "rewrite" refers to more appropriately worded text that is generated to make the original comment less offensive.

[1608] "Analysis results" refers to information such as scores, rewritten sentences, and emotional state obtained after the server analyzes the comments.

[1609] "Gamification elements" are elements that make user behavior fun and motivating, like a game, and specifically include ranking and reward systems.

[1610] "Database" refers to a collection of information for storing comments and analysis results.

[1611] An "emotion engine" refers to a mechanism that analyzes and detects emotional states (such as anger or joy) from user comments.

[1612] The embodiment of the present invention is a system that, when a user inputs a comment, analyzes in real time how the content of the comment will be received by others and scores the degree of offensiveness. A detailed description of this system is provided below.

[1613] System Overview

[1614] The system consists of the following main components:

[1615] User terminal: A device used to input comments and interact with the system. User terminals include personal computers and smartphones.

[1616] Server: Analyzes comments, assigns offensiveness scores, generates rewritten sentences, has an emotion engine that recognizes user emotions, and manages rankings.

[1617] Ranking system: Ranking based on users' average scores, providing a gamification element.

[1618] Hardware and software used

[1619] User device: A device that can connect to the Internet (e.g., smartphone, PC)

[1620] Server: A high-performance computer using cloud services

[1621] Software: generative AI models (e.g., open-source natural language processing models), database management systems, emotion engines

[1622] Data processing and calculation

[1623] 1. Comment input: The user inputs a comment, and the user terminal sends this data to the server.

[1624] 2. Data analysis: The server receives the comments and uses a generative AI model to score the degree of offensiveness, and an emotion engine to analyze the user's emotional state.

[1625] 3. Generating rewritten sentences: The generative AI model generates rewritten sentences for comments that are judged to be highly offensive.

[1626] 4. Presentation of results: The analysis results and rewritten sentences are returned from the server and displayed on the user's terminal.

[1627] 5. Accepting the comment: The user selects whether to accept the rewritten sentence and sends the final comment to the server.

[1628] 6. Update ranking: The server saves the final comments to the database and updates the ranking based on the average score of the users.

[1629] Specific examples

[1630] For example, if a user types a comment like "This news is terrible. Aren't all the people involved stupid?", this comment is sent to the server, which then inputs the following prompt sentence into the generative AI model for analysis:

[1631] Example prompt sentence:

[1632] Comment Analysis:

[1633] Typed comment: "This news is terrible. Are all the people involved stupid?"

[1634] Aggression Score: 80

[1635] Suggested rewrite: "There are many problems with this news. I hope that those involved will improve their response."

[1636] Emotional state: Anger

[1637] After the analysis is complete, the server sends back the results, including the offensiveness score, rewrite suggestions, and emotional state, to the user's device. The user's device displays these results on the screen and prompts the user to decide whether to accept the rewrite suggestions. If the user accepts the rewrite suggestions, the rewritten comment is resubmitted to the server as the final comment. Finally, the final comment is saved in the database, and the ranking is updated based on the user's average score.

[1638] Through this process, users are encouraged to voluntarily post less offensive comments, creating a healthier interactive environment.

[1639] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1640] Step 1: Enter a comment

[1641] A user types a comment into the comment section of a news article or message board.

[1642] What happens: A user types "This news is terrible. Are all the people involved stupid?" This comment is stored as temporary data on the user's device.

[1643] Input: User comment

[1644] Output: Input comments as temporary data

[1645] Step 2: Send a comment analysis request

[1646] The user terminal creates a request to send the entered comment to the server. This request includes the user ID and the comment.

[1647] Specific operation: The user's device packages the user ID and comments in JSON format and sends an HTTP request to the server's API endpoint.

[1648] Input: User ID, Comment

[1649] Output: HTTP request to the server

[1650] Step 3: Receiving and analyzing comments

[1651] The server receives the request and analyzes the comment using a generative AI model, which generates an offensiveness score and a rewrite, and also uses an emotion engine to recognize the user's emotional state.

[1652] Specific operation: The server analyzes the request and creates a prompt sentence to input the comment part into the AI ​​model. The AI ​​model analyzes it and generates an offensiveness score, rewrite sentence, and emotional state.

[1653] Input: User ID, Comment

[1654] Output: Analysis results including aggression score, rewritten sentences, and emotional state

[1655] Step 4: Returning the analysis results

[1656] The server creates a response containing the offensiveness score, suggested rewrites, and the perceived emotion, and sends it back to the user device.

[1657] Specific operation: The server packages the analysis results in JSON format and sends them to the user device as an HTTP response.

[1658] Input: Aggression score, rewritten sentence, emotional state

[1659] Output: HTTP response to the user's device

[1660] Step 5: Presenting and selecting rewrite proposals

[1661] The user device receives the response and displays the analysis results to the user, who can then decide whether to adopt the proposed rewrite.

[1662] Specific operation: The user's device parses the response, displays it on the screen, and presents rewrite suggestions. The user can choose whether to adopt the rewrite suggestions.

[1663] Input: Analysis results (aggression score, rewritten sentence, emotional state)

[1664] Output: User's choice (whether to accept the rewrite proposal or not)

[1665] Step 6: Submitting final comments

[1666] The user device creates a request to resend the last comment to the server, including the user ID and the last comment.

[1667] Specific operation: The user's device packages the user ID and the last comment in JSON format and sends an HTTP request to the server's API endpoint.

[1668] Input: User ID, last comment

[1669] Output: HTTP request to the server

[1670] Step 7: Update the ranking system

[1671] The server saves the final comments to the database and updates the user's average score, which in turn updates the user's rank to generate the latest ranking.

[1672] Specific operation: The server saves the final comment in the database, calculates a new average score based on the user's previous comment scores, updates the ranking database, and sends an update completion message to the user's device.

[1673] Input: Last comment, user's past score

[1674] Output: Updated ranking, update complete message

[1675] (Application example 2)

[1676] 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."

[1677] Current content distribution services often have users posting offensive comments, making it difficult to maintain a healthy dialogue environment. Furthermore, because users often post offensive comments without being aware of their own feelings, a mechanism to curb this is needed. Furthermore, there is a lack of gamification elements to encourage users to voluntarily adopt positive behavior.

[1678] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing user comments in real time and scoring the degree of aggression, means for suggesting improved rewrites for aggressive comments, means for providing a gamification element that ranks comments based on the average points earned by users, means for recognizing users' emotional states in real time and providing that information, and means for applying it to the comment section of the content distribution service. This enables users to voluntarily post less aggressive comments and maintain a healthy dialogue environment.

[1679] "User" refers to a user who posts comments within the content distribution service.

[1680] "Comment" refers to a text message that a user enters within the Content Distribution Service and shares with others.

[1681] "Real-time analysis" refers to the process of instantly analyzing the content of a comment the moment it is entered and providing the results.

[1682] "Offensiveness" refers to a rating that indicates how offensive or harmful a comment may be to others.

[1683] A "rewritten sentence" refers to a sentence in which an offensive comment has been revised to use more neutral or constructive language.

[1684] "Gamification elements" refer to game-like elements that assign scores and ranks to user behavior to increase user motivation.

[1685] "Emotional state" refers to the state of a user's emotions when they are entering a comment, and includes anger, joy, sadness, etc.

[1686] "Content distribution service" refers to an online service that provides users with various content such as video, audio, and text, and allows them to share and comment on it.

[1687] "Server" refers to the computer system that analyzes comments, calculates offensiveness scores, generates rewritten text, recognizes the user's emotional state, and stores and manages data.

[1688] As an embodiment of the present invention, a specific description of the system is given below. This system is composed of the main components of a user terminal, a server, and a ranking system.

[1689] User Device

[1690] User devices are expected to include smartphones, tablets, personal computers, etc. Users enter comments into the comment section of the content distribution service through a comment form. While users are entering their comments, a system is in place to temporarily save them in real time.

[1691] server

[1692] The server analyzes comments, calculates offensiveness scores, generates rewritten sentences, and recognizes the user's emotional state. Specifically, it uses the following software and hardware:

[1693] software:

[1694] Analysis model: We use the Hugging Face text classification model to evaluate the offensiveness of comments.

[1695] Emotion Engine: Recognizes the user's emotional state using the Hugging Face emotion recognition model.

[1696] Hardware:

[1697] High-performance servers: Cloud-based high-performance servers (e.g., Amazon Web Services, Google Cloud Platform) for rapid data processing.

[1698] The server then sends the analyzed comments back to the user's device along with an offensiveness score and a rewritten version of the comment, along with the user's emotional state.

[1699] Ranking System

[1700] The ranking system provides a gamification element by ranking users based on their average score, which is based on a database managed by the server. By posting less offensive comments, users can improve their scores and move up the rankings.

[1701] Specific examples

[1702] For example, if a user types a comment like "This news is terrible. Are all the people involved stupid?", the following process will occur:

[1703] 1. Real-time analysis:

[1704] The comment is immediately sent to a server, where an analysis model calculates an offensiveness score.

[1705] The emotion engine recognizes the user's emotional state as "anger."

[1706] 2. Generating and presenting rewritten sentences:

[1707] Along with the offensiveness score, a rewrite suggestion is generated: "There are many problems with this news. I hope that the people involved will respond better."

[1708] It is sent back to the user's device in real time.

[1709] 3. User Choice:

[1710] Users can choose whether to adopt the rewrite proposal.

[1711] If adopted, the rewritten comment is resubmitted to the server as the final comment.

[1712] 4. Ranking System Update:

[1713] The server saves the final comments in the database and updates the user's average score.

[1714] The rankings are updated and feedback is provided to users.

[1715] Prompt Sentence Examples

[1716] "Please analyze comments entered by users in real time and calculate an offensiveness score. For example, let's analyze the following comment.

[1717] Comment: "This news is terrible. Aren't all the people involved stupid?"

[1718] The above embodiment makes it possible to realize a mechanism that suppresses offensive comments in the comment section and provides a healthy dialogue environment.

[1719] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1720] Step 1:

[1721] The user enters a comment. The user uses a smartphone or personal computer to enter text into the comment field of the content distribution service. This comment is temporarily saved on the user's device. Input: Text data of the comment. Output: Text data temporarily saved on the device.

[1722] Step 2:

[1723] Prepare to send the comment to the server. The user's device generates a request to send the entered comment to the server in real time. This request includes the user ID and comment. Input: User ID and comment text data. Output: Request data to be sent to the server.

[1724] Step 3:

[1725] The server receives the comment. The server receives a request from the user's device and obtains the comment text data and user ID. Input: Request data. Output: Comment text data and user ID.

[1726] Step 4:

[1727] Analyze the comments. The server uses an analysis model (Hugging Face's text classification model) to calculate the comment's aggressiveness score. It also uses an emotion engine (Hugging Face's emotion recognition model) to recognize the user's emotional state. Input: Comment text data. Output: Aggression score and emotional state.

[1728] Step 5:

[1729] Generate a rewritten sentence. The server generates a rewritten sentence that changes the comment to a more neutral or constructive expression based on the offensiveness score. Input: Comment text data and offensiveness score. Output: Rewritten sentence.

[1730] Step 6:

[1731] The analysis results are returned to the user device. The server creates a response including the offensiveness score, rewrite suggestion, and emotional state, and returns it to the user device. Input: offensiveness score, rewrite sentence, emotional state. Output: response data.

[1732] Step 7:

[1733] The analysis results are presented. The user device receives the response from the server and displays the offensiveness score, rewrite suggestions, and emotional state to the user. Input: Response data. Output: Display data on the device.

[1734] Step 8:

[1735] The user makes a choice. The user chooses whether to adopt the proposed rewrite. If the user adopts the rewrite, the content is saved as the final comment. Input: User's choice. Output: Text data of the final comment.

[1736] Step 9:

[1737] Send the final comment to the server. The user terminal creates a request to send the final comment to the server again. This request includes the user ID and the final comment. Input: User ID and text data of the final comment. Output: Request data to be sent to the server.

[1738] Step 10:

[1739] Update the ranking system. The server saves the final comment to the database and updates the user's average score. Based on that score, the user's rank is updated and the latest ranking is generated. Input: Text data of the final comment. Output: Updated ranking data.

[1740] Through the above steps, a system is realized that analyzes users' comments and encourages them to post less offensive comments.

[1741] 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.

[1742] 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.

[1743] 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.

[1744] 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.

[1745] 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.

[1746] 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.

[1747] 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).

[1748] 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.

[1749] 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."

[1750] 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.

[1751] 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).

[1752] 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.

[1753] 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.

[1754] 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.

[1755] 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.

[1756] 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.

[1757] 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.

[1758] 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.

[1759] 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.

[1760] 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.

[1761] 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.

[1762] The following is further disclosed regarding the above embodiment.

[1763] (Claim 1)

[1764] A method to analyze user comments in real time and score the degree of aggression,

[1765] A means of suggesting improved rewrites of offensive comments;

[1766] A means for providing a gamification element that ranks users based on their average score;

[1767] A system including:

[1768] (Claim 2)

[1769] 2. The system according to claim 1, wherein comments entered by a user are sent to a server, and the server can obtain analysis results.

[1770] (Claim 3)

[1771] 10. The system of claim 1, wherein the server receives user comments and generates an offensiveness score and a rewrite suggestion.

[1772] "Example 1"

[1773] (Claim 1)

[1774] A method to analyze user comments in real time and score the degree of aggression,

[1775] A means of suggesting improved rewrites of offensive comments;

[1776] A means for providing a gamification element that ranks users based on their average score;

[1777] a terminal for inputting and sending comments and presenting rewritten sentences;

[1778] a server with an analytical model for generating offensiveness scores and rewrite suggestions;

[1779] A system that displays rankings based on user comment scores, and

[1780] A system including:

[1781] (Claim 2)

[1782] 2. The system according to claim 1, wherein the terminal transmits comments entered by the user to the server, and the server obtains the analysis results.

[1783] (Claim 3)

[1784] 10. The system of claim 1, wherein the server receives user comments and uses a natural language processing model to generate offensiveness scores and rewrite suggestions.

[1785] "Application Example 1"

[1786] (Claim 1)

[1787] A method to analyze user comments in real time and score the degree of aggression,

[1788] A means of suggesting improved rewrites of offensive comments;

[1789] A means for providing a gamification element that ranks users based on their average score;

[1790] A means for displaying a comment's offensiveness score and a rewrite on a smart device;

[1791] A means to update user scoring and rankings;

[1792] A system including:

[1793] (Claim 2)

[1794] The system according to claim 1, wherein comments entered by a user are sent to a server and the analysis results can be obtained on a smart device.

[1795] (Claim 3)

[1796] 2. The system of claim 1, wherein the server receives user comments, generates offensiveness scores and rewrite suggestions, and transmits them to the smart device.

[1797] "Example 2: Combining Emotion Engines"

[1798] (Claim 1)

[1799] a means for users to enter comments;

[1800] means for transmitting comments to a server;

[1801] The server analyzes comments in real time and scores the degree of aggression.

[1802] means for generating a rewrite sentence based on the offensiveness score;

[1803] A method for presenting rewritten sentences to users and prompting them to make a selection;

[1804] A means for providing a gamification element that ranks users based on their average score;

[1805] a means for storing comments and analysis results in a database;

[1806] A system including:

[1807] (Claim 2)

[1808] 2. The system according to claim 1, wherein comments entered by a user are sent to a server, and the server can obtain analysis results.

[1809] (Claim 3)

[1810] 10. The system of claim 1, wherein the server receives user comments, generates offensiveness scores and rewrite suggestions, and recognizes the user's emotional state.

[1811] "Application example 2 when combining emotion engines"

[1812] (Claim 1)

[1813] A method to analyze user comments in real time and score the degree of aggression,

[1814] A means of suggesting improved rewrites of offensive comments;

[1815] A means for providing a gamification element that ranks users based on their average score;

[1816] A means of recognizing and providing information about a user's emotional state in real time;

[1817] A means for applying to the comment section of a content distribution service;

[1818] A system including:

[1819] (Claim 2)

[1820] 2. The system according to claim 1, wherein comments entered by a user are sent to a server, and the server can obtain analysis results.

[1821] (Claim 3)

[1822] 10. The system of claim 1, wherein the server receives the user's comments and generates an offensiveness score, a rewrite suggestion, and an emotional state of the user. [Explanation of symbols]

[1823] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A method to analyze user comments in real time and score the degree of aggression, A means of suggesting improved rewrites of offensive comments; A means for providing a gamification element that ranks users based on their average score; A system including:

2. 2. The system according to claim 1, wherein comments entered by a user are transmitted to a server, and an analysis result is obtained from the server.

3. The system of claim 1 , wherein the server receives user comments and generates offensiveness scores and rewrite suggestions.

Images