system

Abstract

The system aims to prevent the publication of offensive information while facilitating more constructive communication among users. [Solution] The system includes means for analyzing information input in real time using natural language processing means to detect offensive elements, means for issuing a warning to the user when offensive elements are detected and suggesting corrections to the information, and means for evaluating whether the information contains legal risks and providing the user with a warning of legal risks as necessary.

Description

Technical Field

[0004] , , , ,

[0005] , , , ,

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] <00000​​​​​​​This invention provides a system that analyzes input information in real time using natural language processing and detects offensive elements. Furthermore, it includes means for issuing a warning to the user and suggesting appropriate corrections when offensive elements are detected. It also includes means for evaluating whether the input information contains legal risks and, if necessary, providing legal warnings to the user, thereby preventing legal problems. This realizes a technology that prevents the publication of offensive information while enabling users to engage in more constructive communication.

[0006] "Natural language processing" refers to technologies that use computers to analyze meaning from text and audio data and process the information.

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

[0008] "Analysis" is the act of breaking down information and processing it in order to understand its structure and meaning.

[0009] "Aggressive elements" refer to words or phrases that may slander others or cause discord.

[0010] A "warning" is a notification that informs the user of a potential problem or risk.

[0011] A "revised version" is an alternative way of expressing information that is proposed to improve problematic information.

[0012] "Legal risk" refers to the possibility that certain actions or statements may cause legal problems.

[0013] A "user" is an individual or group that operates the system and inputs and receives information.

[0014] "Public disclosure" refers to the information entered being widely shared with other users and the general public.

[0015] "Control means" refers to a technical mechanism for managing and adjusting operations and functions within a system.

Brief Explanation of Drawings

[0016] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which multiple emotions are mapped. [Figure 10] It shows an emotion map to which multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Best Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, a numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, a numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the 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.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0030] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] This invention is a system that analyzes input comments and posts to detect offensive elements and legal risks, thereby facilitating appropriate communication on a large-scale information sharing platform. The system comprises user terminals, a server, and their respective processing modules, and operates in real time.

[0038] First, when a user enters a comment on the information sharing platform, the content is sent from the device to the server. The server uses natural language processing technology to analyze the meaning and context of the comment. The purpose of the analysis is to identify offensive elements and evaluate their meaning.

[0039] If the server detects offensive elements in the input text, the system will immediately issue a warning to the user. The warning will inform the user that the entered content is inappropriate and simultaneously offer suggestions for correction. These suggestions will include specific recommendations for constructive and acceptable ways of expressing the content.

[0040] Furthermore, the server uses a legal risk avoidance module to analyze whether a comment could potentially be legally problematic. For example, if it detects expressions that could be considered defamatory or infringe on privacy, it provides the user with a warning and guidelines about the legal implications.

[0041] Feedback from the server is sent to the user's device, allowing the user to modify their comment based on that feedback. Ultimately, only comments that have been cleaned of offensive elements or legal risks are published. This approach improves the overall quality of communication on the platform.

[0042] For example, if a user enters a comment such as "You're a liar," the system will detect the comment and provide feedback such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is." The aim is to guide users to provide constructive feedback rather than slander.

[0043] This system is designed to prevent aggressive communication and provide a more constructive and safer environment for communication. It is hoped that this initiative will promote mutual understanding and respect among users.

[0044] The following describes the processing flow.

[0045] Step 1:

[0046] A user enters text into a comment field on the information sharing platform. Before the user finishes typing, the information is ready to be sent to the server in real time.

[0047] Step 2:

[0048] The terminal creates an API request to send the entered text to the server and sends it to the server. The server receives the entered text.

[0049] Step 3:

[0050] The server uses natural language processing tools to analyze the input text. Here, it identifies patterns of specific words and phrases to determine if they contain offensive elements.

[0051] Step 4:

[0052] If the server detects any offensive elements, it records those elements and their content. Next, it analyzes the context of the entire comment to identify the parts that were determined to have offensive intent.

[0053] Step 5:

[0054] The server generates a warning message for the user. The warning explains that the input is inappropriate and suggests appropriate corrections.

[0055] Step 6:

[0056] The server uses a legal risk avoidance module to assess whether submitted comments pose legal risks. This assessment involves checking for potential defamation and privacy violations.

[0057] Step 7:

[0058] The server will generate warnings about legal risks as needed and create guidelines to communicate these to users.

[0059] Step 8:

[0060] The server sends a warning message, suggested fixes, and legal warnings to the device. This allows the device to provide real-time feedback to the user.

[0061] Step 9:

[0062] Users revise their comments based on feedback from the server. Once revisions are complete, they review the comments again and submit them if necessary.

[0063] Step 10:

[0064] The revised comments are only published after being verified to be free of offensive elements and legal risks. This control prevents inappropriate content from spreading.

[0065] (Example 1)

[0066] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0067] In today's information and communication environment, offensive language and information with legal risks on online platforms cause many problems. Such information can trigger disputes between users and, in some cases, escalate into legal conflicts. Therefore, in order to maintain safe and constructive communication, it is necessary to detect problematic information in real time and deal with it appropriately. However, many current systems simply delete or block problematic information without showing users how to improve the information. Therefore, a mechanism is needed that enables users to acquire better communication skills and promote mutual understanding.

[0068] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0069] In this invention, the server includes means for analyzing information input and detecting offensive elements using natural language processing technology; means for issuing a warning when offensive elements are detected and automatically generating and presenting suggested corrections to the information using generation technology; means for analyzing whether the information contains legal risks and providing warnings and guidelines for legal risks as necessary; and means for making specific suggestions to encourage improvement to the user when offensive elements or legal risks are detected. As a result, users can correct comments in real time, and only information that has had offensive elements and legal risks removed is published, thus promoting safe and constructive communication.

[0070] "Information input" refers to data that users enter using characters and symbols on an online platform.

[0071] "Natural language processing technology" refers to the technology used by computers to understand, interpret, and generate human language.

[0072] "Offensive elements" refer to expressions or content that insult or offend others.

[0073] "Generative technology" refers to technology that uses artificial intelligence to automatically generate suggested modifications or responses to present to users.

[0074] "Legal risk" refers to situations or factors that could potentially violate the law if one speaks or acts in any way.

[0075] A "warning" is a notification that informs a user that certain information may cause problems.

[0076] "Guidelines" are instructional documents that provide specific steps and suggestions for users to appropriately correct information.

[0077] A "concrete proposal" refers to a clear and practical method for users to improve offensive or legally problematic information.

[0078] This system aims to eliminate offensive elements and avoid legal risks in online information sharing platforms. The specific form of this system is described below.

[0079] The user enters a message into a comment input field on the online platform. Once the user has finished typing, their device sends the message to the server. The device sends the data using a standard communication data transmission protocol, such as HTTP. The server has a processing system built on the Python language, and uses NLTK and generative AI models as natural language processing libraries.

[0080] The server uses this library to analyze the received comment data. During the analysis, it determines the emotions associated with the input, whether it contains aggressive elements, and evaluates whether any legal risks exist. Specifically, it uses text mining techniques to tokenize the comments and calculates an evaluation score using a sentiment analysis algorithm.

[0081] If offensive elements or legal risks are detected, generation technology is used to provide the user with a warning and specific corrective actions. For example, if a user says "You're a liar," the server will automatically generate constructive suggestions such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is."

[0082] The following example prompt is used in the generating AI model: "Evaluate the following comment and provide the problems and suggestions for improvement: 'You're a liar.'" This prompt allows the AI ​​to evaluate the inappropriate elements and suggest appropriate countermeasures.

[0083] Ultimately, users revise their comments based on feedback from the server. The revised comments are sent back to the server and are only published on the platform if they are deemed to no longer be offensive or pose any legal risks. This facilitates safe online communication.

[0084] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0085] Step 1:

[0086] Users enter messages into the comment input field on the information sharing platform. After entering the message, the user's device sends this input data (text comment) to the server. Here, data is entered, and the output is the text data sent to the server.

[0087] Step 2:

[0088] The server analyzes the received text data. It takes the received text data as input and performs data analysis using natural language processing libraries (such as Python's NLTK or generative AI models). Specifically, it tokenizes the comments and evaluates the aggression and sentiment scores of each word. The analysis results are generated as output.

[0089] Step 3:

[0090] The server detects offensive elements and legal risks from the analysis results. It uses the analyzed data as input to filter out the underlying problems. Specifically, it identifies problems by comparing them against pre-configured criteria and lists. The output generates a report indicating the presence or absence of problems and the identified risks.

[0091] Step 4:

[0092] The server generates feedback for the user based on detected issues. Using data on offensive elements and legal risks as input, it leverages a generative AI model to generate appropriate corrective actions and warnings. Specifically, it provides more constructive suggestions in response to statements like "You're a liar." The output is feedback that includes corrective actions and warnings.

[0093] Step 5:

[0094] The server sends the generated feedback to the user's device. As input, it retrieves the feedback content and sends the information to the user's device using a data communication protocol. As output, the user sees the feedback from the server.

[0095] Step 6:

[0096] The user revises their comment based on the feedback received. The user edits the comment, using the server's feedback as input. After re-editing the comment, the user's device sends the revised comment to the server. The revised comment data is then sent back to the server as output.

[0097] Step 7:

[0098] The server re-analyzes the corrected comment to confirm that the problem has been resolved. It receives the corrected comment as input and analyzes it again using natural language processing techniques. The output is a final verification result, and if the comment meets the criteria, it is set to be public.

[0099] (Application Example 1)

[0100] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0101] A problem exists in that aggressive remarks are unintentionally made during conversations between customers and store staff in physical stores, leading to a decline in service quality. Furthermore, insufficient responses to remarks that carry legal risks are also a concern. Therefore, there is a need for means to facilitate smooth communication with customers and improve service quality.

[0102] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0103] In this invention, the server includes means for analyzing conversation input using voice signal processing means to detect aggressive elements, means for issuing a warning and suggesting revisions to the conversation when aggressive elements are detected, and means for evaluating whether the conversation content includes legal risks and providing warnings of legal risks as necessary. This enables store employees and customers to have more constructive and safer conversations in physical stores.

[0104] "Audio signal processing" is a technology that analyzes audio data to understand its meaning and context.

[0105] "Conversation input" refers to audio information collected during a conversation, which records the actual exchange.

[0106] "Aggressive elements" refer to words or expressions in a conversation that could unnecessarily offend the other person.

[0107] "Issuing a warning and suggesting a correction" is the process of informing users when offensive elements are detected and suggesting more appropriate wording.

[0108] "Legal risk" refers to a situation where statements or actions have the potential to cause legal problems.

[0109] "Outputting real-time conversation analysis results on the screen" refers to a function that performs analysis immediately as a conversation takes place and displays the results right away.

[0110] The system for implementing this invention provides a function that analyzes the interaction between store staff and customers in real time in a physical store and promotes smooth communication.

[0111] First, a terminal within the store (e.g., a smartwatch) collects conversations between customers and staff using a speech recognition API (e.g., Google® Cloud Speech-to-Text). This audio data is immediately converted into text. Next, a server uses a natural language processing API (e.g., OpenAI® GPT-4®) to analyze the text data in real time. This analysis identifies aggressive elements and legal risk factors within the conversation.

[0112] When aggressive elements are detected, the server immediately sends feedback to the terminal, informing the user of the identified problem and using a generative AI model to suggest appropriate corrections. This allows store employees to respond to customers quickly and appropriately. For example, a phrase deemed aggressive, such as "That's impossible, isn't it?", might be suggested to be modified to a more constructive expression, such as "Let's think together about whether that suggestion is truly feasible."

[0113] A concrete example of a prompt is, "Please revise the following sentence to be more constructive and customer-friendly: 'That's impossible, isn't it?'" As this example shows, the system optimizes user responses to be more customer-centric.

[0114] In this way, by using a generative AI model to analyze conversations through the cooperation of the server and terminal, and by providing appropriate feedback and suggested corrections immediately, the quality of communication in physical stores can be improved.

[0115] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0116] Step 1:

[0117] The device captures conversations between store employees and customers using its microphone. Audio data is collected as input. A speech recognition API is used to convert this audio data into text data. This converted text data forms the basis for the next processing step.

[0118] Step 2:

[0119] The server receives text data. The converted text is used as input. Using a natural language processing API, the server analyzes the text and detects offensive elements. For example, the word "impossible" might be detected as offensive. The output is the analysis result, including the offensive elements and their location information.

[0120] Step 3:

[0121] The server evaluates whether there are legal risks based on the analysis results. The input is the analysis results from step 2. If legal risks exist, it generates appropriate guidelines along with a legal risk warning. The output provides whether or not there are legal risks and details.

[0122] Step 4:

[0123] The server generates and sends feedback to the user regarding offensive elements and legal risks. The input is the information obtained in steps 2 and 3. A generative AI model is used to present flexible and constructive suggestions for correction. The output is the feedback message.

[0124] Step 5:

[0125] The terminal displays feedback received from the server to the store clerk. This allows the clerk to review the suggested corrections and improve the conversation. The output is displayed on the terminal screen and specifically includes suggested corrections and warnings. The clerk then uses this to correct the conversation.

[0126] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0127] This invention is a system that analyzes user comments and posts on an information sharing platform using a combination of natural language processing and an emotion engine to detect offensive elements and legal risks. When a user inputs information using a terminal, that data is transmitted to the server in real time for analysis.

[0128] On the server, the input information is first processed by natural language processing tools to determine if it contains aggressive elements. Simultaneously, an emotion engine estimates the user's emotions from the text and evaluates the emotional tone. Based on this evaluation, if, for example, strong anger or aggression is detected, a warning message is highlighted. Appropriate correction suggestions are also provided, and if necessary, warnings about legal risks are also issued.

[0129] For example, if a user comments, "Your opinion is completely nonsensical!", the server analyzes this text. Natural language processing identifies "nonsensical" as an aggressive word, and the emotion engine detects a high level of anger. The system then suggests to the user, "That expression is offensive. Why not try asking a question, such as, 'Could you explain that in more detail?'" and simultaneously warns of the legal implications.

[0130] The terminal receives feedback from the server and provides this information to the user in a timely manner. The user can refer to this feedback to revise their comment and repost it in an appropriate form, free from aggression and legal issues.

[0131] The goal of this system is to make communication between users safer and more constructive, and to prevent defamation and slander on online platforms. It also aims to improve users' healthy communication skills through immediate feedback and education.

[0132] The following describes the processing flow.

[0133] Step 1:

[0134] A user enters text into the comment field of the information sharing platform. The user's device monitors the input in real time and prepares it for transmission to the server.

[0135] Step 2:

[0136] The terminal sends the entered text to the server. The server receives the transmitted data and starts the analysis process.

[0137] Step 3:

[0138] The server uses a natural language processing module to analyze the received text. The analysis includes a process to identify whether there are any offensive words or expressions.

[0139] Step 4:

[0140] The server uses an emotion engine to estimate the user's emotional state from the input text. In this step, it primarily identifies emotions such as "anger," "joy," and "sadness," and evaluates their intensity.

[0141] Step 5:

[0142] Based on the analysis results, the server generates a warning message for the user if aggressive elements are detected. At the same time, it adjusts the tone of the warning message and suggested fixes based on the sentiment engine's evaluation.

[0143] Step 6:

[0144] The server uses a legal risk assessment module to check whether the input contains legal risks. If legal issues are found, it develops specific legal warnings for the user.

[0145] Step 7:

[0146] The server sends generated warnings, suggested fixes, and legal warnings to the device. The device then displays this feedback on its screen to provide the user with real-time support.

[0147] Step 8:

[0148] Users can use the displayed feedback to revise their comments to make them more appropriate. After reviewing their comments again, they can resubmit them.

[0149] Step 9:

[0150] The server evaluates the revised comment and publishes it only if it is confirmed to be free of offensive elements and legal risks. This prevents the spread of inappropriate remarks.

[0151] (Example 2)

[0152] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0153] In recent years, there has been an increase in offensive language and legally risky content in online communication. This can lead to conflicts between users and potentially escalate into legal problems. This challenge highlights the need for systems that provide users with immediate feedback and facilitate appropriate communication.

[0154] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0155] In this invention, the server includes means for analyzing information input using natural language processing means to identify aggressive elements, means for analyzing emotions from input data and using a generative model to estimate emotional tone, and means for displaying a warning and suggesting corrections to the information when aggressive elements are detected. This enables immediate feedback to the user and constructive communication free from aggression.

[0156] "Natural language processing means" refers to technologies that enable computers to analyze human language and understand, evaluate, and classify its content.

[0157] "Emotional tone" refers to the overall atmosphere or tendency of emotions and attitudes extracted from writing or speech.

[0158] A "generative model" refers to a machine learning model that learns from large amounts of data and generates new data or information based on those results.

[0159] "Aggressive elements" refer to words or phrases that contain intent or content intended to hurt or offend someone.

[0160] "Legal risk" refers to the risk that one's words or actions may violate the law, resulting in legal problems.

[0161] "Feedback" refers to information provided to users, including evaluations and advice, that serves as a guide for improving or modifying their actions.

[0162] This invention is a system that analyzes user comments and posts on an information sharing platform to detect offensive elements and legal risks. Users input information using a terminal, and this input data is transmitted to the server in real time.

[0163] The server first analyzes the received input data using natural language processing (NLTK) tools. This NLTK processing utilizes software such as Python's NLTK or spaCy libraries. Through this process, the text is tokenized, and the presence or absence of offensive elements is determined.

[0164] Furthermore, the server uses a generative AI model to analyze the emotional tone of the input data. For this purpose, a large-scale language model, for example, is used. The server generates prompt sentences for the input data and sends them to the generative AI model to obtain an emotional score.

[0165] For example, if a user enters the comment, "Your opinion is completely nonsensical!", the server analyzes it. Natural language processing indicates that the word "nonsensical" is offensive, and the generative AI model detects a high level of "anger." The server then suggests to the user, "That expression is offensive. How about revising it to something like, 'Could you explain that in more detail?'" and simultaneously issues a legal warning if necessary.

[0166] An example of a prompt that can be generated is, "Analyze the sentiment of the following text: 'Your opinion is completely nonsensical!'"

[0167] The device receives feedback sent from the server and displays it to the user. Based on this feedback, the user can revise their comment and repost it in a form that is free from aggression and legal issues.

[0168] This process is expected to make communication between users safer and more constructive, and to prevent defamation and slander from occurring on online platforms.

[0169] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0170] Step 1:

[0171] Users enter comments and posts on the information sharing platform using their devices. The entered text data is immediately sent to the server. The input consists of the user's written content itself, which is sent as data.

[0172] Step 2:

[0173] The server analyzes the received input data using natural language processing libraries (e.g., NLTK or spaCy). This process involves tokenization and morphological analysis to evaluate the properties of each word and phrase in the text. The output generates a determination of whether or not the text contains offensive elements.

[0174] Step 3:

[0175] The server sends a prompt sentence to a generative AI model for sentiment analysis. The model calculates a sentiment score based on the input text data and estimates the emotional tone. In this process, the prompt sentence is passed to the model as input, and the sentiment score is returned as output.

[0176] Step 4:

[0177] The server combines natural language processing and sentiment scoring results to generate warning messages and suggestions for correction. If aggressive elements or strong emotions are detected, warnings and suggestions are created to be displayed to the user. These messages are generated as output.

[0178] Step 5:

[0179] A feedback message generated by the server is sent to the terminal. The terminal displays this information to the user, informing them of areas that need correction and presenting suggested corrections. The user can then review their comments based on this information.

[0180] Step 6:

[0181] Users can revise their comments based on feedback from their devices and repost them in a more abusive form. The final output will be improved comments that facilitate appropriate and constructive communication.

[0182] (Application Example 2)

[0183] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0184] In modern information-sharing platforms and communication methods, aggressive communication and legal disputes among users remain significant challenges. In particular, in environments where electronic transactions and communications take place in real time, inappropriate remarks and emotional exchanges can undermine the smooth operation of transactions. These issues necessitate the development of systems that provide a safe and constructive communication environment and manage users' emotions in a healthy manner.

[0185] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0186] In this invention, the server includes means for analyzing information input using natural language processing means and detecting aggressive elements, means for evaluating the emotional tone of the input information using emotion detection means, and means for operating on a personal device to capture the user's emotions in real time and immediately present suggested modifications. This makes communication between users healthy and safe, and facilitates smooth transactions and interactions.

[0187] "Natural language processing methods" are techniques for analyzing human language and converting it into a format that computers can understand.

[0188] "Aggressive elements" refer to words or expressions that may offend others.

[0189] "Legal risk" refers to situations that include actions or statements that may violate the law.

[0190] An "emotion detection method" is a technology that analyzes and evaluates the emotional nuances of input information.

[0191] "Personal devices" refer to electronic devices owned and used by individuals, such as smartphones and tablets.

[0192] "Real-time acquisition" refers to receiving information immediately, analyzing it, and providing processing results instantly.

[0193] A "revised proposal" refers to a suggestion to improve or change the content of the original information.

[0194] "Making communication healthy and safe" means maintaining a state where interactions are polite and avoid harmful misunderstandings and conflicts.

[0195] To realize this invention, a system for analyzing information in real time will be introduced to the information sharing platform. The server will immediately receive information entered by the user and analyze it using natural language processing and an emotion detection engine. In this process, Python and TENSORFLOW® will be used, and Hugging Face Transformers will be used as the natural language processing model. The emotion detection means will evaluate the emotional tone of the entered text and, if aggressive elements are found, will present suggested modifications to the user's personal device in real time.

[0196] The device manages the feedback displayed to the user and provides immediate suggestions for corrections and warnings from the system. This allows users to review their own comments and make corrections as needed.

[0197] For example, if a user comments "This product is a scam!" in an electronic transaction, the emotion detection engine will analyze this statement as strong "anger" and display a suggested correction on the smartphone: "That expression may be misleading. Why not try contacting them in a way that says, 'I am dissatisfied with the condition of the product. Could you suggest a solution?'"

[0198] An example of a prompt would be: "Analyze the message entered by the user to see if it contains offensive language, and if so, generate suggestions for more polite language."

[0199] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0200] Step 1:

[0201] A user enters a message on the information sharing platform. The entered text data is sent to the server in real time via the terminal. The input is the string entered by the user. The output is the raw text data transferred to the server.

[0202] Step 2:

[0203] The server passes the received text data to a natural language processing engine. This engine uses a generative AI model to process the input text and identify offensive words and phrases. The input is raw text data. The output is an evaluation result of whether offensive elements are present.

[0204] Step 3:

[0205] The server uses an emotion detection engine to evaluate the emotional tone of the text. This process uses Hugging Face Transformers to analyze the emotional state of the input message. The input is raw text data. The output is the emotion analysis result.

[0206] Step 4:

[0207] The server generates feedback based on the processed data, taking into account the user's emotions and the presence or absence of aggressive elements. This feedback includes suggested corrections if necessary. The input is the result of natural language processing and sentiment analysis. The output is a feedback message including suggested corrections.

[0208] Step 5:

[0209] The terminal displays feedback messages received from the server to the user. Based on this information, the user can review their own statements and, if necessary, input revised messages. The input is the feedback message, including suggested revisions. The output is the suggested revisions and warning messages presented to the user.

[0210] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0211] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0212] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0213] [Second Embodiment]

[0214] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0215] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0216] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0217] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0218] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0219] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0220] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0221] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0222] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0223] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0224] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0225] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0226] This invention is a system that analyzes input comments and posts to detect offensive elements and legal risks, thereby facilitating appropriate communication on a large-scale information sharing platform. The system comprises user terminals, a server, and their respective processing modules, and operates in real time.

[0227] First, when a user enters a comment on the information sharing platform, the content is sent from the device to the server. The server uses natural language processing technology to analyze the meaning and context of the comment. The purpose of the analysis is to identify offensive elements and evaluate their meaning.

[0228] If the server detects offensive elements in the input text, the system will immediately issue a warning to the user. The warning will inform the user that the entered content is inappropriate and simultaneously offer suggestions for correction. These suggestions will include specific recommendations for constructive and acceptable ways of expressing the content.

[0229] Furthermore, the server uses a legal risk avoidance module to analyze whether a comment could potentially be legally problematic. For example, if it detects expressions that could be considered defamatory or infringe on privacy, it provides the user with a warning and guidelines about the legal implications.

[0230] Feedback from the server is sent to the user's device, allowing the user to modify their comment based on that feedback. Ultimately, only comments that have been cleaned of offensive elements or legal risks are published. This approach improves the overall quality of communication on the platform.

[0231] For example, if a user enters a comment such as "You're a liar," the system will detect the comment and provide feedback such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is." The aim is to guide users to provide constructive feedback rather than slander.

[0232] This system is designed to prevent aggressive communication and provide a more constructive and safer environment for communication. It is hoped that this initiative will promote mutual understanding and respect among users.

[0233] The following describes the processing flow.

[0234] Step 1:

[0235] A user enters text into a comment field on the information sharing platform. Before the user finishes typing, the information is ready to be sent to the server in real time.

[0236] Step 2:

[0237] The terminal creates an API request to send the entered text to the server and sends it to the server. The server receives the entered text.

[0238] Step 3:

[0239] The server uses natural language processing tools to analyze the input text. Here, it identifies patterns of specific words and phrases to determine if they contain offensive elements.

[0240] Step 4:

[0241] If the server detects any offensive elements, it records those elements and their content. Next, it analyzes the context of the entire comment to identify the parts that were determined to have offensive intent.

[0242] Step 5:

[0243] The server generates a warning message for the user. The warning explains that the input is inappropriate and suggests appropriate corrections.

[0244] Step 6:

[0245] The server uses a legal risk avoidance module to assess whether submitted comments pose legal risks. This assessment involves checking for potential defamation and privacy violations.

[0246] Step 7:

[0247] The server will generate warnings about legal risks as needed and create guidelines to communicate these to users.

[0248] Step 8:

[0249] The server sends a warning message, suggested fixes, and legal warnings to the device. This allows the device to provide real-time feedback to the user.

[0250] Step 9:

[0251] Users revise their comments based on feedback from the server. Once revisions are complete, they review the comments again and submit them if necessary.

[0252] Step 10:

[0253] The revised comments are only published after being verified to be free of offensive elements and legal risks. This control prevents inappropriate content from spreading.

[0254] (Example 1)

[0255] Next, we will describe Example 1. 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."

[0256] In today's information and communication environment, offensive language and information with legal risks on online platforms cause many problems. Such information can trigger disputes between users and, in some cases, escalate into legal conflicts. Therefore, in order to maintain safe and constructive communication, it is necessary to detect problematic information in real time and deal with it appropriately. However, many current systems simply delete or block problematic information without showing users how to improve the information. Therefore, a mechanism is needed that enables users to acquire better communication skills and promote mutual understanding.

[0257] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0258] In this invention, the server includes means for analyzing information input and detecting offensive elements using natural language processing technology; means for issuing a warning when offensive elements are detected and automatically generating and presenting suggested corrections to the information using generation technology; means for analyzing whether the information contains legal risks and providing warnings and guidelines for legal risks as necessary; and means for making specific suggestions to encourage improvement to the user when offensive elements or legal risks are detected. As a result, users can correct comments in real time, and only information that has had offensive elements and legal risks removed is published, thus promoting safe and constructive communication.

[0259] "Information input" refers to data that users enter using characters and symbols on an online platform.

[0260] "Natural language processing technology" refers to the technology used by computers to understand, interpret, and generate human language.

[0261] "Offensive elements" refer to expressions or content that insult or offend others.

[0262] "Generative technology" refers to technology that uses artificial intelligence to automatically generate suggested modifications or responses to present to users.

[0263] "Legal risk" refers to situations or factors that could potentially violate the law if one speaks or acts in any way.

[0264] A "warning" is a notification that informs a user that certain information may cause problems.

[0265] "Guidelines" are instructional documents that provide specific steps and suggestions for users to appropriately correct information.

[0266] A "concrete proposal" refers to a clear and practical method for users to improve offensive or legally problematic information.

[0267] This system aims to eliminate offensive elements and avoid legal risks in online information sharing platforms. The specific form of this system is described below.

[0268] The user enters a message into a comment input field on the online platform. Once the user has finished typing, their device sends the message to the server. The device sends the data using a standard communication data transmission protocol, such as HTTP. The server has a processing system built on the Python language, and uses NLTK and generative AI models as natural language processing libraries.

[0269] The server uses this library to analyze the received comment data. During the analysis, it determines the emotions associated with the input, whether it contains aggressive elements, and evaluates whether any legal risks exist. Specifically, it uses text mining techniques to tokenize the comments and calculates an evaluation score using a sentiment analysis algorithm.

[0270] If offensive elements or legal risks are detected, generation technology is used to provide the user with a warning and specific corrective actions. For example, if a user says "You're a liar," the server will automatically generate constructive suggestions such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is."

[0271] The following example prompt is used in the generating AI model: "Evaluate the following comment and provide the problems and suggestions for improvement: 'You're a liar.'" This prompt allows the AI ​​to evaluate the inappropriate elements and suggest appropriate countermeasures.

[0272] Ultimately, users revise their comments based on feedback from the server. The revised comments are sent back to the server and are only published on the platform if they are deemed to no longer be offensive or pose any legal risks. This facilitates safe online communication.

[0273] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0274] Step 1:

[0275] Users enter messages into the comment input field on the information sharing platform. After entering the message, the user's device sends this input data (text comment) to the server. Here, data is entered, and the output is the text data sent to the server.

[0276] Step 2:

[0277] The server analyzes the received text data. It takes the received text data as input and performs data analysis using natural language processing libraries (such as Python's NLTK or generative AI models). Specifically, it tokenizes the comments and evaluates the aggression and sentiment scores of each word. The analysis results are generated as output.

[0278] Step 3:

[0279] The server detects offensive elements and legal risks from the analysis results. It uses the analyzed data as input to filter out the underlying problems. Specifically, it identifies problems by comparing them against pre-configured criteria and lists. The output generates a report indicating the presence or absence of problems and the identified risks.

[0280] Step 4:

[0281] The server generates feedback for the user based on the detected problems. Using data related to aggressive elements and legal risks as input, it utilizes a generative AI model to generate appropriate amendments and warning messages. Specifically, for a statement like "You are a liar," it provides more constructive suggestions. As output, feedback containing amendments and warnings is obtained.

[0282] Step 5:

[0283] The server sends the generated feedback to the user's terminal. Using the feedback content as input, it transmits information to the user's terminal using a data communication protocol. As output, the user views the feedback from the server.

[0284] Step 6:

[0285] The user revises the comment based on the received feedback. Using the feedback from the server as input, the user performs editing work while referring to it. After re - editing the comment, the user's terminal sends the revised comment to the server. As output, the revised comment data is sent to the server.

[0286] Step 7:

[0287] The server re - analyzes the revised comment to confirm that the problem has been solved. Receiving the revised comment as input, it performs analysis again using natural language processing technology. As output, a final confirmation result is generated, and if the comment meets the criteria, it is set for publication.

[0288] (Application Example 1)

[0289] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server," and the smart glasses 214 are referred to as the "terminal."

[0290] A problem exists in that aggressive remarks are unintentionally made during conversations between customers and store staff in physical stores, leading to a decline in service quality. Furthermore, insufficient responses to remarks that carry legal risks are also a concern. Therefore, there is a need for means to facilitate smooth communication with customers and improve service quality.

[0291] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0292] In this invention, the server includes means for analyzing conversation input using voice signal processing means to detect aggressive elements, means for issuing a warning and suggesting revisions to the conversation when aggressive elements are detected, and means for evaluating whether the conversation content includes legal risks and providing warnings of legal risks as necessary. This enables store employees and customers to have more constructive and safer conversations in physical stores.

[0293] "Audio signal processing" is a technology that analyzes audio data to understand its meaning and context.

[0294] "Conversation input" refers to audio information collected during a conversation, which records the actual exchange.

[0295] "Aggressive elements" refer to words or expressions in a conversation that could unnecessarily offend the other person.

[0296] "Issuing a warning and suggesting a correction" is the process of informing users when offensive elements are detected and suggesting more appropriate wording.

[0297] "Legal risk" refers to a situation where statements or actions have the potential to cause legal problems.

[0298] "Outputting real-time conversation analysis results on the screen" refers to a function that performs analysis immediately as a conversation takes place and displays the results right away.

[0299] The system for implementing this invention provides a function that analyzes the interaction between store staff and customers in real time in a physical store and promotes smooth communication.

[0300] First, a terminal within the store (e.g., a smartwatch) collects conversations between customers and staff using a speech recognition API (e.g., Google Cloud Speech-to-Text). This audio data is immediately converted into text. Next, a server uses a natural language processing API (e.g., OpenAI GPT-4) to analyze the text data in real time. This analysis identifies aggressive elements and legal risk factors within the conversation.

[0301] When aggressive elements are detected, the server immediately sends feedback to the terminal, informing the user of the identified problem and using a generative AI model to suggest appropriate corrections. This allows store employees to respond to customers quickly and appropriately. For example, a phrase deemed aggressive, such as "That's impossible, isn't it?", might be suggested to be modified to a more constructive expression, such as "Let's think together about whether that suggestion is truly feasible."

[0302] A concrete example of a prompt is, "Please revise the following sentence to be more constructive and customer-friendly: 'That's impossible, isn't it?'" As this example shows, the system optimizes user responses to be more customer-centric.

[0303] In this way, by using a generative AI model to analyze conversations through the cooperation of the server and terminal, and by providing appropriate feedback and suggested corrections immediately, the quality of communication in physical stores can be improved.

[0304] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0305] Step 1:

[0306] The terminal captures the conversation between the store clerk and the customer inside the store using a microphone. As input, voice data is collected. This voice data is converted into text data using a speech recognition API. This converted text data serves as the basis for the next processing.

[0307] Step 2:

[0308] The server receives the text data. As input, the converted text is used. Using a natural language processing API, the server analyzes the text and detects aggressive elements. For example, the word "impossible" may be detected as aggressive. The output is an analysis result that includes the aggressive elements and their position information.

[0309] Step 3:

[0310] The server evaluates whether there is a legal risk based on the analysis result. The input is the analysis result of Step 2. If a legal risk exists, appropriate guidelines are generated along with a warning of the legal risk. As output, the presence and details of the legal risk are obtained.

[0311] Step 4:

[0312] The server generates feedback on the aggressive elements and legal risks for the user and sends it to the terminal. The input is the information obtained in Step 2 and Step 3. Utilizing a generation AI model, flexible and constructive amendments are presented. The output is a feedback message.

[0313] Step 5:

[0314] The terminal displays the feedback received from the server to the store clerk. As a result, the store clerk can confirm the presented amendments and improve the conversation. The output is the display content on the terminal screen, specifically including the amendments and warning messages. The store clerk performs an operation to correct the conversation based on this.

[0315] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0316] This invention is a system that analyzes user comments and posts on an information sharing platform using a combination of natural language processing and an emotion engine to detect offensive elements and legal risks. When a user inputs information using a terminal, that data is transmitted to the server in real time for analysis.

[0317] On the server, the input information is first processed by natural language processing tools to determine if it contains aggressive elements. Simultaneously, an emotion engine estimates the user's emotions from the text and evaluates the emotional tone. Based on this evaluation, if, for example, strong anger or aggression is detected, a warning message is highlighted. Appropriate correction suggestions are also provided, and if necessary, warnings about legal risks are also issued.

[0318] For example, if a user comments, "Your opinion is completely nonsensical!", the server analyzes this text. Natural language processing identifies "nonsensical" as an aggressive word, and the emotion engine detects a high level of anger. The system then suggests to the user, "That expression is offensive. Why not try asking a question, such as, 'Could you explain that in more detail?'" and simultaneously warns of the legal implications.

[0319] The terminal receives feedback from the server and provides this information to the user in a timely manner. The user can refer to this feedback to revise their comment and repost it in an appropriate form, free from aggression and legal issues.

[0320] The goal of this system is to make communication between users safer and more constructive, and to prevent defamation and slander on online platforms. It also aims to improve users' healthy communication skills through immediate feedback and education.

[0321] The following describes the processing flow.

[0322] Step 1:

[0323] A user enters text into the comment field of the information sharing platform. The user's device monitors the input in real time and prepares it for transmission to the server.

[0324] Step 2:

[0325] The terminal sends the entered text to the server. The server receives the transmitted data and starts the analysis process.

[0326] Step 3:

[0327] The server uses a natural language processing module to analyze the received text. The analysis includes a process to identify whether there are any offensive words or expressions.

[0328] Step 4:

[0329] The server uses an emotion engine to estimate the user's emotional state from the input text. In this step, it primarily identifies emotions such as "anger," "joy," and "sadness," and evaluates their intensity.

[0330] Step 5:

[0331] Based on the analysis results, the server generates a warning message for the user if aggressive elements are detected. At the same time, it adjusts the tone of the warning message and suggested fixes based on the sentiment engine's evaluation.

[0332] Step 6:

[0333] The server uses a legal risk assessment module to check whether the input contains legal risks. If legal issues are found, it develops specific legal warnings for the user.

[0334] Step 7:

[0335] The server sends generated warnings, suggested fixes, and legal warnings to the device. The device then displays this feedback on its screen to provide the user with real-time support.

[0336] Step 8:

[0337] Users can use the displayed feedback to revise their comments to make them more appropriate. After reviewing their comments again, they can resubmit them.

[0338] Step 9:

[0339] The server evaluates the revised comment and publishes it only if it is confirmed to be free of offensive elements and legal risks. This prevents the spread of inappropriate remarks.

[0340] (Example 2)

[0341] Next, we will describe Example 2. 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".

[0342] In recent years, there has been an increase in offensive language and legally risky content in online communication. This can lead to conflicts between users and potentially escalate into legal problems. This challenge highlights the need for systems that provide users with immediate feedback and facilitate appropriate communication.

[0343] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0344] In this invention, the server includes means for analyzing information input using natural language processing means to identify aggressive elements, means for analyzing emotions from input data and using a generative model to estimate emotional tone, and means for displaying a warning and suggesting corrections to the information when aggressive elements are detected. This enables immediate feedback to the user and constructive communication free from aggression.

[0345] "Natural language processing means" refers to technologies that enable computers to analyze human language and understand, evaluate, and classify its content.

[0346] "Emotional tone" refers to the overall atmosphere or tendency of emotions and attitudes extracted from writing or speech.

[0347] A "generative model" refers to a machine learning model that learns from large amounts of data and generates new data or information based on those results.

[0348] "Aggressive elements" refer to words or phrases that contain intent or content intended to hurt or offend someone.

[0349] "Legal risk" refers to the risk that one's words or actions may violate the law, resulting in legal problems.

[0350] "Feedback" refers to information provided to users, including evaluations and advice, that serves as a guide for improving or modifying their actions.

[0351] This invention is a system that analyzes user comments and posts on an information sharing platform to detect offensive elements and legal risks. Users input information using a terminal, and this input data is transmitted to the server in real time.

[0352] The server first analyzes the received input data using natural language processing (NLTK) tools. This NLTK processing utilizes software such as Python's NLTK or spaCy libraries. Through this process, the text is tokenized, and the presence or absence of offensive elements is determined.

[0353] Furthermore, the server uses a generative AI model to analyze the emotional tone of the input data. For this purpose, a large-scale language model, for example, is used. The server generates prompt sentences for the input data and sends them to the generative AI model to obtain an emotional score.

[0354] For example, if a user enters the comment, "Your opinion is completely nonsensical!", the server analyzes it. Natural language processing indicates that the word "nonsensical" is offensive, and the generative AI model detects a high level of "anger." The server then suggests to the user, "That expression is offensive. How about revising it to something like, 'Could you explain that in more detail?'" and simultaneously issues a legal warning if necessary.

[0355] An example of a prompt that can be generated is, "Analyze the sentiment of the following text: 'Your opinion is completely nonsensical!'"

[0356] The device receives feedback sent from the server and displays it to the user. Based on this feedback, the user can revise their comment and repost it in a form that is free from aggression and legal issues.

[0357] This process is expected to make communication between users safer and more constructive, and to prevent defamation and slander from occurring on online platforms.

[0358] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0359] Step 1:

[0360] Users enter comments and posts on the information sharing platform using their devices. The entered text data is immediately sent to the server. The input consists of the user's written content itself, which is sent as data.

[0361] Step 2:

[0362] The server analyzes the received input data using natural language processing libraries (e.g., NLTK or spaCy). This process involves tokenization and morphological analysis to evaluate the properties of each word and phrase in the text. The output generates a determination of whether or not the text contains offensive elements.

[0363] Step 3:

[0364] The server sends a prompt sentence to a generative AI model for sentiment analysis. The model calculates a sentiment score based on the input text data and estimates the emotional tone. In this process, the prompt sentence is passed to the model as input, and the sentiment score is returned as output.

[0365] Step 4:

[0366] The server combines natural language processing and sentiment scoring results to generate warning messages and suggestions for correction. If aggressive elements or strong emotions are detected, warnings and suggestions are created to be displayed to the user. These messages are generated as output.

[0367] Step 5:

[0368] A feedback message generated by the server is sent to the terminal. The terminal displays this information to the user, informing them of areas that need correction and presenting suggested corrections. The user can then review their comments based on this information.

[0369] Step 6:

[0370] Users can revise their comments based on feedback from their devices and repost them in a more abusive form. The final output will be improved comments that facilitate appropriate and constructive communication.

[0371] (Application Example 2)

[0372] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0373] In modern information-sharing platforms and communication methods, aggressive communication and legal disputes among users remain significant challenges. In particular, in environments where electronic transactions and communications take place in real time, inappropriate remarks and emotional exchanges can undermine the smooth operation of transactions. These issues necessitate the development of systems that provide a safe and constructive communication environment and manage users' emotions in a healthy manner.

[0374] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0375] In this invention, the server includes means for analyzing information input using natural language processing means and detecting aggressive elements, means for evaluating the emotional tone of the input information using emotion detection means, and means for operating on a personal device to capture the user's emotions in real time and immediately present suggested modifications. This makes communication between users healthy and safe, and facilitates smooth transactions and interactions.

[0376] "Natural language processing methods" are techniques for analyzing human language and converting it into a format that computers can understand.

[0377] "Aggressive elements" refer to words or expressions that may offend others.

[0378] "Legal risk" refers to situations that include actions or statements that may violate the law.

[0379] An "emotion detection method" is a technology that analyzes and evaluates the emotional nuances of input information.

[0380] "Personal devices" refer to electronic devices owned and used by individuals, such as smartphones and tablets.

[0381] "Real-time acquisition" refers to receiving information immediately, analyzing it, and providing processing results instantly.

[0382] A "revised proposal" refers to a suggestion to improve or change the content of the original information.

[0383] "Making communication healthy and safe" means maintaining a state where interactions are polite and avoid harmful misunderstandings and conflicts.

[0384] To realize this invention, a system for analyzing information in real time will be introduced to the information sharing platform. The server will immediately receive information entered by the user and analyze it using natural language processing and sentiment detection engines. In this process, Python and TensorFlow will be used, and Hugging Face Transformers will be used as the natural language processing model. The sentiment detection means will evaluate the emotional tone of the input text and, if aggressive elements are found, will present suggested modifications to the user's personal device in real time.

[0385] The device manages the feedback displayed to the user and provides immediate suggestions for corrections and warnings from the system. This allows users to review their own comments and make corrections as needed.

[0386] For example, if a user comments "This product is a scam!" in an electronic transaction, the emotion detection engine will analyze this statement as strong "anger" and display a suggested correction on the smartphone: "That expression may be misleading. Why not try contacting them in a way that says, 'I am dissatisfied with the condition of the product. Could you suggest a solution?'"

[0387] An example of a prompt would be: "Analyze the message entered by the user to see if it contains offensive language, and if so, generate suggestions for more polite language."

[0388] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0389] Step 1:

[0390] A user enters a message on the information sharing platform. The entered text data is sent to the server in real time via the terminal. The input is the string entered by the user. The output is the raw text data transferred to the server.

[0391] Step 2:

[0392] The server passes the received text data to a natural language processing engine. This engine uses a generative AI model to process the input text and identify offensive words and phrases. The input is raw text data. The output is an evaluation result of whether offensive elements are present.

[0393] Step 3:

[0394] The server uses an emotion detection engine to evaluate the emotional tone of the text. This process uses Hugging Face Transformers to analyze the emotional state of the input message. The input is raw text data. The output is the emotion analysis result.

[0395] Step 4:

[0396] The server generates feedback based on the processed data, taking into account the user's emotions and the presence or absence of aggressive elements. This feedback includes suggested corrections if necessary. The input is the result of natural language processing and sentiment analysis. The output is a feedback message including suggested corrections.

[0397] Step 5:

[0398] The terminal displays feedback messages received from the server to the user. Based on this information, the user can review their own statements and, if necessary, input revised messages. The input is the feedback message, including suggested revisions. The output is the suggested revisions and warning messages presented to the user.

[0399] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0400] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0401] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0402] [Third Embodiment]

[0403] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0404] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0405] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0406] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0407] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0408] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0409] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0410] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0411] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0412] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0413] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0414] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0415] This invention is a system that analyzes input comments and posts to detect offensive elements and legal risks, thereby facilitating appropriate communication on a large-scale information sharing platform. The system comprises user terminals, a server, and their respective processing modules, and operates in real time.

[0416] First, when a user enters a comment on the information sharing platform, the content is sent from the device to the server. The server uses natural language processing technology to analyze the meaning and context of the comment. The purpose of the analysis is to identify offensive elements and evaluate their meaning.

[0417] If the server detects offensive elements in the input text, the system will immediately issue a warning to the user. The warning will inform the user that the entered content is inappropriate and simultaneously offer suggestions for correction. These suggestions will include specific recommendations for constructive and acceptable ways of expressing the content.

[0418] Furthermore, the server uses a legal risk avoidance module to analyze whether a comment could potentially be legally problematic. For example, if it detects expressions that could be considered defamatory or infringe on privacy, it provides the user with a warning and guidelines about the legal implications.

[0419] Feedback from the server is sent to the user's device, allowing the user to modify their comment based on that feedback. Ultimately, only comments that have been cleaned of offensive elements or legal risks are published. This approach improves the overall quality of communication on the platform.

[0420] For example, if a user enters a comment such as "You're a liar," the system will detect the comment and provide feedback such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is." The aim is to guide users to provide constructive feedback rather than slander.

[0421] This system is designed to prevent aggressive communication and provide a more constructive and safer environment for communication. It is hoped that this initiative will promote mutual understanding and respect among users.

[0422] The following describes the processing flow.

[0423] Step 1:

[0424] A user enters text into a comment field on the information sharing platform. Before the user finishes typing, the information is ready to be sent to the server in real time.

[0425] Step 2:

[0426] The terminal creates an API request to send the entered text to the server and sends it to the server. The server receives the entered text.

[0427] Step 3:

[0428] The server uses natural language processing tools to analyze the input text. Here, it identifies patterns of specific words and phrases to determine if they contain offensive elements.

[0429] Step 4:

[0430] If the server detects any offensive elements, it records those elements and their content. Next, it analyzes the context of the entire comment to identify the parts that were determined to have offensive intent.

[0431] Step 5:

[0432] The server generates a warning message for the user. The warning explains that the input is inappropriate and suggests appropriate corrections.

[0433] Step 6:

[0434] The server uses a legal risk avoidance module to assess whether submitted comments pose legal risks. This assessment involves checking for potential defamation and privacy violations.

[0435] Step 7:

[0436] The server will generate warnings about legal risks as needed and create guidelines to communicate these to users.

[0437] Step 8:

[0438] The server sends a warning message, suggested fixes, and legal warnings to the device. This allows the device to provide real-time feedback to the user.

[0439] Step 9:

[0440] Users revise their comments based on feedback from the server. Once revisions are complete, they review the comments again and submit them if necessary.

[0441] Step 10:

[0442] The revised comments are only published after being verified to be free of offensive elements and legal risks. This control prevents inappropriate content from spreading.

[0443] (Example 1)

[0444] Next, we will describe Example 1. 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."

[0445] In today's information and communication environment, offensive language and information with legal risks on online platforms cause many problems. Such information can trigger disputes between users and, in some cases, escalate into legal conflicts. Therefore, in order to maintain safe and constructive communication, it is necessary to detect problematic information in real time and deal with it appropriately. However, many current systems simply delete or block problematic information without showing users how to improve the information. Therefore, a mechanism is needed that enables users to acquire better communication skills and promote mutual understanding.

[0446] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0447] In this invention, the server includes means for analyzing information input and detecting offensive elements using natural language processing technology; means for issuing a warning when offensive elements are detected and automatically generating and presenting suggested corrections to the information using generation technology; means for analyzing whether the information contains legal risks and providing warnings and guidelines for legal risks as necessary; and means for making specific suggestions to encourage improvement to the user when offensive elements or legal risks are detected. As a result, users can correct comments in real time, and only information that has had offensive elements and legal risks removed is published, thus promoting safe and constructive communication.

[0448] "Information input" refers to data that users enter using characters and symbols on an online platform.

[0449] "Natural language processing technology" refers to the technology used by computers to understand, interpret, and generate human language.

[0450] "Offensive elements" refer to expressions or content that insult or offend others.

[0451] "Generative technology" refers to technology that uses artificial intelligence to automatically generate suggested modifications or responses to present to users.

[0452] "Legal risk" refers to situations or factors that could potentially violate the law if one speaks or acts in any way.

[0453] A "warning" is a notification that informs a user that certain information may cause problems.

[0454] "Guidelines" are instructional documents that provide specific steps and suggestions for users to appropriately correct information.

[0455] A "concrete proposal" refers to a clear and practical method for users to improve offensive or legally problematic information.

[0456] This system aims to eliminate offensive elements and avoid legal risks in online information sharing platforms. The specific form of this system is described below.

[0457] The user enters a message into a comment input field on the online platform. Once the user has finished typing, their device sends the message to the server. The device sends the data using a standard communication data transmission protocol, such as HTTP. The server has a processing system built on the Python language, and uses NLTK and generative AI models as natural language processing libraries.

[0458] The server uses this library to analyze the received comment data. During the analysis, it determines the emotions associated with the input, whether it contains aggressive elements, and evaluates whether any legal risks exist. Specifically, it uses text mining techniques to tokenize the comments and calculates an evaluation score using a sentiment analysis algorithm.

[0459] If offensive elements or legal risks are detected, generation technology is used to provide the user with a warning and specific corrective actions. For example, if a user says "You're a liar," the server will automatically generate constructive suggestions such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is."

[0460] The following example prompt is used in the generating AI model: "Evaluate the following comment and provide the problems and suggestions for improvement: 'You're a liar.'" This prompt allows the AI ​​to evaluate the inappropriate elements and suggest appropriate countermeasures.

[0461] Ultimately, users revise their comments based on feedback from the server. The revised comments are sent back to the server and are only published on the platform if they are deemed to no longer be offensive or pose any legal risks. This facilitates safe online communication.

[0462] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0463] Step 1:

[0464] Users enter messages into the comment input field on the information sharing platform. After entering the message, the user's device sends this input data (text comment) to the server. Here, data is entered, and the output is the text data sent to the server.

[0465] Step 2:

[0466] The server analyzes the received text data. It takes the received text data as input and performs data analysis using natural language processing libraries (such as Python's NLTK or generative AI models). Specifically, it tokenizes the comments and evaluates the aggression and sentiment scores of each word. The analysis results are generated as output.

[0467] Step 3:

[0468] The server detects offensive elements and legal risks from the analysis results. It uses the analyzed data as input to filter out the underlying problems. Specifically, it identifies problems by comparing them against pre-configured criteria and lists. The output generates a report indicating the presence or absence of problems and the identified risks.

[0469] Step 4:

[0470] The server generates feedback for the user based on detected issues. Using data on offensive elements and legal risks as input, it leverages a generative AI model to generate appropriate corrective actions and warnings. Specifically, it provides more constructive suggestions in response to statements like "You're a liar." The output is feedback that includes corrective actions and warnings.

[0471] Step 5:

[0472] The server sends the generated feedback to the user's device. As input, it retrieves the feedback content and sends the information to the user's device using a data communication protocol. As output, the user sees the feedback from the server.

[0473] Step 6:

[0474] The user revises their comment based on the feedback received. The user edits the comment, using the server's feedback as input. After re-editing the comment, the user's device sends the revised comment to the server. The revised comment data is then sent back to the server as output.

[0475] Step 7:

[0476] The server re-analyzes the corrected comment to confirm that the problem has been resolved. It receives the corrected comment as input and analyzes it again using natural language processing techniques. The output is a final verification result, and if the comment meets the criteria, it is set to be public.

[0477] (Application Example 1)

[0478] Next, we will explain Application Example 1. In the following explanation, 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."

[0479] A problem exists in that aggressive remarks are unintentionally made during conversations between customers and store staff in physical stores, leading to a decline in service quality. Furthermore, insufficient responses to remarks that carry legal risks are also a concern. Therefore, there is a need for means to facilitate smooth communication with customers and improve service quality.

[0480] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0481] In this invention, the server includes means for analyzing conversation input using voice signal processing means to detect aggressive elements, means for issuing a warning and suggesting revisions to the conversation when aggressive elements are detected, and means for evaluating whether the conversation content includes legal risks and providing warnings of legal risks as necessary. This enables store employees and customers to have more constructive and safer conversations in physical stores.

[0482] "Audio signal processing" is a technology that analyzes audio data to understand its meaning and context.

[0483] "Conversation input" refers to audio information collected during a conversation, which records the actual exchange.

[0484] "Aggressive elements" refer to words or expressions in a conversation that could unnecessarily offend the other person.

[0485] "Issuing a warning and suggesting a correction" is the process of informing users when offensive elements are detected and suggesting more appropriate wording.

[0486] "Legal risk" refers to a situation where statements or actions have the potential to cause legal problems.

[0487] "Outputting real-time conversation analysis results on the screen" refers to a function that performs analysis immediately as a conversation takes place and displays the results right away.

[0488] The system for implementing this invention provides a function that analyzes the interaction between store staff and customers in real time in a physical store and promotes smooth communication.

[0489] First, a terminal within the store (e.g., a smartwatch) collects conversations between customers and staff using a speech recognition API (e.g., Google Cloud Speech-to-Text). This audio data is immediately converted into text. Next, a server uses a natural language processing API (e.g., OpenAI GPT-4) to analyze the text data in real time. This analysis identifies aggressive elements and legal risk factors within the conversation.

[0490] When aggressive elements are detected, the server immediately sends feedback to the terminal, informing the user of the identified problem and using a generative AI model to suggest appropriate corrections. This allows store employees to respond to customers quickly and appropriately. For example, a phrase deemed aggressive, such as "That's impossible, isn't it?", might be suggested to be modified to a more constructive expression, such as "Let's think together about whether that suggestion is truly feasible."

[0491] A concrete example of a prompt is, "Please revise the following sentence to be more constructive and customer-friendly: 'That's impossible, isn't it?'" As this example shows, the system optimizes user responses to be more customer-centric.

[0492] In this way, by using a generative AI model to analyze conversations through the cooperation of the server and terminal, and by providing appropriate feedback and suggested corrections immediately, the quality of communication in physical stores can be improved.

[0493] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0494] Step 1:

[0495] The device captures conversations between store employees and customers using its microphone. Audio data is collected as input. A speech recognition API is used to convert this audio data into text data. This converted text data forms the basis for the next processing step.

[0496] Step 2:

[0497] The server receives text data. The converted text is used as input. Using a natural language processing API, the server analyzes the text and detects offensive elements. For example, the word "impossible" might be detected as offensive. The output is the analysis result, including the offensive elements and their location information.

[0498] Step 3:

[0499] The server evaluates whether there are legal risks based on the analysis results. The input is the analysis results from step 2. If legal risks exist, it generates appropriate guidelines along with a legal risk warning. The output provides whether or not there are legal risks and details.

[0500] Step 4:

[0501] The server generates and sends feedback to the user regarding offensive elements and legal risks. The input is the information obtained in steps 2 and 3. A generative AI model is used to present flexible and constructive suggestions for correction. The output is the feedback message.

[0502] Step 5:

[0503] The terminal displays feedback received from the server to the store clerk. This allows the clerk to review the suggested corrections and improve the conversation. The output is displayed on the terminal screen and specifically includes suggested corrections and warnings. The clerk then uses this to correct the conversation.

[0504] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0505] This invention is a system that analyzes user comments and posts on an information sharing platform using a combination of natural language processing and an emotion engine to detect offensive elements and legal risks. When a user inputs information using a terminal, that data is transmitted to the server in real time for analysis.

[0506] On the server, the input information is first processed by natural language processing tools to determine if it contains aggressive elements. Simultaneously, an emotion engine estimates the user's emotions from the text and evaluates the emotional tone. Based on this evaluation, if, for example, strong anger or aggression is detected, a warning message is highlighted. Appropriate correction suggestions are also provided, and if necessary, warnings about legal risks are also issued.

[0507] For example, if a user comments, "Your opinion is completely nonsensical!", the server analyzes this text. Natural language processing identifies "nonsensical" as an aggressive word, and the emotion engine detects a high level of anger. The system then suggests to the user, "That expression is offensive. Why not try asking a question, such as, 'Could you explain that in more detail?'" and simultaneously warns of the legal implications.

[0508] The terminal receives feedback from the server and provides this information to the user in a timely manner. The user can refer to this feedback to revise their comment and repost it in an appropriate form, free from aggression and legal issues.

[0509] The goal of this system is to make communication between users safer and more constructive, and to prevent defamation and slander on online platforms. It also aims to improve users' healthy communication skills through immediate feedback and education.

[0510] The following describes the processing flow.

[0511] Step 1:

[0512] A user enters text into the comment field of the information sharing platform. The user's device monitors the input in real time and prepares it for transmission to the server.

[0513] Step 2:

[0514] The terminal sends the entered text to the server. The server receives the transmitted data and starts the analysis process.

[0515] Step 3:

[0516] The server uses a natural language processing module to analyze the received text. The analysis includes a process to identify whether there are any offensive words or expressions.

[0517] Step 4:

[0518] The server uses an emotion engine to estimate the user's emotional state from the input text. In this step, it primarily identifies emotions such as "anger," "joy," and "sadness," and evaluates their intensity.

[0519] Step 5:

[0520] Based on the analysis results, the server generates a warning message for the user if aggressive elements are detected. At the same time, it adjusts the tone of the warning message and suggested fixes based on the sentiment engine's evaluation.

[0521] Step 6:

[0522] The server uses a legal risk assessment module to check whether the input contains legal risks. If legal issues are found, it develops specific legal warnings for the user.

[0523] Step 7:

[0524] The server sends generated warnings, suggested fixes, and legal warnings to the device. The device then displays this feedback on its screen to provide the user with real-time support.

[0525] Step 8:

[0526] Users can use the displayed feedback to revise their comments to make them more appropriate. After reviewing their comments again, they can resubmit them.

[0527] Step 9:

[0528] The server evaluates the revised comment and publishes it only if it is confirmed to be free of offensive elements and legal risks. This prevents the spread of inappropriate remarks.

[0529] (Example 2)

[0530] Next, we will describe Example 2. 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."

[0531] In recent years, there has been an increase in offensive language and legally risky content in online communication. This can lead to conflicts between users and potentially escalate into legal problems. This challenge highlights the need for systems that provide users with immediate feedback and facilitate appropriate communication.

[0532] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0533] In this invention, the server includes means for analyzing information input using natural language processing means to identify aggressive elements, means for analyzing emotions from input data and using a generative model to estimate emotional tone, and means for displaying a warning and suggesting corrections to the information when aggressive elements are detected. This enables immediate feedback to the user and constructive communication free from aggression.

[0534] "Natural language processing means" refers to technologies that enable computers to analyze human language and understand, evaluate, and classify its content.

[0535] "Emotional tone" refers to the overall atmosphere or tendency of emotions and attitudes extracted from writing or speech.

[0536] A "generative model" refers to a machine learning model that learns from large amounts of data and generates new data or information based on those results.

[0537] "Aggressive elements" refer to words or phrases that contain intent or content intended to hurt or offend someone.

[0538] "Legal risk" refers to the risk that one's words or actions may violate the law, resulting in legal problems.

[0539] "Feedback" refers to information provided to users, including evaluations and advice, that serves as a guide for improving or modifying their actions.

[0540] This invention is a system that analyzes user comments and posts on an information sharing platform to detect offensive elements and legal risks. Users input information using a terminal, and this input data is transmitted to the server in real time.

[0541] The server first analyzes the received input data using natural language processing (NLTK) tools. This NLTK processing utilizes software such as Python's NLTK or spaCy libraries. Through this process, the text is tokenized, and the presence or absence of offensive elements is determined.

[0542] Furthermore, the server uses a generative AI model to analyze the emotional tone of the input data. For this purpose, a large-scale language model, for example, is used. The server generates prompt sentences for the input data and sends them to the generative AI model to obtain an emotional score.

[0543] For example, if a user enters the comment, "Your opinion is completely nonsensical!", the server analyzes it. Natural language processing indicates that the word "nonsensical" is offensive, and the generative AI model detects a high level of "anger." The server then suggests to the user, "That expression is offensive. How about revising it to something like, 'Could you explain that in more detail?'" and simultaneously issues a legal warning if necessary.

[0544] An example of a prompt that can be generated is, "Analyze the sentiment of the following text: 'Your opinion is completely nonsensical!'"

[0545] The device receives feedback sent from the server and displays it to the user. Based on this feedback, the user can revise their comment and repost it in a form that is free from aggression and legal issues.

[0546] This process is expected to make communication between users safer and more constructive, and to prevent defamation and slander from occurring on online platforms.

[0547] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0548] Step 1:

[0549] Users enter comments and posts on the information sharing platform using their devices. The entered text data is immediately sent to the server. The input consists of the user's written content itself, which is sent as data.

[0550] Step 2:

[0551] The server analyzes the received input data using natural language processing libraries (e.g., NLTK or spaCy). This process involves tokenization and morphological analysis to evaluate the properties of each word and phrase in the text. The output generates a determination of whether or not the text contains offensive elements.

[0552] Step 3:

[0553] The server sends a prompt sentence to a generative AI model for sentiment analysis. The model calculates a sentiment score based on the input text data and estimates the emotional tone. In this process, the prompt sentence is passed to the model as input, and the sentiment score is returned as output.

[0554] Step 4:

[0555] The server combines natural language processing and sentiment scoring results to generate warning messages and suggestions for correction. If aggressive elements or strong emotions are detected, warnings and suggestions are created to be displayed to the user. These messages are generated as output.

[0556] Step 5:

[0557] A feedback message generated by the server is sent to the terminal. The terminal displays this information to the user, informing them of areas that need correction and presenting suggested corrections. The user can then review their comments based on this information.

[0558] Step 6:

[0559] Users can revise their comments based on feedback from their devices and repost them in a more abusive form. The final output will be improved comments that facilitate appropriate and constructive communication.

[0560] (Application Example 2)

[0561] Next, we will explain application example 2. In the following explanation, 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."

[0562] In modern information-sharing platforms and communication methods, aggressive communication and legal disputes among users remain significant challenges. In particular, in environments where electronic transactions and communications take place in real time, inappropriate remarks and emotional exchanges can undermine the smooth operation of transactions. These issues necessitate the development of systems that provide a safe and constructive communication environment and manage users' emotions in a healthy manner.

[0563] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0564] In this invention, the server includes means for analyzing information input using natural language processing means and detecting aggressive elements, means for evaluating the emotional tone of the input information using emotion detection means, and means for operating on a personal device to capture the user's emotions in real time and immediately present suggested modifications. This makes communication between users healthy and safe, and facilitates smooth transactions and interactions.

[0565] "Natural language processing methods" are techniques for analyzing human language and converting it into a format that computers can understand.

[0566] "Aggressive elements" refer to words or expressions that may offend others.

[0567] "Legal risk" refers to situations that include actions or statements that may violate the law.

[0568] An "emotion detection method" is a technology that analyzes and evaluates the emotional nuances of input information.

[0569] "Personal devices" refer to electronic devices owned and used by individuals, such as smartphones and tablets.

[0570] "Real-time acquisition" refers to receiving information immediately, analyzing it, and providing processing results instantly.

[0571] A "revised proposal" refers to a suggestion to improve or change the content of the original information.

[0572] "Making communication healthy and safe" means maintaining a state where interactions are polite and avoid harmful misunderstandings and conflicts.

[0573] To realize this invention, a system for analyzing information in real time will be introduced to the information sharing platform. The server will immediately receive information entered by the user and analyze it using natural language processing and sentiment detection engines. In this process, Python and TensorFlow will be used, and Hugging Face Transformers will be used as the natural language processing model. The sentiment detection means will evaluate the emotional tone of the input text and, if aggressive elements are found, will present suggested modifications to the user's personal device in real time.

[0574] The device manages the feedback displayed to the user and provides immediate suggestions for corrections and warnings from the system. This allows users to review their own comments and make corrections as needed.

[0575] For example, if a user comments "This product is a scam!" in an electronic transaction, the emotion detection engine will analyze this statement as strong "anger" and display a suggested correction on the smartphone: "That expression may be misleading. Why not try contacting them in a way that says, 'I am dissatisfied with the condition of the product. Could you suggest a solution?'"

[0576] An example of a prompt would be: "Analyze the message entered by the user to see if it contains offensive language, and if so, generate suggestions for more polite language."

[0577] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0578] Step 1:

[0579] A user enters a message on the information sharing platform. The entered text data is sent to the server in real time via the terminal. The input is the string entered by the user. The output is the raw text data transferred to the server.

[0580] Step 2:

[0581] The server passes the received text data to a natural language processing engine. This engine uses a generative AI model to process the input text and identify offensive words and phrases. The input is raw text data. The output is an evaluation result of whether offensive elements are present.

[0582] Step 3:

[0583] The server uses an emotion detection engine to evaluate the emotional tone of the text. This process uses Hugging Face Transformers to analyze the emotional state of the input message. The input is raw text data. The output is the emotion analysis result.

[0584] Step 4:

[0585] The server generates feedback based on the processed data, taking into account the user's emotions and the presence or absence of aggressive elements. This feedback includes suggested corrections if necessary. The input is the result of natural language processing and sentiment analysis. The output is a feedback message including suggested corrections.

[0586] Step 5:

[0587] The terminal displays feedback messages received from the server to the user. Based on this information, the user can review their own statements and, if necessary, input revised messages. The input is the feedback message, including suggested revisions. The output is the suggested revisions and warning messages presented to the user.

[0588] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0589] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0590] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0591] [Fourth Embodiment]

[0592] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0593] As shown in Figure 7, the 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.

[0594] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0595] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0596] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0597] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0598] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0599] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0600] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0601] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0602] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0603] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0604] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0605] This invention is a system that analyzes input comments and posts to detect offensive elements and legal risks, thereby facilitating appropriate communication on a large-scale information sharing platform. The system comprises user terminals, a server, and their respective processing modules, and operates in real time.

[0606] First, when a user enters a comment on the information sharing platform, the content is sent from the device to the server. The server uses natural language processing technology to analyze the meaning and context of the comment. The purpose of the analysis is to identify offensive elements and evaluate their meaning.

[0607] If the server detects offensive elements in the input text, the system will immediately issue a warning to the user. The warning will inform the user that the entered content is inappropriate and simultaneously offer suggestions for correction. These suggestions will include specific recommendations for constructive and acceptable ways of expressing the content.

[0608] Furthermore, the server uses a legal risk avoidance module to analyze whether a comment could potentially be legally problematic. For example, if it detects expressions that could be considered defamatory or infringe on privacy, it provides the user with a warning and guidelines about the legal implications.

[0609] Feedback from the server is sent to the user's device, allowing the user to modify their comment based on that feedback. Ultimately, only comments that have been cleaned of offensive elements or legal risks are published. This approach improves the overall quality of communication on the platform.

[0610] For example, if a user enters a comment such as "You're a liar," the system will detect the comment and provide feedback such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is." The aim is to guide users to provide constructive feedback rather than slander.

[0611] This system is designed to prevent aggressive communication and provide a more constructive and safer environment for communication. It is hoped that this initiative will promote mutual understanding and respect among users.

[0612] The following describes the processing flow.

[0613] Step 1:

[0614] A user enters text into a comment field on the information sharing platform. Before the user finishes typing, the information is ready to be sent to the server in real time.

[0615] Step 2:

[0616] The terminal creates an API request to send the entered text to the server and sends it to the server. The server receives the entered text.

[0617] Step 3:

[0618] The server uses natural language processing tools to analyze the input text. Here, it identifies patterns of specific words and phrases to determine if they contain offensive elements.

[0619] Step 4:

[0620] If the server detects any offensive elements, it records those elements and their content. Next, it analyzes the context of the entire comment to identify the parts that were determined to have offensive intent.

[0621] Step 5:

[0622] The server generates a warning message for the user. The warning explains that the input is inappropriate and suggests appropriate corrections.

[0623] Step 6:

[0624] The server uses a legal risk avoidance module to assess whether submitted comments pose legal risks. This assessment involves checking for potential defamation and privacy violations.

[0625] Step 7:

[0626] The server will generate warnings about legal risks as needed and create guidelines to communicate these to users.

[0627] Step 8:

[0628] The server sends a warning message, suggested fixes, and legal warnings to the device. This allows the device to provide real-time feedback to the user.

[0629] Step 9:

[0630] Users revise their comments based on feedback from the server. Once revisions are complete, they review the comments again and submit them if necessary.

[0631] Step 10:

[0632] The revised comments are only published after being verified to be free of offensive elements and legal risks. This control prevents inappropriate content from spreading.

[0633] (Example 1)

[0634] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0635] In today's information and communication environment, offensive language and information with legal risks on online platforms cause many problems. Such information can trigger disputes between users and, in some cases, escalate into legal conflicts. Therefore, in order to maintain safe and constructive communication, it is necessary to detect problematic information in real time and deal with it appropriately. However, many current systems simply delete or block problematic information without showing users how to improve the information. Therefore, a mechanism is needed that enables users to acquire better communication skills and promote mutual understanding.

[0636] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0637] In this invention, the server includes means for analyzing information input and detecting offensive elements using natural language processing technology; means for issuing a warning when offensive elements are detected and automatically generating and presenting suggested corrections to the information using generation technology; means for analyzing whether the information contains legal risks and providing warnings and guidelines for legal risks as necessary; and means for making specific suggestions to encourage improvement to the user when offensive elements or legal risks are detected. As a result, users can correct comments in real time, and only information that has had offensive elements and legal risks removed is published, thus promoting safe and constructive communication.

[0638] "Information input" refers to data that users enter using characters and symbols on an online platform.

[0639] "Natural language processing technology" refers to the technology used by computers to understand, interpret, and generate human language.

[0640] "Offensive elements" refer to expressions or content that insult or offend others.

[0641] "Generative technology" refers to technology that uses artificial intelligence to automatically generate suggested modifications or responses to present to users.

[0642] "Legal risk" refers to situations or factors that could potentially violate the law if one speaks or acts in any way.

[0643] A "warning" is a notification that informs a user that certain information may cause problems.

[0644] "Guidelines" are instructional documents that provide specific steps and suggestions for users to appropriately correct information.

[0645] A "concrete proposal" refers to a clear and practical method for users to improve offensive or legally problematic information.

[0646] This system aims to eliminate offensive elements and avoid legal risks in online information sharing platforms. The specific form of this system is described below.

[0647] The user enters a message into a comment input field on the online platform. Once the user has finished typing, their device sends the message to the server. The device sends the data using a standard communication data transmission protocol, such as HTTP. The server has a processing system built on the Python language, and uses NLTK and generative AI models as natural language processing libraries.

[0648] The server uses this library to analyze the received comment data. During the analysis, it determines the emotions associated with the input, whether it contains aggressive elements, and evaluates whether any legal risks exist. Specifically, it uses text mining techniques to tokenize the comments and calculates an evaluation score using a sentiment analysis algorithm.

[0649] If offensive elements or legal risks are detected, generation technology is used to provide the user with a warning and specific corrective actions. For example, if a user says "You're a liar," the server will automatically generate constructive suggestions such as, "Is that accusation based on facts? It would be helpful if you could specify what the problem is."

[0650] The following example prompt is used in the generating AI model: "Evaluate the following comment and provide the problems and suggestions for improvement: 'You're a liar.'" This prompt allows the AI ​​to evaluate the inappropriate elements and suggest appropriate countermeasures.

[0651] Ultimately, users revise their comments based on feedback from the server. The revised comments are sent back to the server and are only published on the platform if they are deemed to no longer be offensive or pose any legal risks. This facilitates safe online communication.

[0652] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0653] Step 1:

[0654] Users enter messages into the comment input field on the information sharing platform. After entering the message, the user's device sends this input data (text comment) to the server. Here, data is entered, and the output is the text data sent to the server.

[0655] Step 2:

[0656] The server analyzes the received text data. It takes the received text data as input and performs data analysis using natural language processing libraries (such as Python's NLTK or generative AI models). Specifically, it tokenizes the comments and evaluates the aggression and sentiment scores of each word. The analysis results are generated as output.

[0657] Step 3:

[0658] The server detects offensive elements and legal risks from the analysis results. It uses the analyzed data as input to filter out the underlying problems. Specifically, it identifies problems by comparing them against pre-configured criteria and lists. The output generates a report indicating the presence or absence of problems and the identified risks.

[0659] Step 4:

[0660] The server generates feedback for the user based on detected issues. Using data on offensive elements and legal risks as input, it leverages a generative AI model to generate appropriate corrective actions and warnings. Specifically, it provides more constructive suggestions in response to statements like "You're a liar." The output is feedback that includes corrective actions and warnings.

[0661] Step 5:

[0662] The server sends the generated feedback to the user's device. As input, it retrieves the feedback content and sends the information to the user's device using a data communication protocol. As output, the user sees the feedback from the server.

[0663] Step 6:

[0664] The user revises their comment based on the feedback received. The user edits the comment, using the server's feedback as input. After re-editing the comment, the user's device sends the revised comment to the server. The revised comment data is then sent back to the server as output.

[0665] Step 7:

[0666] The server re-analyzes the corrected comment to confirm that the problem has been resolved. It receives the corrected comment as input and analyzes it again using natural language processing techniques. The output is a final verification result, and if the comment meets the criteria, it is set to be public.

[0667] (Application Example 1)

[0668] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0669] A problem exists in that aggressive remarks are unintentionally made during conversations between customers and store staff in physical stores, leading to a decline in service quality. Furthermore, insufficient responses to remarks that carry legal risks are also a concern. Therefore, there is a need for means to facilitate smooth communication with customers and improve service quality.

[0670] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0671] In this invention, the server includes means for analyzing conversation input using voice signal processing means to detect aggressive elements, means for issuing a warning and suggesting revisions to the conversation when aggressive elements are detected, and means for evaluating whether the conversation content includes legal risks and providing warnings of legal risks as necessary. This enables store employees and customers to have more constructive and safer conversations in physical stores.

[0672] "Audio signal processing" is a technology that analyzes audio data to understand its meaning and context.

[0673] "Conversation input" refers to audio information collected during a conversation, which records the actual exchange.

[0674] "Aggressive elements" refer to words or expressions in a conversation that could unnecessarily offend the other person.

[0675] "Issuing a warning and suggesting a correction" is the process of informing users when offensive elements are detected and suggesting more appropriate wording.

[0676] "Legal risk" refers to a situation where statements or actions have the potential to cause legal problems.

[0677] "Outputting real-time conversation analysis results on the screen" refers to a function that performs analysis immediately as a conversation takes place and displays the results right away.

[0678] The system for implementing this invention provides a function that analyzes the interaction between store staff and customers in real time in a physical store and promotes smooth communication.

[0679] First, a terminal within the store (e.g., a smartwatch) collects conversations between customers and staff using a speech recognition API (e.g., Google Cloud Speech-to-Text). This audio data is immediately converted into text. Next, a server uses a natural language processing API (e.g., OpenAI GPT-4) to analyze the text data in real time. This analysis identifies aggressive elements and legal risk factors within the conversation.

[0680] When aggressive elements are detected, the server immediately sends feedback to the terminal, informing the user of the identified problem and using a generative AI model to suggest appropriate corrections. This allows store employees to respond to customers quickly and appropriately. For example, a phrase deemed aggressive, such as "That's impossible, isn't it?", might be suggested to be modified to a more constructive expression, such as "Let's think together about whether that suggestion is truly feasible."

[0681] A concrete example of a prompt is, "Please revise the following sentence to be more constructive and customer-friendly: 'That's impossible, isn't it?'" As this example shows, the system optimizes user responses to be more customer-centric.

[0682] In this way, by using a generative AI model to analyze conversations through the cooperation of the server and terminal, and by providing appropriate feedback and suggested corrections immediately, the quality of communication in physical stores can be improved.

[0683] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0684] Step 1:

[0685] The device captures conversations between store employees and customers using its microphone. Audio data is collected as input. A speech recognition API is used to convert this audio data into text data. This converted text data forms the basis for the next processing step.

[0686] Step 2:

[0687] The server receives text data. The converted text is used as input. Using a natural language processing API, the server analyzes the text and detects offensive elements. For example, the word "impossible" might be detected as offensive. The output is the analysis result, including the offensive elements and their location information.

[0688] Step 3:

[0689] The server evaluates whether there are legal risks based on the analysis results. The input is the analysis results from step 2. If legal risks exist, it generates appropriate guidelines along with a legal risk warning. The output provides whether or not there are legal risks and details.

[0690] Step 4:

[0691] The server generates and sends feedback to the user regarding offensive elements and legal risks. The input is the information obtained in steps 2 and 3. A generative AI model is used to present flexible and constructive suggestions for correction. The output is the feedback message.

[0692] Step 5:

[0693] The terminal displays feedback received from the server to the store clerk. This allows the clerk to review the suggested corrections and improve the conversation. The output is displayed on the terminal screen and specifically includes suggested corrections and warnings. The clerk then uses this to correct the conversation.

[0694] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0695] This invention is a system that analyzes user comments and posts on an information sharing platform using a combination of natural language processing and an emotion engine to detect offensive elements and legal risks. When a user inputs information using a terminal, that data is transmitted to the server in real time for analysis.

[0696] On the server, the input information is first processed by natural language processing tools to determine if it contains aggressive elements. Simultaneously, an emotion engine estimates the user's emotions from the text and evaluates the emotional tone. Based on this evaluation, if, for example, strong anger or aggression is detected, a warning message is highlighted. Appropriate correction suggestions are also provided, and if necessary, warnings about legal risks are also issued.

[0697] For example, if a user comments, "Your opinion is completely nonsensical!", the server analyzes this text. Natural language processing identifies "nonsensical" as an aggressive word, and the emotion engine detects a high level of anger. The system then suggests to the user, "That expression is offensive. Why not try asking a question, such as, 'Could you explain that in more detail?'" and simultaneously warns of the legal implications.

[0698] The terminal receives feedback from the server and provides this information to the user in a timely manner. The user can refer to this feedback to revise their comment and repost it in an appropriate form, free from aggression and legal issues.

[0699] The goal of this system is to make communication between users safer and more constructive, and to prevent defamation and slander on online platforms. It also aims to improve users' healthy communication skills through immediate feedback and education.

[0700] The following describes the processing flow.

[0701] Step 1:

[0702] A user enters text into the comment field of the information sharing platform. The user's device monitors the input in real time and prepares it for transmission to the server.

[0703] Step 2:

[0704] The terminal sends the entered text to the server. The server receives the transmitted data and starts the analysis process.

[0705] Step 3:

[0706] The server uses a natural language processing module to analyze the received text. The analysis includes a process to identify whether there are any offensive words or expressions.

[0707] Step 4:

[0708] The server uses an emotion engine to estimate the user's emotional state from the input text. In this step, it primarily identifies emotions such as "anger," "joy," and "sadness," and evaluates their intensity.

[0709] Step 5:

[0710] Based on the analysis results, the server generates a warning message for the user if aggressive elements are detected. At the same time, it adjusts the tone of the warning message and suggested fixes based on the sentiment engine's evaluation.

[0711] Step 6:

[0712] The server uses a legal risk assessment module to check whether the input contains legal risks. If legal issues are found, it develops specific legal warnings for the user.

[0713] Step 7:

[0714] The server sends generated warnings, suggested fixes, and legal warnings to the device. The device then displays this feedback on its screen to provide the user with real-time support.

[0715] Step 8:

[0716] Users can use the displayed feedback to revise their comments to make them more appropriate. After reviewing their comments again, they can resubmit them.

[0717] Step 9:

[0718] The server evaluates the revised comment and publishes it only if it is confirmed to be free of offensive elements and legal risks. This prevents the spread of inappropriate remarks.

[0719] (Example 2)

[0720] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0721] In recent years, there has been an increase in offensive language and legally risky content in online communication. This can lead to conflicts between users and potentially escalate into legal problems. This challenge highlights the need for systems that provide users with immediate feedback and facilitate appropriate communication.

[0722] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0723] In this invention, the server includes means for analyzing information input using natural language processing means to identify aggressive elements, means for analyzing emotions from input data and using a generative model to estimate emotional tone, and means for displaying a warning and suggesting corrections to the information when aggressive elements are detected. This enables immediate feedback to the user and constructive communication free from aggression.

[0724] "Natural language processing means" refers to technologies that enable computers to analyze human language and understand, evaluate, and classify its content.

[0725] "Emotional tone" refers to the overall atmosphere or tendency of emotions and attitudes extracted from writing or speech.

[0726] A "generative model" refers to a machine learning model that learns from large amounts of data and generates new data or information based on those results.

[0727] "Aggressive elements" refer to words or phrases that contain intent or content intended to hurt or offend someone.

[0728] "Legal risk" refers to the risk that one's words or actions may violate the law, resulting in legal problems.

[0729] "Feedback" refers to information provided to users, including evaluations and advice, that serves as a guide for improving or modifying their actions.

[0730] This invention is a system that analyzes user comments and posts on an information sharing platform to detect offensive elements and legal risks. Users input information using a terminal, and this input data is transmitted to the server in real time.

[0731] The server first analyzes the received input data using natural language processing (NLTK) tools. This NLTK processing utilizes software such as Python's NLTK or spaCy libraries. Through this process, the text is tokenized, and the presence or absence of offensive elements is determined.

[0732] Furthermore, the server uses a generative AI model to analyze the emotional tone of the input data. For this purpose, a large-scale language model, for example, is used. The server generates prompt sentences for the input data and sends them to the generative AI model to obtain an emotional score.

[0733] For example, if a user enters the comment, "Your opinion is completely nonsensical!", the server analyzes it. Natural language processing indicates that the word "nonsensical" is offensive, and the generative AI model detects a high level of "anger." The server then suggests to the user, "That expression is offensive. How about revising it to something like, 'Could you explain that in more detail?'" and simultaneously issues a legal warning if necessary.

[0734] An example of a prompt that can be generated is, "Analyze the sentiment of the following text: 'Your opinion is completely nonsensical!'"

[0735] The device receives feedback sent from the server and displays it to the user. Based on this feedback, the user can revise their comment and repost it in a form that is free from aggression and legal issues.

[0736] This process is expected to make communication between users safer and more constructive, and to prevent defamation and slander from occurring on online platforms.

[0737] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0738] Step 1:

[0739] Users enter comments and posts on the information sharing platform using their devices. The entered text data is immediately sent to the server. The input consists of the user's written content itself, which is sent as data.

[0740] Step 2:

[0741] The server analyzes the received input data using natural language processing libraries (e.g., NLTK or spaCy). This process involves tokenization and morphological analysis to evaluate the properties of each word and phrase in the text. The output generates a determination of whether or not the text contains offensive elements.

[0742] Step 3:

[0743] The server sends a prompt sentence to a generative AI model for sentiment analysis. The model calculates a sentiment score based on the input text data and estimates the emotional tone. In this process, the prompt sentence is passed to the model as input, and the sentiment score is returned as output.

[0744] Step 4:

[0745] The server combines natural language processing and sentiment scoring results to generate warning messages and suggestions for correction. If aggressive elements or strong emotions are detected, warnings and suggestions are created to be displayed to the user. These messages are generated as output.

[0746] Step 5:

[0747] A feedback message generated by the server is sent to the terminal. The terminal displays this information to the user, informing them of areas that need correction and presenting suggested corrections. The user can then review their comments based on this information.

[0748] Step 6:

[0749] Users can revise their comments based on feedback from their devices and repost them in a more abusive form. The final output will be improved comments that facilitate appropriate and constructive communication.

[0750] (Application Example 2)

[0751] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0752] In modern information-sharing platforms and communication methods, aggressive communication and legal disputes among users remain significant challenges. In particular, in environments where electronic transactions and communications take place in real time, inappropriate remarks and emotional exchanges can undermine the smooth operation of transactions. These issues necessitate the development of systems that provide a safe and constructive communication environment and manage users' emotions in a healthy manner.

[0753] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0754] In this invention, the server includes means for analyzing information input using natural language processing means and detecting aggressive elements, means for evaluating the emotional tone of the input information using emotion detection means, and means for operating on a personal device to capture the user's emotions in real time and immediately present suggested modifications. This makes communication between users healthy and safe, and facilitates smooth transactions and interactions.

[0755] "Natural language processing methods" are techniques for analyzing human language and converting it into a format that computers can understand.

[0756] "Aggressive elements" refer to words or expressions that may offend others.

[0757] "Legal risk" refers to situations that include actions or statements that may violate the law.

[0758] An "emotion detection method" is a technology that analyzes and evaluates the emotional nuances of input information.

[0759] "Personal devices" refer to electronic devices owned and used by individuals, such as smartphones and tablets.

[0760] "Real-time acquisition" refers to receiving information immediately, analyzing it, and providing processing results instantly.

[0761] A "revised proposal" refers to a suggestion to improve or change the content of the original information.

[0762] "Making communication healthy and safe" means maintaining a state where interactions are polite and avoid harmful misunderstandings and conflicts.

[0763] To realize this invention, a system for analyzing information in real time will be introduced to the information sharing platform. The server will immediately receive information entered by the user and analyze it using natural language processing and sentiment detection engines. In this process, Python and TensorFlow will be used, and Hugging Face Transformers will be used as the natural language processing model. The sentiment detection means will evaluate the emotional tone of the input text and, if aggressive elements are found, will present suggested modifications to the user's personal device in real time.

[0764] The device manages the feedback displayed to the user and provides immediate suggestions for corrections and warnings from the system. This allows users to review their own comments and make corrections as needed.

[0765] For example, if a user comments "This product is a scam!" in an electronic transaction, the emotion detection engine will analyze this statement as strong "anger" and display a suggested correction on the smartphone: "That expression may be misleading. Why not try contacting them in a way that says, 'I am dissatisfied with the condition of the product. Could you suggest a solution?'"

[0766] An example of a prompt would be: "Analyze the message entered by the user to see if it contains offensive language, and if so, generate suggestions for more polite language."

[0767] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0768] Step 1:

[0769] A user enters a message on the information sharing platform. The entered text data is sent to the server in real time via the terminal. The input is the string entered by the user. The output is the raw text data transferred to the server.

[0770] Step 2:

[0771] The server passes the received text data to a natural language processing engine. This engine uses a generative AI model to process the input text and identify offensive words and phrases. The input is raw text data. The output is an evaluation result of whether offensive elements are present.

[0772] Step 3:

[0773] The server uses an emotion detection engine to evaluate the emotional tone of the text. This process uses Hugging Face Transformers to analyze the emotional state of the input message. The input is raw text data. The output is the emotion analysis result.

[0774] Step 4:

[0775] The server generates feedback based on the processed data, taking into account the user's emotions and the presence or absence of aggressive elements. This feedback includes suggested corrections if necessary. The input is the result of natural language processing and sentiment analysis. The output is a feedback message including suggested corrections.

[0776] Step 5:

[0777] The terminal displays feedback messages received from the server to the user. Based on this information, the user can review their own statements and, if necessary, input revised messages. The input is the feedback message, including suggested revisions. The output is the suggested revisions and warning messages presented to the user.

[0778] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0779] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0780] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0781] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0782] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0783] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0784] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0785] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0786] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0787] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0788] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0789] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0790] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0791] 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.

[0792] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0793] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0794] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0795] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0796] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0797] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0798] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0799] The following is further disclosed regarding the embodiments described above.

[0800] (Claim 1)

[0801] A means for analyzing information input using natural language processing and detecting aggressive elements,

[0802] A means of issuing a warning when offensive elements are detected and suggesting corrections to the information,

[0803] A means of assessing whether the information contains legal risks and providing warnings about legal risks as necessary,

[0804] A system that includes this.

[0805] (Claim 2)

[0806] The system according to claim 1, comprising means for analyzing input information and generating a response based on the relevant information, and presenting the generated information response on a screen.

[0807] (Claim 3)

[0808] The system according to claim 1, further comprising control means for ensuring that information is made public only after it has been modified to not contain offensive elements.

[0809] "Example 1"

[0810] (Claim 1)

[0811] A means of analyzing information input and detecting aggressive elements using natural language processing technology,

[0812] A means of issuing a warning when offensive elements are detected and automatically generating and presenting proposed corrections to the information using generation technology,

[0813] A means to analyze whether the information contains legal risks and, if necessary, to provide warnings and guidelines regarding those legal risks.

[0814] A means of providing specific suggestions to encourage users to make improvements when offensive elements or legal risks are detected,

[0815] A system that includes this.

[0816] (Claim 2)

[0817] The system according to claim 1, further comprising means for presenting a response automatically generated based on analyzed input information to a display device.

[0818] (Claim 3)

[0819] The system according to claim 1, comprising means for controlling whether information is made public only after it has been modified to remove offensive elements and legal risks.

[0820] "Application Example 1"

[0821] (Claim 1)

[0822] A means for analyzing conversation input using speech signal processing means and detecting aggressive elements,

[0823] A means of issuing a warning when aggressive elements are detected and suggesting ways to modify the conversation,

[0824] A means of evaluating whether the content of a conversation contains legal risks and providing warnings of legal risks as necessary,

[0825] A means of outputting the results of conversation analysis to the screen in real time,

[0826] A system that includes this.

[0827] (Claim 2)

[0828] The system according to claim 1, comprising a means for generating response candidates and presenting an improvement dialogue based on the proposed revisions.

[0829] (Claim 3)

[0830] The system according to claim 1, further comprising control means for ensuring that conversation results are shared only if the conversation has been modified to not contain offensive elements.

[0831] "Example 2 of combining an emotion engine"

[0832] (Claim 1)

[0833] A means of analyzing information input using natural language processing and identifying aggressive elements,

[0834] A method that uses a generative model to analyze emotions from input data and estimate emotional tone,

[0835] A means of displaying a warning when offensive elements are detected and suggesting corrections to the information,

[0836] A means of assessing whether the information contains legal risks and providing warnings of legal risks as necessary,

[0837] A means of providing users with feedback and encouraging them to revise their comments,

[0838] A system that includes this.

[0839] (Claim 2)

[0840] The system according to claim 1, comprising a structure that processes received information and generates a response based on the corresponding data, and displays the generated response.

[0841] (Claim 3)

[0842] The system according to claim 1, comprising means for controlling whether information is made public only after it has been modified to not contain offensive elements.

[0843] "Application example 2 when combining with an emotional engine"

[0844] (Claim 1)

[0845] A means for analyzing information input using natural language processing and detecting aggressive elements,

[0846] A means of issuing a warning when offensive elements are detected and suggesting corrections to the information,

[0847] A means of assessing whether the information contains legal risks and providing warnings about legal risks as necessary,

[0848] A means for evaluating the emotional tone of input information using an emotion detection means,

[0849] A means of operating on a personal device, capturing the user's emotions in real time, and immediately presenting suggested modifications,

[0850] A system that includes this.

[0851] (Claim 2)

[0852] The system according to claim 1, comprising means for analyzing input information and generating a response based on the relevant information, and presenting the generated information response on a screen.

[0853] (Claim 3)

[0854] The system according to claim 1, further comprising control means for displaying appropriate suggestions on the user's screen when emotional elements are detected, and for ensuring that the information is only made public after it has been modified to be free of offensive elements. [Explanation of Symbols]

[0855] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for analyzing information input using natural language processing and detecting aggressive elements, A means of issuing a warning when offensive elements are detected and suggesting corrections to the information, A means of assessing whether the information contains legal risks and providing warnings about legal risks as necessary, A system that includes this.

2. The system according to claim 1, comprising means for analyzing input information and generating a response based on the relevant information, and presenting the generated information response on a screen.

3. The system according to claim 1, further comprising control means for ensuring that information is made public only after it has been modified to not contain offensive elements.

Images