system

Abstract

Provide a system. 【Solution means】 Means for analyzing information generated by an information sender by an information analysis device and evaluating risks based on legal and ethical standards; Means for presenting an amendment to the information sender based on the risk assessment; Means for continuously monitoring external reactions to information transmitted by an information transmission device and detecting abnormal reactions; Means for generating a response text based on past data and providing it to the information sender when an abnormal reaction is detected; Means for analyzing the collected external reactions and providing a summary thereof to the information sender; Means for enhancing natural language processing using a generation AI model for the analysis of posted information and the generation of amendments; Means for dynamically adjusting the input prompt text to the generation AI model to improve analysis accuracy; A system including the above.

Description

Technical Field

[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 and includes 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] In information dissemination platforms such as SNS, there is a possibility that an information disseminator may inadvertently disseminate information containing legal or ethical issues. Also, if a resulting conflagration occurs, appropriate and prompt response is required, but the information disseminator often has difficulty determining how to handle it. Furthermore, the response to the disseminated content imposing an excessive mental burden on the information disseminator has also been a problem. The present invention aims to solve these problems and enable an information disseminator to disseminate information with confidence.

Means for Solving the Problems

[0005] This invention provides a system that uses an information analysis device to analyze information generated by an information sender in advance, thereby evaluating risks based on legal and ethical standards. Based on the risk evaluation, the system proposes revisions to the information sender, enabling them to correct problematic content in advance. The system also includes a function to continuously monitor external reactions to information sent by the information sender and detect abnormal reactions such as online firestorms. This allows the system to automatically generate appropriate response drafts based on past data when abnormal reactions occur and provide them to the information sender promptly. Furthermore, by analyzing the collected external reactions and presenting a summary to the information sender, the system reduces the mental burden on the sender and provides suggestions for improvement for future communication activities. This enables information senders to communicate information with minimal risk.

[0006] An "information analysis device" is a device that receives information generated by an information sender and has the function of analyzing and evaluating it based on legal and ethical standards.

[0007] An "information provider" refers to an individual or group that creates and disseminates information on platforms such as social media.

[0008] "Legal and ethical standards" refer to criteria for judging the appropriateness of information from the perspective of law and socially required ethics.

[0009] "Risk assessment" is the process of detecting potential legal or ethical issues contained in information and determining their importance and impact.

[0010] A "revised proposal" refers to specific changes or suggestions aimed at mitigating or eliminating the risks pointed out to the information provider.

[0011] An "information dissemination device" refers to a device or software used to post information on platforms such as social networking services (SNS).

[0012] "External reactions" refers to all activities such as comments, reactions, and shares received from others after information has been released.

[0013] An "abnormal reaction" refers to a reaction that is particularly negative or involves a sudden change compared to a typical reaction, and includes things like online outrage and criticism.

[0014] A "response message" refers to a series of automatically generated messages proposed by the system as an appropriate response to an abnormal reaction.

[0015] "Collected external reactions" refers to the aggregate of all feedback, comments, and reactions received from others on the platform after the information has been disseminated. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This 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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This 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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] Shows an emotion map to which a plurality of 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.

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 according to the attached drawings.

[0018] First, the language used in the following description will be described.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of 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, the 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, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[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, a 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 provides a system that checks information before posting it to social media and other platforms to prevent problems from occurring. This system mainly includes an information analysis device, a terminal, and a server.

[0038] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0039] When the server receives submitted data, it analyzes the data using natural language processing techniques. The purpose of the analysis is to check whether the submitted content poses any legal or ethical risks. For example, it identifies whether it contains racist or potentially misleading language. Based on these results, the server lists the potential risks present in the submission and generates specific suggestions for revision.

[0040] Next, the server sends the generated risk assessment and proposed corrections to the user's device. On the device, this information is displayed visually to the user, allowing them to modify their post considering the identified risks.

[0041] After a post is published, the server continuously monitors external reactions to that post. The server collects the reactions and determines if they are abnormal. For example, a sudden surge in critical comments or a topic becoming a major topic of discussion in a short period of time may be considered a sign of an impending online firestorm.

[0042] If a controversy is detected, the server quickly references similar past case studies and generates a draft apology. This draft apology is crafted with sincerity and specificity in mind and is promptly provided to the user.

[0043] Furthermore, the server analyzes the collected external reactions and generates a summary of user feedback. This summary shows the proportion of positive, negative, and neutral opinions and includes specific suggestions for improvement that will help users with future posts.

[0044] As a concrete example, consider a scenario where a public relations representative for a company posts an introduction to a new product. Suppose this post contains an inappropriate comparison to a competing product. The system detects this and suggests revisions from an ethical standpoint. The PR representative adjusts the post according to the suggestions, implements preventative measures, and then publishes it. If, after publication, criticism from the competing company increases, the server provides a suitable apology draft and also offers suggestions for improving future posts.

[0045] In this way, the system of the present invention supports information providers in communicating smoothly while minimizing risks.

[0046] The following describes the processing flow.

[0047] Step 1:

[0048] The user creates a post on their device. The device receives the user input as text data.

[0049] Step 2:

[0050] The device presses the "Confirm" button. Afterward, the device prepares to send the posted data to the server.

[0051] Step 3:

[0052] The server receives the submitted data. The server executes a natural language processing algorithm to analyze the information.

[0053] Step 4:

[0054] The server evaluates the content of posts based on legal and ethical standards. If there are any risk points, they are identified and highlighted.

[0055] Step 5:

[0056] The server generates revised plans along with the evaluation results. These plans include specific suggestions for mitigating risks.

[0057] Step 6:

[0058] The server sends the generated evaluation results and suggested corrections to the terminal. The results are then displayed to the user on the terminal.

[0059] Step 7:

[0060] Users edit their posts through their devices. They make corrections as needed based on the identified risks and then finalize the post.

[0061] Step 8:

[0062] The user publishes a post, and the device uploads that information to the social networking platform.

[0063] Step 9:

[0064] The server continuously monitors external reactions to posts after they are published, paying particular attention to criticism and sudden increases in the number of comments.

[0065] Step 10:

[0066] When the server detects an abnormal response, it automatically generates an appropriate response by referring to past case studies.

[0067] Step 11:

[0068] The server generates a draft response message and sends it to the terminal. The user reviews it and customizes it as needed.

[0069] Step 12:

[0070] The server analyzes external reaction comments and summarizes them as positive, negative, or neutral.

[0071] Step 13:

[0072] The server sends a summary and improvement suggestions to the user's device, which helps them in creating future posts.

[0073] (Example 1)

[0074] 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."

[0075] There is a lack of solutions to prevent legal and ethical problems when information providers publish information on social media and other online platforms, and to respond quickly and appropriately to reactions after publication. Furthermore, in a situation where unexpected external reactions and criticisms to published information are rapidly increasing, there is a need for a mechanism that efficiently presents countermeasures and reduces the burden on information providers.

[0076] 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.

[0077] In this invention, the server includes means for analyzing information generated by the information sender using natural language processing technology and evaluating risks based on legal and ethical standards; means for generating revised versions based on prompt sentences using a generation AI model and presenting them to the information sender; and means for continuously monitoring external reactions and detecting abnormal reactions. As a result, the information sender can grasp risks in advance, revise their expression to an appropriate one, and respond quickly to reactions after publication.

[0078] An "information analysis device" is a device that analyzes data generated by information providers and has the function of evaluating risks based on legal and ethical standards.

[0079] "Natural language processing technology" is a method that understands text data semantically and grammatically, and analyzes the ambiguity and complexity unique to human language.

[0080] A "generative AI model" is an artificial intelligence model that learns from past data and generates new text or suggestions.

[0081] A "prompt" is a sentence of instruction or question input to an AI model to obtain a specific response or output.

[0082] "Risk assessment" is the process of determining whether the content of information may cause legal or ethical problems before it is made public.

[0083] A "proposal for revision" refers to specific changes proposed to make the information being disseminated more appropriate and problem-free.

[0084] An "abnormal reaction" refers to a response that exhibits critical or emotional feedback from an external source that exceeds the normal range.

[0085] A "draft response" is a document provided by the information provider to enable them to respond quickly to detected abnormal reactions.

[0086] A "summary" is a shortened version of collected external responses, extracting the key points and providing them to the information provider.

[0087] This invention is a system that minimizes the risks for information providers when posting information on online platforms. The system mainly consists of a server, terminals, and users.

[0088] server

[0089] The server receives posted data sent from the user's terminal. This data is analyzed using natural language processing techniques. Specifically, Python natural language processing libraries (e.g., spaCy, NLTK) are used to evaluate the legal and ethical risks of the posted content. The server utilizes a generative AI model to generate suggested revisions based on the provided prompt text. These suggested revisions are notified to the user via their terminal. The server also monitors external reactions after the post is published and detects any abnormal reactions. If an abnormal reaction is detected, it generates a suggested response based on past data and provides it to the user promptly.

[0090] terminal

[0091] The device collects user-entered posting data and sends it to the server. It then receives suggested revisions and risk assessments from the server, which are displayed visually to the user. This allows the user to review their posts and make revisions as needed.

[0092] User

[0093] Users can use this system to create their posts, reviewing risk assessments and suggested revisions along the way. For example, suppose a corporate public relations representative creating a new product description includes ethically problematic language. In this case, the system will detect the language and suggest revisions. The user can then adjust the content based on the feedback and make an appropriate post.

[0094] For example, a prompt might include instructions such as, "Please revise this expression to be neutral." Based on these instructions, the AI ​​generates specific revisions to mitigate risk. This allows users to communicate information more safely and effectively.

[0095] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0096] Step 1:

[0097] Users create post data on their devices. This post data may include text, images, or videos. The information entered by the user is first temporarily stored on the device. At this time, a post text is generated based on the user's goals and intentions.

[0098] Step 2:

[0099] The terminal prepares to send the created post data to the server. The terminal converts the data to the appropriate format and sends it to the server using the encrypted HTTP protocol, ensuring data security over network communication.

[0100] Step 3:

[0101] The server receives posted data sent from the terminal. Based on the received data, it analyzes the data using natural language processing techniques. This analysis process breaks down the text, extracts keywords and phrases, and assesses the legal or ethical risks. For example, it calculates a sensitivity score for each sentence and detects inappropriate phrases.

[0102] Step 4:

[0103] Based on the analysis results, the server generates suggested corrections for areas where risks were detected. Specifically, it uses a generation AI model to receive the prompt "Please correct this expression to be neutral" and generates a new expression. During this process, data is input to the model and suggested corrections are output.

[0104] Step 5:

[0105] The server sends the generated revised proposals and risk assessments to the terminal. The transmitted data is formatted in a way that is easily understandable to the user.

[0106] Step 6:

[0107] The terminal displays the risk assessment and proposed corrections received from the server to the user. Using a visual GUI, the terminal highlights risk areas and displays proposed corrections in pop-ups or side panels. This allows users to easily review their feedback.

[0108] Step 7:

[0109] Users review their posts based on the suggested revisions and make corrections as needed. The revised posts are then reviewed and ready for publication.

[0110] Step 8:

[0111] The server continuously monitors external reactions to a post even after it has been published. The server collects data in real time and analyzes whether any abnormal reactions are occurring. This analysis includes comment frequency and sentiment analysis.

[0112] Step 9:

[0113] If the server detects an abnormal response, it will generate a suggested response based on similar past cases and provide it to the user. This suggested response will include apologies and suggestions for corrections to help the user respond quickly.

[0114] (Application Example 1)

[0115] 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."

[0116] In modern digital communication, information dissemination via social media is commonplace, but at the same time, inappropriate posts that carry legal and ethical risks have become a problem. Therefore, it is necessary to detect these risks in advance and address them appropriately. However, conventional systems have difficulty responding in real time, and there are challenges in making highly accurate risk assessments and corrective suggestions.

[0117] 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.

[0118] In this invention, the server includes means for analyzing information generated by information senders using an information analysis device and evaluating the risks based on legal and ethical standards; means for enhancing natural language processing using a generative AI model to analyze posted information and generate proposed revisions; and means for dynamically adjusting input prompt sentences to the generative AI model to improve analysis accuracy. This makes it possible to evaluate the risks in SNS posts in advance with high accuracy and to quickly present appropriate revision suggestions.

[0119] An "information analysis device" is a device that analyzes data generated by information providers and assesses risks based on legal and ethical standards.

[0120] A "generative AI model" is an artificial intelligence model designed for advanced natural language processing, and is used for analyzing posted information and generating suggested revisions.

[0121] A "prompt sentence" is a sentence used as input to a generative AI model, and it is dynamically adjusted to improve the accuracy of the analysis.

[0122] "External reactions" refer to comments and feedback received from third parties regarding information made public through information dissemination devices.

[0123] An "abnormal reaction" refers to a characteristic response that is not typical, such as a sudden surge in critical comments or rapid growth in public attention.

[0124] A "response draft" is a draft text, generated based on past data and successful cases, intended to be an apology or response, provided to the information provider when an abnormal reaction is detected.

[0125] A "revision proposal" is a specific suggestion for correction presented to the information provider when legal or ethical risks are detected in the content of their post.

[0126] A "summary" is information that shows the proportion of positive, negative, and neutral opinions, compiled to provide information providers with the results of an analysis of collected external reactions.

[0127] To implement this invention, the server must be equipped with an information analysis device, a generative AI model, and a prompt sentence adjustment system. The server receives posted data from information senders and performs analysis using natural language processing technology. This analysis utilizes a generative AI model to perform a highly accurate risk assessment of the text. At this time, the prompt sentences input to the generative AI model are dynamically adjusted to improve the accuracy of the analysis.

[0128] Based on the risk assessment generated from the analysis results, the server presents the user with suggested revisions on their device. The user can then adjust their post according to the suggested revisions displayed on their device and resubmit it. This feature enables the dissemination of information that complies with legal and ethical standards.

[0129] Furthermore, after the information is published, the server continuously monitors external reactions. If an abnormal reaction is detected, the server uses a generative AI model, based on past cases, to generate an appropriate response and provide it to the user. In addition, the server analyzes the collected external reactions and presents them to the information provider as a summary, which can be used to improve future posts.

[0130] For example, if a user is about to post an opinion about a new product, this system can pre-scan the content, identify potentially misleading parts, and suggest neutral wording.

[0131] An example of a prompt message might be: "Regarding a new social media post, it says: 'Our new product is the best on the market and overwhelmingly superior to the competition.' Please assess whether this post contains any ethical or legal risks and provide suggestions for revisions if necessary."

[0132] In this way, the entire system enables information providers to consistently manage risks both before and after the event, and to facilitate smooth communication.

[0133] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0134] Step 1:

[0135] Users create post data using their devices. This post data is prepared to be sent from the device to the server for sharing on the SNS platform. The input is free-form text data created by the user, and the output is text data ready to be sent to the server.

[0136] Step 2:

[0137] The server uses a generative AI model to perform natural language processing to process the received posted data. The server prepares prompt sentences and inputs these prompt sentences into the AI ​​model for analysis. The input consists of the user's posted text and prompt sentences, and the output is an assessment of legal and ethical risks.

[0138] Step 3:

[0139] Based on the generated risk assessment, the server creates proposed revisions to the posted content. These revisions include specific suggestions for improving particular parts of the information. The input is the result of the risk assessment, and the output is the proposed revisions.

[0140] Step 4:

[0141] The suggested revisions are displayed on the terminal's user interface and presented to the user. The user reviews them and modifies the post as needed. The input is the suggested revisions sent from the server, and the output is the updated post after the user has made the necessary corrections.

[0142] Step 5:

[0143] After the information is officially released, the server continuously monitors external reactions on the SNS platform. If an abnormal reaction is detected, the server refers to past data and generates a response message. The input is the external reaction data, and the output is the generated response message.

[0144] Step 6:

[0145] The server analyzes the collected external reactions, creates a summary, and provides it to the user. This summary shows the proportion of positive, negative, and neutral opinions, which can be used to help create future posts. The input is the external reaction data, and the output is a summary of the analysis results.

[0146] 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.

[0147] This invention provides a system that enables information providers on social media and other platforms to manage information more appropriately and reduce risks such as online firestorms. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0148] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0149] The server uses natural language processing techniques to analyze the received posted data. This analysis assesses the potential legal and ethical risks contained in the posted content. It also uses an emotion engine to recognize the user's emotional state, for example, whether the user is stressed or relaxed.

[0150] Once the risk assessment is complete, the server generates risk points along with proposed solutions based on them. Furthermore, these solutions are adjusted to best suit the user's emotional state. For example, if the user is stressed, the explanation might be presented more gently.

[0151] Once the proposed revisions are generated, the server sends them to the terminal. The terminal then presents the proposed revisions and related risk information to the user, who then modifies their post accordingly.

[0152] After a post is published, the server continuously monitors external reactions to it. The server analyzes this reaction data and detects negative reactions that exceed the normal range (abnormal reactions). If an abnormal reaction is detected, the server creates a draft response based on past data. In this process, the tone of the draft response may be adjusted to match the emotional state of the information provider.

[0153] As a concrete example, consider a scenario where a company posts about a new product, and that post receives criticism from some customers. The system, in order to alleviate the stress on the company's public relations representative (user), activates an emotion engine to generate and quickly deliver an appropriate response in the right tone. As a result, the public relations representative can respond appropriately, contributing to maintaining the company's image.

[0154] Thus, the system of the present invention functions effectively to minimize risks on social media while taking into account the emotions of information providers.

[0155] The following describes the processing flow.

[0156] Step 1:

[0157] The user creates a post on their device. The device collects the user's input data in real time, and the sentiment engine begins to recognize the user's emotional state.

[0158] Step 2:

[0159] The user presses the "Confirm" button. The device prepares to send the posted data and the user's emotional state information to the server.

[0160] Step 3:

[0161] The server receives the posted data. The server uses natural language processing algorithms to analyze the content of the post and assess the legal and ethical risks.

[0162] Step 4:

[0163] The server uses the results of the emotion engine's analysis to consider the user's emotional state. In particular, it examines whether the user is experiencing anxiety or stress.

[0164] Step 5:

[0165] The server identifies risk points in a post and generates suggested revisions as needed. These revisions are adjusted to reflect the user's emotional state. For example, if the user is feeling stressed, more detailed explanations will be provided.

[0166] Step 6:

[0167] The server sends the risk assessment results and the adjusted proposed modifications to the terminal.

[0168] Step 7:

[0169] The terminal displays information sent from the server to the user. The user reviews the posted content based on the displayed risk assessment and suggested revisions, and makes corrections as needed.

[0170] Step 8:

[0171] The user confirms the revised post and publishes it. The device then reviews the post and uploads it to the social media platform.

[0172] Step 9:

[0173] After a post is published, the server continuously monitors external reactions. Through analysis of these reactions, it detects abnormal reactions, such as an increase in negative comments.

[0174] Step 10:

[0175] When an abnormal response is detected, the server automatically generates an appropriate response by referring to past case studies. This response is also adjusted to take the user's emotional state into consideration.

[0176] Step 11:

[0177] The server generates a response message and sends it to the terminal. The user reviews it on the terminal and makes any necessary customizations.

[0178] Step 12:

[0179] The server performs a detailed analysis of external reactions, summarizes the results, and provides them to the user. This allows the user to obtain useful feedback for future activities.

[0180] (Example 2)

[0181] 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".

[0182] When information providers disseminate information on social media and other platforms, there is a lack of means to prevent legal and ethical risks, as well as problems arising from negative reactions from external parties. Furthermore, generating revised versions or responses without considering the emotional state of the information provider can actually cause stress and inappropriate responses. Therefore, it is necessary to resolve these issues and enable information providers to disseminate information with confidence.

[0183] 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.

[0184] In this invention, the server includes means for analyzing information generated by the information sender and evaluating the risks based on legal and ethical standards, means for suggesting revisions while considering the emotional state of the information sender, and means for continuously monitoring external reactions to the information and detecting abnormal reactions. This enables responses that are appropriate to the emotional state of the information sender, resulting in more appropriate and effective risk management.

[0185] An "information analysis device" is a device that analyzes information generated by information providers and has the function of evaluating the risks from a legal and ethical standpoint.

[0186] An "information provider" refers to an individual or organization that disseminates information on social media or other platforms.

[0187] "Risk assessment" is the process of identifying the legal and ethical risks inherent in the content of information and measuring their severity.

[0188] A "revised proposal" is a suggestion to improve or adjust the information content based on a risk assessment.

[0189] "Emotional state" is an indicator used to assess the psychological health and emotional arousal of an information provider, and includes states of stress and relaxation.

[0190] An "abnormal response" is data indicating a negative external reaction that exceeds the normal range.

[0191] A "response draft" is a draft of a statement generated to facilitate appropriate communication in response to an abnormal reaction.

[0192] "Collected external reactions" refer to comments and feedback received from external sources regarding the information that has been disseminated.

[0193] This invention provides a system for reducing legal and ethical risks and appropriately managing external reactions when information providers disseminate information on social media, etc. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0194] The user creates a social media post on their device. The device then prepares to send the generated post data to the server. The user can create the post content using a standard text editor or a dedicated application.

[0195] The server uses natural language processing techniques to analyze the content of posts received from terminals. Modern deep learning models (e.g., BERT and GPT) are used as generative AI models for this analysis. This process is necessary to assess the legal and ethical risks of the posts. Furthermore, an emotion engine is utilized to recognize the user's emotional state.

[0196] After the analysis is complete, the server generates a revised version based on the risk assessment. This revised version is optimized according to the user's emotional state. For example, if the user is stressed, the revised version can be more detailed and thoughtful. The generated revised version is sent from the server to the terminal and presented to the user. The user can then revise their post based on this revised version.

[0197] As a concrete example, consider a situation where a company posts about a new product on social media, and the post is met with criticism from some customers. In this situation, the system can take into account the emotional state of the public relations representative (user) and generate and present an appropriate response, enabling a swift and effective response.

[0198] An example of a prompt might be the text, "Generate a response to critical comments on the new product."

[0199] In this way, the system provides practical means for information providers to minimize risks on social media.

[0200] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0201] Step 1:

[0202] The user creates a social media post using their device. As input, the user enters the text for the post using a text editor or a dedicated app. This data is temporarily stored on the device as a draft for later transmission to the server. Once the user completes the post and issues a send command, the process moves to the next step.

[0203] Step 2:

[0204] The terminal prepares to send user-created post data to the server. Specifically, the terminal encrypts the post data in a secure manner and structures the data according to the transmission protocol. The input is the user-prepared post data, and the output is encrypted data ready for transmission.

[0205] Step 3:

[0206] The server analyzes the posted data received from the terminal. This analysis uses natural language processing techniques, particularly generative AI models, to syntactically parse the text data and assess legal and ethical risks. The received posted data is used as input, and a risk assessment report is generated as output. Furthermore, an emotion engine is used to analyze the user's emotional state and assess their stress and relaxation levels.

[0207] Step 4:

[0208] The server generates revised suggestions based on risk assessment. This revision generation utilizes a generative AI model, which suggests optimizing the posted content while considering the user's emotional state. The inputs used are the risk assessment report and emotional state data, and the output is the revised suggestions presented to the user.

[0209] Step 5:

[0210] The server sends the generated proposed corrections and risk information to the terminal. The terminal then displays the received information to the user. Specifically, the terminal screen displays the proposed corrections and risk points, and the interface is designed to make it easy for the user to review them. The input is the proposed correction data from the server, and the output is visual feedback provided to the user.

[0211] Step 6:

[0212] After a post is published, the server monitors external reactions. It collects new comments and feedback via SNS APIs and analyzes them. The input is unstructured external data, and the output is the detection of abnormal reactions. Based on these results, the server generates appropriate response drafts and adjusts the tone according to the emotional state of the information provider.

[0213] (Application Example 2)

[0214] 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".

[0215] In online information dissemination, negative reactions and online firestorms can have a significant impact on individuals and e-commerce, making it crucial to mitigate risks beforehand. Furthermore, emotionally driven statements by information providers can lead to trouble. Against this backdrop, a system is needed that provides appropriate revisions and response templates while considering the emotional state of the information provider.

[0216] 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.

[0217] In this invention, the server includes means for evaluating the emotional state of the information sender using an information analysis device, means for making constructive suggestions based on a risk assessment of the review or comment, and means for adjusting the content of the post using the emotion analysis device. This enables information senders to send information calmly and constructively while mitigating risks.

[0218] An "information analysis device" is a device that analyzes information generated by information providers and assesses the risks based on legal and ethical standards.

[0219] A "sentiment analysis device" is a device that evaluates the emotional state of information providers and helps them adjust the content of their posts.

[0220] A "risk assessment-based revision proposal" is a revised proposal presented to the information provider in order to mitigate the legal and ethical risks inherent in the posted content.

[0221] An "abnormal reaction" refers to a negative external response to information transmitted by an information transmission device that exceeds the normal range.

[0222] A "response draft" is a proposed response generated in response to negative reactions received by an information provider, often by referring to past successful examples.

[0223] A "constructive suggestion" is a specific proposal to make the information provider's review or comment more positive and fair.

[0224] The system of this invention reduces risks and promotes constructive communication by using natural language processing and sentiment analysis technologies when users transmit information. The entire system consists of a terminal, a server, and related analysis devices.

[0225] The terminal is responsible for receiving information entered by the user and sending that data to the server. The server uses an information analysis device to perform natural language processing (for example, using the Transformers library from TENSORFLOW® or Hugging Face) and evaluate the legal and ethical risks of the entered information. In addition, an emotion analysis device determines the emotional state of the information sender and identifies emotions such as stress and anger.

[0226] Based on this analysis, the server generates revised proposals aligned with the risk assessment. These proposals aim to soften responses and include more positive and specific language. Draft response statements based on past cases are also created and presented to the information provider. This allows users to improve their information dissemination from a calm and objective perspective.

[0227] As a concrete example, suppose a user writes a review about a product. If the review contains emotional and negative content, the system assesses the risk and suggests, "Try to provide constructive feedback by specifically mentioning areas for improvement in the product." This process guides user submissions to be fair and easily accepted.

[0228] An example of a prompt message generated by the AI ​​model is, "Please enter your post content. In particular, please revise it to soften negative sentiments and make it constructive and fair." This allows users to follow the prompt and adjust their own expression.

[0229] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0230] Step 1:

[0231] Users create information posts using their devices. Here, users input the content of reviews and comments. The entered text data is temporarily stored on the device.

[0232] Step 2:

[0233] The terminal sends the entered text data to the server. The terminal transfers the data to the server using a specific protocol (e.g., HTTP or HTTPS). Text data is sent as input.

[0234] Step 3:

[0235] The server analyzes the received text data using a natural language processing model. Here, libraries such as TensorFlow and Transformers are used to analyze the text and evaluate the legal and ethical risks inherent in its content. The input text data is the target of the analysis, and a risk assessment is generated as output.

[0236] Step 4:

[0237] Simultaneously, the server uses an emotion analysis device to evaluate the user's emotional state. Based on the input text, it performs emotion analysis and detects the emotions the user may be experiencing (e.g., stress, anger, etc.). The output is the evaluation result of the emotional state.

[0238] Step 5:

[0239] The server generates suggested revisions for the user based on risk assessment and emotional state assessment. Using a generation AI model, it specifically points out areas for improvement in the posted content and encourages constructive feedback. The output is the proposed revisions.

[0240] Step 6:

[0241] The server sends the generated revised draft and response draft to the terminal. The revised draft specifically outlines how the information to be communicated should be improved. The terminal receives the output from the server and displays the suggestions to the user.

[0242] Step 7:

[0243] Users adjust and finalize their information based on the suggested revisions displayed on their devices. During this process, they refer to the prompt message, "Please enter your post content. In particular, please revise it to mitigate negative sentiment and ensure it is constructive and fair." The final revised content is then disseminated externally in a way that reduces risk.

[0244] 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.

[0245] Data generation model 58 is a type of 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.

[0246] 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.

[0247] [Second Embodiment]

[0248] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0249] 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.

[0250] 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).

[0251] 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.

[0252] 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.

[0253] 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).

[0254] 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.

[0255] 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.

[0256] 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.

[0257] 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.

[0258] 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.

[0259] 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".

[0260] This invention provides a system that checks information before posting it to social media and other platforms to prevent problems from occurring. This system mainly includes an information analysis device, a terminal, and a server.

[0261] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0262] When the server receives submitted data, it analyzes the data using natural language processing techniques. The purpose of the analysis is to check whether the submitted content poses any legal or ethical risks. For example, it identifies whether it contains racist or potentially misleading language. Based on these results, the server lists the potential risks present in the submission and generates specific suggestions for revision.

[0263] Next, the server sends the generated risk assessment and proposed corrections to the user's device. On the device, this information is displayed visually to the user, allowing them to modify their post considering the identified risks.

[0264] After a post is published, the server continuously monitors external reactions to that post. The server collects the reactions and determines if they are abnormal. For example, a sudden surge in critical comments or a topic becoming a major topic of discussion in a short period of time may be considered a sign of an impending online firestorm.

[0265] If a controversy is detected, the server quickly references similar past case studies and generates a draft apology. This draft apology is crafted with sincerity and specificity in mind and is promptly provided to the user.

[0266] Furthermore, the server analyzes the collected external reactions and generates a summary of user feedback. This summary shows the proportion of positive, negative, and neutral opinions and includes specific suggestions for improvement that will help users with future posts.

[0267] As a concrete example, consider a scenario where a public relations representative for a company posts an introduction to a new product. Suppose this post contains an inappropriate comparison to a competing product. The system detects this and suggests revisions from an ethical standpoint. The PR representative adjusts the post according to the suggestions, implements preventative measures, and then publishes it. If, after publication, criticism from the competing company increases, the server provides a suitable apology draft and also offers suggestions for improving future posts.

[0268] In this way, the system of the present invention supports information providers in communicating smoothly while minimizing risks.

[0269] The following describes the processing flow.

[0270] Step 1:

[0271] The user creates a post on their device. The device receives the user input as text data.

[0272] Step 2:

[0273] The device presses the "Confirm" button. Afterward, the device prepares to send the posted data to the server.

[0274] Step 3:

[0275] The server receives the submitted data. The server executes a natural language processing algorithm to analyze the information.

[0276] Step 4:

[0277] The server evaluates the content of posts based on legal and ethical standards. If there are any risk points, they are identified and highlighted.

[0278] Step 5:

[0279] The server generates revised plans along with the evaluation results. These plans include specific suggestions for mitigating risks.

[0280] Step 6:

[0281] The server sends the generated evaluation results and suggested corrections to the terminal. The results are then displayed to the user on the terminal.

[0282] Step 7:

[0283] The user modifies the post through the terminal. Appropriate modifications are made based on the pointed-out risks, and the post is finally finalized.

[0284] Step 8:

[0285] The user publishes the post, and the terminal uploads the information to the SNS platform.

[0286] Step 9:

[0287] The server continues to monitor the external reactions to the post after publication. In particular, it pays attention to criticism and a sudden increase in the number of comments.

[0288] Step 10:

[0289] When the server detects an abnormal reaction, it automatically generates an appropriate response text by referring to past case studies.

[0290] Step 11:

[0291] The server sends the response text generated to the terminal. The user checks it and customizes it if necessary.

[0292] Step 12:

[0293] The server analyzes the external reaction comments and summarizes the positive, negative, and neutral evaluations.

[0294] Step 13:

[0295] The server sends the summary and improvement suggestions to the terminal. Thus, the user can use them for future post creation.

[0296] (Example 1)

[0297] Next, 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".

[0298] There is a lack of solutions to prevent legal and ethical problems when information providers publish information on social media and other online platforms, and to respond quickly and appropriately to reactions after publication. Furthermore, in a situation where unexpected external reactions and criticisms to published information are rapidly increasing, there is a need for a mechanism that efficiently presents countermeasures and reduces the burden on information providers.

[0299] 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.

[0300] In this invention, the server includes means for analyzing information generated by the information sender using natural language processing technology and evaluating risks based on legal and ethical standards; means for generating revised versions based on prompt sentences using a generation AI model and presenting them to the information sender; and means for continuously monitoring external reactions and detecting abnormal reactions. As a result, the information sender can grasp risks in advance, revise their expression to an appropriate one, and respond quickly to reactions after publication.

[0301] An "information analysis device" is a device that analyzes data generated by information providers and has the function of evaluating risks based on legal and ethical standards.

[0302] "Natural language processing technology" is a method that understands text data semantically and grammatically, and analyzes the ambiguity and complexity unique to human language.

[0303] A "generative AI model" is an artificial intelligence model that learns from past data and generates new text or suggestions.

[0304] A "prompt" is a sentence of instruction or question input to an AI model to obtain a specific response or output.

[0305] "Risk assessment" is a process of judging the possibility that the content may cause legal or ethical problems before the information is made public.

[0306] "Amendment" refers to the specific revised content proposed to change the transmitted information into a more appropriate and problem-free expression.

[0307] "Abnormal reaction" refers to a reaction indicating critical or emotional feedback from the outside that exceeds the normal range.

[0308] "Response text" is the text provided for the information sender to quickly respond to the detected abnormal reaction.

[0309] "Summary" is the content that shortens the entire external reaction collected, extracts important points, and provides them to the information sender.

[0310] This invention is a system that minimizes the risks when an information sender posts information on an online platform. The system mainly consists of a server, a terminal, and a user.

[0311] Server

[0312] The server receives the posted data transmitted from the user's terminal. Analyze this data using natural language processing technology. Specifically, use Python's natural language processing libraries (e.g., spaCy, NLTK) to evaluate the legal and ethical risks of the posted content. The server utilizes a generative AI model to generate amendments based on the provided prompt text. This amendment is notified to the user through the terminal. Also, the server monitors external reactions even after the post is made public and detects abnormal reactions. If an abnormal reaction is detected, a response text is generated referring to past data and provided to the user promptly.

[0313] Terminal

[0314] The device collects user-entered posting data and sends it to the server. It then receives suggested revisions and risk assessments from the server, which are displayed visually to the user. This allows the user to review their posts and make revisions as needed.

[0315] User

[0316] Users can use this system to create their posts, reviewing risk assessments and suggested revisions along the way. For example, suppose a corporate public relations representative creating a new product description includes ethically problematic language. In this case, the system will detect the language and suggest revisions. The user can then adjust the content based on the feedback and make an appropriate post.

[0317] For example, a prompt might include instructions such as, "Please revise this expression to be neutral." Based on these instructions, the AI ​​generates specific revisions to mitigate risk. This allows users to communicate information more safely and effectively.

[0318] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0319] Step 1:

[0320] Users create post data on their devices. This post data may include text, images, or videos. The information entered by the user is first temporarily stored on the device. At this time, a post text is generated based on the user's goals and intentions.

[0321] Step 2:

[0322] The terminal prepares to send the created post data to the server. The terminal converts the data to the appropriate format and sends it to the server using the encrypted HTTP protocol, ensuring data security over network communication.

[0323] Step 3:

[0324] The server receives posted data sent from the terminal. Based on the received data, it analyzes the data using natural language processing techniques. This analysis process breaks down the text, extracts keywords and phrases, and assesses the legal or ethical risks. For example, it calculates a sensitivity score for each sentence and detects inappropriate phrases.

[0325] Step 4:

[0326] Based on the analysis results, the server generates suggested corrections for areas where risks were detected. Specifically, it uses a generation AI model to receive the prompt "Please correct this expression to be neutral" and generates a new expression. During this process, data is input to the model and suggested corrections are output.

[0327] Step 5:

[0328] The server sends the generated revised proposals and risk assessments to the terminal. The transmitted data is formatted in a way that is easily understandable to the user.

[0329] Step 6:

[0330] The terminal displays the risk assessment and proposed corrections received from the server to the user. Using a visual GUI, the terminal highlights risk areas and displays proposed corrections in pop-ups or side panels. This allows users to easily review their feedback.

[0331] Step 7:

[0332] Users review their posts based on the suggested revisions and make corrections as needed. The revised posts are then reviewed and ready for publication.

[0333] Step 8:

[0334] The server continuously monitors external reactions to a post even after it has been published. The server collects data in real time and analyzes whether any abnormal reactions are occurring. This analysis includes comment frequency and sentiment analysis.

[0335] Step 9:

[0336] If the server detects an abnormal response, it will generate a suggested response based on similar past cases and provide it to the user. This suggested response will include apologies and suggestions for corrections to help the user respond quickly.

[0337] (Application Example 1)

[0338] 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 glasses 214 will be referred to as the "terminal."

[0339] In modern digital communication, information dissemination via social media is commonplace, but at the same time, inappropriate posts that carry legal and ethical risks have become a problem. Therefore, it is necessary to detect these risks in advance and address them appropriately. However, conventional systems have difficulty responding in real time, and there are challenges in making highly accurate risk assessments and corrective suggestions.

[0340] 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.

[0341] In this invention, the server includes means for analyzing information generated by information senders using an information analysis device and evaluating the risks based on legal and ethical standards; means for enhancing natural language processing using a generative AI model to analyze posted information and generate proposed revisions; and means for dynamically adjusting input prompt sentences to the generative AI model to improve analysis accuracy. This makes it possible to evaluate the risks in SNS posts in advance with high accuracy and to quickly present appropriate revision suggestions.

[0342] An "information analysis device" is a device that analyzes data generated by information providers and assesses risks based on legal and ethical standards.

[0343] A "generative AI model" is an artificial intelligence model designed for advanced natural language processing, and is used for analyzing posted information and generating suggested revisions.

[0344] A "prompt sentence" is a sentence used as input to a generative AI model, and it is dynamically adjusted to improve the accuracy of the analysis.

[0345] "External reactions" refer to comments and feedback received from third parties regarding information made public through information dissemination devices.

[0346] An "abnormal reaction" refers to a characteristic response that is not typical, such as a sudden surge in critical comments or rapid growth in public attention.

[0347] A "response draft" is a draft text, generated based on past data and successful cases, intended to be an apology or response, provided to the information provider when an abnormal reaction is detected.

[0348] A "revision proposal" is a specific suggestion for correction presented to the information provider when legal or ethical risks are detected in the content of their post.

[0349] A "summary" is information that shows the proportion of positive, negative, and neutral opinions, compiled to provide information providers with the results of an analysis of collected external reactions.

[0350] To implement this invention, the server must be equipped with an information analysis device, a generative AI model, and a prompt sentence adjustment system. The server receives posted data from information senders and performs analysis using natural language processing technology. This analysis utilizes a generative AI model to perform a highly accurate risk assessment of the text. At this time, the prompt sentences input to the generative AI model are dynamically adjusted to improve the accuracy of the analysis.

[0351] Based on the risk assessment generated from the analysis results, the server presents the user with suggested revisions on their device. The user can then adjust their post according to the suggested revisions displayed on their device and resubmit it. This feature enables the dissemination of information that complies with legal and ethical standards.

[0352] Furthermore, after the information is published, the server continuously monitors external reactions. If an abnormal reaction is detected, the server uses a generative AI model, based on past cases, to generate an appropriate response and provide it to the user. In addition, the server analyzes the collected external reactions and presents them to the information provider as a summary, which can be used to improve future posts.

[0353] For example, if a user is about to post an opinion about a new product, this system can pre-scan the content, identify potentially misleading parts, and suggest neutral wording.

[0354] An example of a prompt message might be: "Regarding a new social media post, it says: 'Our new product is the best on the market and overwhelmingly superior to the competition.' Please assess whether this post contains any ethical or legal risks and provide suggestions for revisions if necessary."

[0355] In this way, the entire system enables information providers to consistently manage risks both before and after the event, and to facilitate smooth communication.

[0356] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0357] Step 1:

[0358] Users create post data using their devices. This post data is prepared to be sent from the device to the server for sharing on the SNS platform. The input is free-form text data created by the user, and the output is text data ready to be sent to the server.

[0359] Step 2:

[0360] The server uses a generative AI model to perform natural language processing to process the received posted data. The server prepares prompt sentences and inputs these prompt sentences into the AI ​​model for analysis. The input consists of the user's posted text and prompt sentences, and the output is an assessment of legal and ethical risks.

[0361] Step 3:

[0362] Based on the generated risk assessment, the server creates proposed revisions to the posted content. These revisions include specific suggestions for improving particular parts of the information. The input is the result of the risk assessment, and the output is the proposed revisions.

[0363] Step 4:

[0364] The suggested revisions are displayed on the terminal's user interface and presented to the user. The user reviews them and modifies the post as needed. The input is the suggested revisions sent from the server, and the output is the updated post after the user has made the necessary corrections.

[0365] Step 5:

[0366] After the information is officially released, the server continuously monitors external reactions on the SNS platform. If an abnormal reaction is detected, the server refers to past data and generates a response message. The input is the external reaction data, and the output is the generated response message.

[0367] Step 6:

[0368] The server analyzes the collected external reactions, creates a summary, and provides it to the user. This summary shows the proportion of positive, negative, and neutral opinions, which can be used to help create future posts. The input is the external reaction data, and the output is a summary of the analysis results.

[0369] 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.

[0370] This invention provides a system that enables information providers on social media and other platforms to manage information more appropriately and reduce risks such as online firestorms. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0371] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0372] The server uses natural language processing techniques to analyze the received posted data. This analysis assesses the potential legal and ethical risks contained in the posted content. It also uses an emotion engine to recognize the user's emotional state, for example, whether the user is stressed or relaxed.

[0373] Once the risk assessment is complete, the server generates risk points along with proposed solutions based on them. Furthermore, these solutions are adjusted to best suit the user's emotional state. For example, if the user is stressed, the explanation might be presented more gently.

[0374] Once the proposed revisions are generated, the server sends them to the terminal. The terminal then presents the proposed revisions and related risk information to the user, who then modifies their post accordingly.

[0375] After a post is published, the server continuously monitors external reactions to it. The server analyzes this reaction data and detects negative reactions that exceed the normal range (abnormal reactions). If an abnormal reaction is detected, the server creates a draft response based on past data. In this process, the tone of the draft response may be adjusted to match the emotional state of the information provider.

[0376] As a concrete example, consider a scenario where a company posts about a new product, and that post receives criticism from some customers. The system, in order to alleviate the stress on the company's public relations representative (user), activates an emotion engine to generate and quickly deliver an appropriate response in the right tone. As a result, the public relations representative can respond appropriately, contributing to maintaining the company's image.

[0377] Thus, the system of the present invention functions effectively to minimize risks on social media while taking into account the emotions of information providers.

[0378] The following describes the processing flow.

[0379] Step 1:

[0380] The user creates a post on their device. The device collects the user's input data in real time, and the sentiment engine begins to recognize the user's emotional state.

[0381] Step 2:

[0382] The user presses the "Confirm" button. The device prepares to send the posted data and the user's emotional state information to the server.

[0383] Step 3:

[0384] The server receives the posted data. The server uses natural language processing algorithms to analyze the content of the post and assess the legal and ethical risks.

[0385] Step 4:

[0386] The server uses the results of the emotion engine's analysis to consider the user's emotional state. In particular, it examines whether the user is experiencing anxiety or stress.

[0387] Step 5:

[0388] The server identifies risk points in a post and generates suggested revisions as needed. These revisions are adjusted to reflect the user's emotional state. For example, if the user is feeling stressed, more detailed explanations will be provided.

[0389] Step 6:

[0390] The server sends the risk assessment results and the adjusted proposed modifications to the terminal.

[0391] Step 7:

[0392] The terminal displays information sent from the server to the user. The user reviews the posted content based on the displayed risk assessment and suggested revisions, and makes corrections as needed.

[0393] Step 8:

[0394] The user confirms the revised post and publishes it. The device then reviews the post and uploads it to the social media platform.

[0395] Step 9:

[0396] After a post is published, the server continuously monitors external reactions. Through analysis of these reactions, it detects abnormal reactions, such as an increase in negative comments.

[0397] Step 10:

[0398] When an abnormal response is detected, the server automatically generates an appropriate response by referring to past case studies. This response is also adjusted to take the user's emotional state into consideration.

[0399] Step 11:

[0400] The server generates a response message and sends it to the terminal. The user reviews it on the terminal and makes any necessary customizations.

[0401] Step 12:

[0402] The server performs a detailed analysis of external reactions, summarizes the results, and provides them to the user. This allows the user to obtain useful feedback for future activities.

[0403] (Example 2)

[0404] 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".

[0405] When information providers disseminate information on social media and other platforms, there is a lack of means to prevent legal and ethical risks, as well as problems arising from negative reactions from external parties. Furthermore, generating revised versions or responses without considering the emotional state of the information provider can actually cause stress and inappropriate responses. Therefore, it is necessary to resolve these issues and enable information providers to disseminate information with confidence.

[0406] 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.

[0407] In this invention, the server includes means for analyzing information generated by the information sender and evaluating the risks based on legal and ethical standards, means for suggesting revisions while considering the emotional state of the information sender, and means for continuously monitoring external reactions to the information and detecting abnormal reactions. This enables responses that are appropriate to the emotional state of the information sender, resulting in more appropriate and effective risk management.

[0408] An "information analysis device" is a device that analyzes information generated by information providers and has the function of evaluating the risks from a legal and ethical standpoint.

[0409] An "information provider" refers to an individual or organization that disseminates information on social media or other platforms.

[0410] "Risk assessment" is the process of identifying the legal and ethical risks inherent in the content of information and measuring their severity.

[0411] A "revised proposal" is a suggestion to improve or adjust the information content based on a risk assessment.

[0412] "Emotional state" is an indicator used to assess the psychological health and emotional arousal of an information provider, and includes states of stress and relaxation.

[0413] An "abnormal response" is data indicating a negative external reaction that exceeds the normal range.

[0414] A "response draft" is a draft of a statement generated to facilitate appropriate communication in response to an abnormal reaction.

[0415] "Collected external reactions" refer to comments and feedback received from external sources regarding the information that has been disseminated.

[0416] This invention provides a system for reducing legal and ethical risks and appropriately managing external reactions when information providers disseminate information on social media, etc. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0417] The user creates a social media post on their device. The device then prepares to send the generated post data to the server. The user can create the post content using a standard text editor or a dedicated application.

[0418] The server uses natural language processing techniques to analyze the content of posts received from terminals. Modern deep learning models (e.g., BERT and GPT) are used as generative AI models for this analysis. This process is necessary to assess the legal and ethical risks of the posts. Furthermore, an emotion engine is utilized to recognize the user's emotional state.

[0419] After the analysis is complete, the server generates a revised version based on the risk assessment. This revised version is optimized according to the user's emotional state. For example, if the user is stressed, the revised version can be more detailed and thoughtful. The generated revised version is sent from the server to the terminal and presented to the user. The user can then revise their post based on this revised version.

[0420] As a concrete example, consider a situation where a company posts about a new product on social media, and the post is met with criticism from some customers. In this situation, the system can take into account the emotional state of the public relations representative (user) and generate and present an appropriate response, enabling a swift and effective response.

[0421] An example of a prompt might be the text, "Generate a response to critical comments on the new product."

[0422] In this way, the system provides practical means for information providers to minimize risks on social media.

[0423] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0424] Step 1:

[0425] The user creates a social media post using their device. As input, the user enters the text for the post using a text editor or a dedicated app. This data is temporarily stored on the device as a draft for later transmission to the server. Once the user completes the post and issues a send command, the process moves to the next step.

[0426] Step 2:

[0427] The terminal prepares to send user-created post data to the server. Specifically, the terminal encrypts the post data in a secure manner and structures the data according to the transmission protocol. The input is the user-prepared post data, and the output is encrypted data ready for transmission.

[0428] Step 3:

[0429] The server analyzes the posted data received from the terminal. This analysis uses natural language processing techniques, particularly generative AI models, to syntactically parse the text data and assess legal and ethical risks. The received posted data is used as input, and a risk assessment report is generated as output. Furthermore, an emotion engine is used to analyze the user's emotional state and assess their stress and relaxation levels.

[0430] Step 4:

[0431] The server generates revised suggestions based on risk assessment. This revision generation utilizes a generative AI model, which suggests optimizing the posted content while considering the user's emotional state. The inputs used are the risk assessment report and emotional state data, and the output is the revised suggestions presented to the user.

[0432] Step 5:

[0433] The server sends the generated proposed corrections and risk information to the terminal. The terminal then displays the received information to the user. Specifically, the terminal screen displays the proposed corrections and risk points, and the interface is designed to make it easy for the user to review them. The input is the proposed correction data from the server, and the output is visual feedback provided to the user.

[0434] Step 6:

[0435] After a post is published, the server monitors external reactions. It collects new comments and feedback via SNS APIs and analyzes them. The input is unstructured external data, and the output is the detection of abnormal reactions. Based on these results, the server generates appropriate response drafts and adjusts the tone according to the emotional state of the information provider.

[0436] (Application Example 2)

[0437] 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."

[0438] In online information dissemination, negative reactions and online firestorms can have a significant impact on individuals and e-commerce, making it crucial to mitigate risks beforehand. Furthermore, emotionally driven statements by information providers can lead to trouble. Against this backdrop, a system is needed that provides appropriate revisions and response templates while considering the emotional state of the information provider.

[0439] 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.

[0440] In this invention, the server includes means for evaluating the emotional state of the information sender using an information analysis device, means for making constructive suggestions based on a risk assessment of the review or comment, and means for adjusting the content of the post using the emotion analysis device. This enables information senders to send information calmly and constructively while mitigating risks.

[0441] An "information analysis device" is a device that analyzes information generated by information providers and assesses the risks based on legal and ethical standards.

[0442] A "sentiment analysis device" is a device that evaluates the emotional state of information providers and helps them adjust the content of their posts.

[0443] A "risk assessment-based revision proposal" is a revised proposal presented to the information provider in order to mitigate the legal and ethical risks inherent in the posted content.

[0444] An "abnormal reaction" refers to a negative external response to information transmitted by an information transmission device that exceeds the normal range.

[0445] A "response draft" is a proposed response generated in response to negative reactions received by an information provider, often by referring to past successful examples.

[0446] A "constructive suggestion" is a specific proposal to make the information provider's review or comment more positive and fair.

[0447] The system of this invention reduces risks and promotes constructive communication by using natural language processing and sentiment analysis technologies when users transmit information. The entire system consists of a terminal, a server, and related analysis devices.

[0448] The terminal is responsible for receiving information entered by the user and sending that data to the server. The server uses an information analysis device to perform natural language processing (for example, using TensorFlow or the Transformers library from Hugging Face) and evaluate the legal and ethical risks of the entered information. In addition, an emotion analysis device determines the emotional state of the information sender and identifies emotions such as stress and anger.

[0449] Based on this analysis, the server generates revised proposals aligned with the risk assessment. These proposals aim to soften responses and include more positive and specific language. Draft response statements based on past cases are also created and presented to the information provider. This allows users to improve their information dissemination from a calm and objective perspective.

[0450] As a concrete example, suppose a user writes a review about a product. If the review contains emotional and negative content, the system assesses the risk and suggests, "Try to provide constructive feedback by specifically mentioning areas for improvement in the product." This process guides user submissions to be fair and easily accepted.

[0451] An example of a prompt message generated by the AI ​​model is, "Please enter your post content. In particular, please revise it to soften negative sentiments and make it constructive and fair." This allows users to follow the prompt and adjust their own expression.

[0452] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0453] Step 1:

[0454] Users create information posts using their devices. Here, users input the content of reviews and comments. The entered text data is temporarily stored on the device.

[0455] Step 2:

[0456] The terminal sends the entered text data to the server. The terminal transfers the data to the server using a specific protocol (e.g., HTTP or HTTPS). Text data is sent as input.

[0457] Step 3:

[0458] The server analyzes the received text data using a natural language processing model. Here, libraries such as TensorFlow and Transformers are used to analyze the text and evaluate the legal and ethical risks inherent in its content. The input text data is the target of the analysis, and a risk assessment is generated as output.

[0459] Step 4:

[0460] Simultaneously, the server uses an emotion analysis device to evaluate the user's emotional state. Based on the input text, it performs emotion analysis and detects the emotions the user may be experiencing (e.g., stress, anger, etc.). The output is the evaluation result of the emotional state.

[0461] Step 5:

[0462] The server generates suggested revisions for the user based on risk assessment and emotional state assessment. Using a generation AI model, it specifically points out areas for improvement in the posted content and encourages constructive feedback. The output is the proposed revisions.

[0463] Step 6:

[0464] The server sends the generated revised draft and response draft to the terminal. The revised draft specifically outlines how the information to be communicated should be improved. The terminal receives the output from the server and displays the suggestions to the user.

[0465] Step 7:

[0466] Users adjust and finalize their information based on the suggested revisions displayed on their devices. During this process, they refer to the prompt message, "Please enter your post content. In particular, please revise it to mitigate negative sentiment and ensure it is constructive and fair." The final revised content is then disseminated externally in a way that reduces risk.

[0467] 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.

[0468] 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.

[0469] 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.

[0470] [Third Embodiment]

[0471] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0472] 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.

[0473] 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).

[0474] 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.

[0475] 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.

[0476] 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).

[0477] 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.

[0478] 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.

[0479] 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.

[0480] 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.

[0481] 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.

[0482] 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".

[0483] This invention provides a system that checks information before posting it to social media and other platforms to prevent problems from occurring. This system mainly includes an information analysis device, a terminal, and a server.

[0484] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0485] When the server receives submitted data, it analyzes the data using natural language processing techniques. The purpose of the analysis is to check whether the submitted content poses any legal or ethical risks. For example, it identifies whether it contains racist or potentially misleading language. Based on these results, the server lists the potential risks present in the submission and generates specific suggestions for revision.

[0486] Next, the server sends the generated risk assessment and proposed corrections to the user's device. On the device, this information is displayed visually to the user, allowing them to modify their post considering the identified risks.

[0487] After a post is published, the server continuously monitors external reactions to that post. The server collects the reactions and determines if they are abnormal. For example, a sudden surge in critical comments or a topic becoming a major topic of discussion in a short period of time may be considered a sign of an impending online firestorm.

[0488] If a controversy is detected, the server quickly references similar past case studies and generates a draft apology. This draft apology is crafted with sincerity and specificity in mind and is promptly provided to the user.

[0489] Furthermore, the server analyzes the collected external reactions and generates a summary of user feedback. This summary shows the proportion of positive, negative, and neutral opinions and includes specific suggestions for improvement that will help users with future posts.

[0490] As a concrete example, consider a scenario where a public relations representative for a company posts an introduction to a new product. Suppose this post contains an inappropriate comparison to a competing product. The system detects this and suggests revisions from an ethical standpoint. The PR representative adjusts the post according to the suggestions, implements preventative measures, and then publishes it. If, after publication, criticism from the competing company increases, the server provides a suitable apology draft and also offers suggestions for improving future posts.

[0491] In this way, the system of the present invention supports information providers in communicating smoothly while minimizing risks.

[0492] The following describes the processing flow.

[0493] Step 1:

[0494] The user creates a post on their device. The device receives the user input as text data.

[0495] Step 2:

[0496] The device presses the "Confirm" button. Afterward, the device prepares to send the posted data to the server.

[0497] Step 3:

[0498] The server receives the submitted data. The server executes a natural language processing algorithm to analyze the information.

[0499] Step 4:

[0500] The server evaluates the content of posts based on legal and ethical standards. If there are any risk points, they are identified and highlighted.

[0501] Step 5:

[0502] The server generates revised plans along with the evaluation results. These plans include specific suggestions for mitigating risks.

[0503] Step 6:

[0504] The server sends the generated evaluation results and suggested corrections to the terminal. The results are then displayed to the user on the terminal.

[0505] Step 7:

[0506] Users edit their posts through their devices. They make corrections as needed based on the identified risks and then finalize the post.

[0507] Step 8:

[0508] The user publishes a post, and the device uploads that information to the social networking platform.

[0509] Step 9:

[0510] The server continuously monitors external reactions to posts after they are published, paying particular attention to criticism and sudden increases in the number of comments.

[0511] Step 10:

[0512] When the server detects an abnormal response, it automatically generates an appropriate response by referring to past case studies.

[0513] Step 11:

[0514] The server generates a draft response message and sends it to the terminal. The user reviews it and customizes it as needed.

[0515] Step 12:

[0516] The server analyzes external reaction comments and summarizes them as positive, negative, or neutral.

[0517] Step 13:

[0518] The server sends a summary and improvement suggestions to the user's device, which helps them in creating future posts.

[0519] (Example 1)

[0520] 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."

[0521] There is a lack of solutions to prevent legal and ethical problems when information providers publish information on social media and other online platforms, and to respond quickly and appropriately to reactions after publication. Furthermore, in a situation where unexpected external reactions and criticisms to published information are rapidly increasing, there is a need for a mechanism that efficiently presents countermeasures and reduces the burden on information providers.

[0522] 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.

[0523] In this invention, the server includes means for analyzing information generated by the information sender using natural language processing technology and evaluating risks based on legal and ethical standards; means for generating revised versions based on prompt sentences using a generation AI model and presenting them to the information sender; and means for continuously monitoring external reactions and detecting abnormal reactions. As a result, the information sender can grasp risks in advance, revise their expression to an appropriate one, and respond quickly to reactions after publication.

[0524] An "information analysis device" is a device that analyzes data generated by information providers and has the function of evaluating risks based on legal and ethical standards.

[0525] "Natural language processing technology" is a method that understands text data semantically and grammatically, and analyzes the ambiguity and complexity unique to human language.

[0526] A "generative AI model" is an artificial intelligence model that learns from past data and generates new text or suggestions.

[0527] A "prompt" is a sentence of instruction or question input to an AI model to obtain a specific response or output.

[0528] "Risk assessment" is the process of determining whether the content of information may cause legal or ethical problems before it is made public.

[0529] A "proposal for revision" refers to specific changes proposed to make the information being disseminated more appropriate and problem-free.

[0530] An "abnormal reaction" refers to a response that exhibits critical or emotional feedback from an external source that exceeds the normal range.

[0531] A "draft response" is a document provided by the information provider to enable them to respond quickly to detected abnormal reactions.

[0532] A "summary" is a shortened version of collected external responses, extracting the key points and providing them to the information provider.

[0533] This invention is a system that minimizes the risks for information providers when posting information on online platforms. The system mainly consists of a server, terminals, and users.

[0534] server

[0535] The server receives posted data sent from the user's terminal. This data is analyzed using natural language processing techniques. Specifically, Python natural language processing libraries (e.g., spaCy, NLTK) are used to evaluate the legal and ethical risks of the posted content. The server utilizes a generative AI model to generate suggested revisions based on the provided prompt text. These suggested revisions are notified to the user via their terminal. The server also monitors external reactions after the post is published and detects any abnormal reactions. If an abnormal reaction is detected, it generates a suggested response based on past data and provides it to the user promptly.

[0536] terminal

[0537] The device collects user-entered posting data and sends it to the server. It then receives suggested revisions and risk assessments from the server, which are displayed visually to the user. This allows the user to review their posts and make revisions as needed.

[0538] User

[0539] Users can use this system to create their posts, reviewing risk assessments and suggested revisions along the way. For example, suppose a corporate public relations representative creating a new product description includes ethically problematic language. In this case, the system will detect the language and suggest revisions. The user can then adjust the content based on the feedback and make an appropriate post.

[0540] For example, a prompt might include instructions such as, "Please revise this expression to be neutral." Based on these instructions, the AI ​​generates specific revisions to mitigate risk. This allows users to communicate information more safely and effectively.

[0541] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0542] Step 1:

[0543] Users create post data on their devices. This post data may include text, images, or videos. The information entered by the user is first temporarily stored on the device. At this time, a post text is generated based on the user's goals and intentions.

[0544] Step 2:

[0545] The terminal prepares to send the created post data to the server. The terminal converts the data to the appropriate format and sends it to the server using the encrypted HTTP protocol, ensuring data security over network communication.

[0546] Step 3:

[0547] The server receives posted data sent from the terminal. Based on the received data, it analyzes the data using natural language processing techniques. This analysis process breaks down the text, extracts keywords and phrases, and assesses the legal or ethical risks. For example, it calculates a sensitivity score for each sentence and detects inappropriate phrases.

[0548] Step 4:

[0549] Based on the analysis results, the server generates suggested corrections for areas where risks were detected. Specifically, it uses a generation AI model to receive the prompt "Please correct this expression to be neutral" and generates a new expression. During this process, data is input to the model and suggested corrections are output.

[0550] Step 5:

[0551] The server sends the generated revised proposals and risk assessments to the terminal. The transmitted data is formatted in a way that is easily understandable to the user.

[0552] Step 6:

[0553] The terminal displays the risk assessment and proposed corrections received from the server to the user. Using a visual GUI, the terminal highlights risk areas and displays proposed corrections in pop-ups or side panels. This allows users to easily review their feedback.

[0554] Step 7:

[0555] Users review their posts based on the suggested revisions and make corrections as needed. The revised posts are then reviewed and ready for publication.

[0556] Step 8:

[0557] The server continuously monitors external reactions to a post even after it has been published. The server collects data in real time and analyzes whether any abnormal reactions are occurring. This analysis includes comment frequency and sentiment analysis.

[0558] Step 9:

[0559] If the server detects an abnormal response, it will generate a suggested response based on similar past cases and provide it to the user. This suggested response will include apologies and suggestions for corrections to help the user respond quickly.

[0560] (Application Example 1)

[0561] 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."

[0562] In modern digital communication, information dissemination via social media is commonplace, but at the same time, inappropriate posts that carry legal and ethical risks have become a problem. Therefore, it is necessary to detect these risks in advance and address them appropriately. However, conventional systems have difficulty responding in real time, and there are challenges in making highly accurate risk assessments and corrective suggestions.

[0563] 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.

[0564] In this invention, the server includes means for analyzing information generated by information senders using an information analysis device and evaluating the risks based on legal and ethical standards; means for enhancing natural language processing using a generative AI model to analyze posted information and generate proposed revisions; and means for dynamically adjusting input prompt sentences to the generative AI model to improve analysis accuracy. This makes it possible to evaluate the risks in SNS posts in advance with high accuracy and to quickly present appropriate revision suggestions.

[0565] An "information analysis device" is a device that analyzes data generated by information providers and assesses risks based on legal and ethical standards.

[0566] A "generative AI model" is an artificial intelligence model designed for advanced natural language processing, and is used for analyzing posted information and generating suggested revisions.

[0567] A "prompt sentence" is a sentence used as input to a generative AI model, and it is dynamically adjusted to improve the accuracy of the analysis.

[0568] "External reactions" refer to comments and feedback received from third parties regarding information made public through information dissemination devices.

[0569] An "abnormal reaction" refers to a characteristic response that is not typical, such as a sudden surge in critical comments or rapid growth in public attention.

[0570] A "response draft" is a draft text, generated based on past data and successful cases, intended to be an apology or response, provided to the information provider when an abnormal reaction is detected.

[0571] A "revision proposal" is a specific suggestion for correction presented to the information provider when legal or ethical risks are detected in the content of their post.

[0572] A "summary" is information that shows the proportion of positive, negative, and neutral opinions, compiled to provide information providers with the results of an analysis of collected external reactions.

[0573] To implement this invention, the server must be equipped with an information analysis device, a generative AI model, and a prompt sentence adjustment system. The server receives posted data from information senders and performs analysis using natural language processing technology. This analysis utilizes a generative AI model to perform a highly accurate risk assessment of the text. At this time, the prompt sentences input to the generative AI model are dynamically adjusted to improve the accuracy of the analysis.

[0574] Based on the risk assessment generated from the analysis results, the server presents the user with suggested revisions on their device. The user can then adjust their post according to the suggested revisions displayed on their device and resubmit it. This feature enables the dissemination of information that complies with legal and ethical standards.

[0575] Furthermore, after the information is published, the server continuously monitors external reactions. If an abnormal reaction is detected, the server uses a generative AI model, based on past cases, to generate an appropriate response and provide it to the user. In addition, the server analyzes the collected external reactions and presents them to the information provider as a summary, which can be used to improve future posts.

[0576] For example, if a user is about to post an opinion about a new product, this system can pre-scan the content, identify potentially misleading parts, and suggest neutral wording.

[0577] An example of a prompt message might be: "Regarding a new social media post, it says: 'Our new product is the best on the market and overwhelmingly superior to the competition.' Please assess whether this post contains any ethical or legal risks and provide suggestions for revisions if necessary."

[0578] In this way, the entire system enables information providers to consistently manage risks both before and after the event, and to facilitate smooth communication.

[0579] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0580] Step 1:

[0581] Users create post data using their devices. This post data is prepared to be sent from the device to the server for sharing on the SNS platform. The input is free-form text data created by the user, and the output is text data ready to be sent to the server.

[0582] Step 2:

[0583] The server uses a generative AI model to perform natural language processing to process the received posted data. The server prepares prompt sentences and inputs these prompt sentences into the AI ​​model for analysis. The input consists of the user's posted text and prompt sentences, and the output is an assessment of legal and ethical risks.

[0584] Step 3:

[0585] Based on the generated risk assessment, the server creates proposed revisions to the posted content. These revisions include specific suggestions for improving particular parts of the information. The input is the result of the risk assessment, and the output is the proposed revisions.

[0586] Step 4:

[0587] The suggested revisions are displayed on the terminal's user interface and presented to the user. The user reviews them and modifies the post as needed. The input is the suggested revisions sent from the server, and the output is the updated post after the user has made the necessary corrections.

[0588] Step 5:

[0589] After the information is officially released, the server continuously monitors external reactions on the SNS platform. If an abnormal reaction is detected, the server refers to past data and generates a response message. The input is the external reaction data, and the output is the generated response message.

[0590] Step 6:

[0591] The server analyzes the collected external reactions, creates a summary, and provides it to the user. This summary shows the proportion of positive, negative, and neutral opinions, which can be used to help create future posts. The input is the external reaction data, and the output is a summary of the analysis results.

[0592] 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.

[0593] This invention provides a system that enables information providers on social media and other platforms to manage information more appropriately and reduce risks such as online firestorms. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0594] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0595] The server uses natural language processing techniques to analyze the received posted data. This analysis assesses the potential legal and ethical risks contained in the posted content. It also uses an emotion engine to recognize the user's emotional state, for example, whether the user is stressed or relaxed.

[0596] Once the risk assessment is complete, the server generates risk points along with proposed solutions based on them. Furthermore, these solutions are adjusted to best suit the user's emotional state. For example, if the user is stressed, the explanation might be presented more gently.

[0597] Once the proposed revisions are generated, the server sends them to the terminal. The terminal then presents the proposed revisions and related risk information to the user, who then modifies their post accordingly.

[0598] After a post is published, the server continuously monitors external reactions to it. The server analyzes this reaction data and detects negative reactions that exceed the normal range (abnormal reactions). If an abnormal reaction is detected, the server creates a draft response based on past data. In this process, the tone of the draft response may be adjusted to match the emotional state of the information provider.

[0599] As a concrete example, consider a scenario where a company posts about a new product, and that post receives criticism from some customers. The system, in order to alleviate the stress on the company's public relations representative (user), activates an emotion engine to generate and quickly deliver an appropriate response in the right tone. As a result, the public relations representative can respond appropriately, contributing to maintaining the company's image.

[0600] Thus, the system of the present invention functions effectively to minimize risks on social media while taking into account the emotions of information providers.

[0601] The following describes the processing flow.

[0602] Step 1:

[0603] The user creates a post on their device. The device collects the user's input data in real time, and the sentiment engine begins to recognize the user's emotional state.

[0604] Step 2:

[0605] The user presses the "Confirm" button. The device prepares to send the posted data and the user's emotional state information to the server.

[0606] Step 3:

[0607] The server receives the posted data. The server uses natural language processing algorithms to analyze the content of the post and assess the legal and ethical risks.

[0608] Step 4:

[0609] The server uses the results of the emotion engine's analysis to consider the user's emotional state. In particular, it examines whether the user is experiencing anxiety or stress.

[0610] Step 5:

[0611] The server identifies risk points in a post and generates suggested revisions as needed. These revisions are adjusted to reflect the user's emotional state. For example, if the user is feeling stressed, more detailed explanations will be provided.

[0612] Step 6:

[0613] The server sends the risk assessment results and the adjusted proposed modifications to the terminal.

[0614] Step 7:

[0615] The terminal displays information sent from the server to the user. The user reviews the posted content based on the displayed risk assessment and suggested revisions, and makes corrections as needed.

[0616] Step 8:

[0617] The user confirms the revised post and publishes it. The device then reviews the post and uploads it to the social media platform.

[0618] Step 9:

[0619] After a post is published, the server continuously monitors external reactions. Through analysis of these reactions, it detects abnormal reactions, such as an increase in negative comments.

[0620] Step 10:

[0621] When an abnormal response is detected, the server automatically generates an appropriate response by referring to past case studies. This response is also adjusted to take the user's emotional state into consideration.

[0622] Step 11:

[0623] The server generates a response message and sends it to the terminal. The user reviews it on the terminal and makes any necessary customizations.

[0624] Step 12:

[0625] The server performs a detailed analysis of external reactions, summarizes the results, and provides them to the user. This allows the user to obtain useful feedback for future activities.

[0626] (Example 2)

[0627] 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."

[0628] When information providers disseminate information on social media and other platforms, there is a lack of means to prevent legal and ethical risks, as well as problems arising from negative reactions from external parties. Furthermore, generating revised versions or responses without considering the emotional state of the information provider can actually cause stress and inappropriate responses. Therefore, it is necessary to resolve these issues and enable information providers to disseminate information with confidence.

[0629] 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.

[0630] In this invention, the server includes means for analyzing information generated by the information sender and evaluating the risks based on legal and ethical standards, means for suggesting revisions while considering the emotional state of the information sender, and means for continuously monitoring external reactions to the information and detecting abnormal reactions. This enables responses that are appropriate to the emotional state of the information sender, resulting in more appropriate and effective risk management.

[0631] An "information analysis device" is a device that analyzes information generated by information providers and has the function of evaluating the risks from a legal and ethical standpoint.

[0632] An "information provider" refers to an individual or organization that disseminates information on social media or other platforms.

[0633] "Risk assessment" is the process of identifying the legal and ethical risks inherent in the content of information and measuring their severity.

[0634] A "revised proposal" is a suggestion to improve or adjust the information content based on a risk assessment.

[0635] "Emotional state" is an indicator used to assess the psychological health and emotional arousal of an information provider, and includes states of stress and relaxation.

[0636] An "abnormal response" is data indicating a negative external reaction that exceeds the normal range.

[0637] A "response draft" is a draft of a statement generated to facilitate appropriate communication in response to an abnormal reaction.

[0638] "Collected external reactions" refer to comments and feedback received from external sources regarding the information that has been disseminated.

[0639] This invention provides a system for reducing legal and ethical risks and appropriately managing external reactions when information providers disseminate information on social media, etc. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0640] The user creates a social media post on their device. The device then prepares to send the generated post data to the server. The user can create the post content using a standard text editor or a dedicated application.

[0641] The server uses natural language processing techniques to analyze the content of posts received from terminals. Modern deep learning models (e.g., BERT and GPT) are used as generative AI models for this analysis. This process is necessary to assess the legal and ethical risks of the posts. Furthermore, an emotion engine is utilized to recognize the user's emotional state.

[0642] After the analysis is complete, the server generates a revised version based on the risk assessment. This revised version is optimized according to the user's emotional state. For example, if the user is stressed, the revised version can be more detailed and thoughtful. The generated revised version is sent from the server to the terminal and presented to the user. The user can then revise their post based on this revised version.

[0643] As a concrete example, consider a situation where a company posts about a new product on social media, and the post is met with criticism from some customers. In this situation, the system can take into account the emotional state of the public relations representative (user) and generate and present an appropriate response, enabling a swift and effective response.

[0644] An example of a prompt might be the text, "Generate a response to critical comments on the new product."

[0645] In this way, the system provides practical means for information providers to minimize risks on social media.

[0646] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0647] Step 1:

[0648] The user creates a social media post using their device. As input, the user enters the text for the post using a text editor or a dedicated app. This data is temporarily stored on the device as a draft for later transmission to the server. Once the user completes the post and issues a send command, the process moves to the next step.

[0649] Step 2:

[0650] The terminal prepares to send user-created post data to the server. Specifically, the terminal encrypts the post data in a secure manner and structures the data according to the transmission protocol. The input is the user-prepared post data, and the output is encrypted data ready for transmission.

[0651] Step 3:

[0652] The server analyzes the posted data received from the terminal. This analysis uses natural language processing techniques, particularly generative AI models, to syntactically parse the text data and assess legal and ethical risks. The received posted data is used as input, and a risk assessment report is generated as output. Furthermore, an emotion engine is used to analyze the user's emotional state and assess their stress and relaxation levels.

[0653] Step 4:

[0654] The server generates revised suggestions based on risk assessment. This revision generation utilizes a generative AI model, which suggests optimizing the posted content while considering the user's emotional state. The inputs used are the risk assessment report and emotional state data, and the output is the revised suggestions presented to the user.

[0655] Step 5:

[0656] The server sends the generated proposed corrections and risk information to the terminal. The terminal then displays the received information to the user. Specifically, the terminal screen displays the proposed corrections and risk points, and the interface is designed to make it easy for the user to review them. The input is the proposed correction data from the server, and the output is visual feedback provided to the user.

[0657] Step 6:

[0658] After a post is published, the server monitors external reactions. It collects new comments and feedback via SNS APIs and analyzes them. The input is unstructured external data, and the output is the detection of abnormal reactions. Based on these results, the server generates appropriate response drafts and adjusts the tone according to the emotional state of the information provider.

[0659] (Application Example 2)

[0660] 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."

[0661] In online information dissemination, negative reactions and online firestorms can have a significant impact on individuals and e-commerce, making it crucial to mitigate risks beforehand. Furthermore, emotionally driven statements by information providers can lead to trouble. Against this backdrop, a system is needed that provides appropriate revisions and response templates while considering the emotional state of the information provider.

[0662] 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.

[0663] In this invention, the server includes means for evaluating the emotional state of the information sender using an information analysis device, means for making constructive suggestions based on a risk assessment of the review or comment, and means for adjusting the content of the post using the emotion analysis device. This enables information senders to send information calmly and constructively while mitigating risks.

[0664] An "information analysis device" is a device that analyzes information generated by information providers and assesses the risks based on legal and ethical standards.

[0665] A "sentiment analysis device" is a device that evaluates the emotional state of information providers and helps them adjust the content of their posts.

[0666] A "risk assessment-based revision proposal" is a revised proposal presented to the information provider in order to mitigate the legal and ethical risks inherent in the posted content.

[0667] An "abnormal reaction" refers to a negative external response to information transmitted by an information transmission device that exceeds the normal range.

[0668] A "response draft" is a proposed response generated in response to negative reactions received by an information provider, often by referring to past successful examples.

[0669] A "constructive suggestion" is a specific proposal to make the information provider's review or comment more positive and fair.

[0670] The system of this invention reduces risks and promotes constructive communication by using natural language processing and sentiment analysis technologies when users transmit information. The entire system consists of a terminal, a server, and related analysis devices.

[0671] The terminal is responsible for receiving information entered by the user and sending that data to the server. The server uses an information analysis device to perform natural language processing (for example, using TensorFlow or the Transformers library from Hugging Face) and evaluate the legal and ethical risks of the entered information. In addition, an emotion analysis device determines the emotional state of the information sender and identifies emotions such as stress and anger.

[0672] Based on this analysis, the server generates revised proposals aligned with the risk assessment. These proposals aim to soften responses and include more positive and specific language. Draft response statements based on past cases are also created and presented to the information provider. This allows users to improve their information dissemination from a calm and objective perspective.

[0673] As a concrete example, suppose a user writes a review about a product. If the review contains emotional and negative content, the system assesses the risk and suggests, "Try to provide constructive feedback by specifically mentioning areas for improvement in the product." This process guides user submissions to be fair and easily accepted.

[0674] An example of a prompt message generated by the AI ​​model is, "Please enter your post content. In particular, please revise it to soften negative sentiments and make it constructive and fair." This allows users to follow the prompt and adjust their own expression.

[0675] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0676] Step 1:

[0677] Users create information posts using their devices. Here, users input the content of reviews and comments. The entered text data is temporarily stored on the device.

[0678] Step 2:

[0679] The terminal sends the entered text data to the server. The terminal transfers the data to the server using a specific protocol (e.g., HTTP or HTTPS). Text data is sent as input.

[0680] Step 3:

[0681] The server analyzes the received text data using a natural language processing model. Here, libraries such as TensorFlow and Transformers are used to analyze the text and evaluate the legal and ethical risks inherent in its content. The input text data is the target of the analysis, and a risk assessment is generated as output.

[0682] Step 4:

[0683] Simultaneously, the server uses an emotion analysis device to evaluate the user's emotional state. Based on the input text, it performs emotion analysis and detects the emotions the user may be experiencing (e.g., stress, anger, etc.). The output is the evaluation result of the emotional state.

[0684] Step 5:

[0685] The server generates suggested revisions for the user based on risk assessment and emotional state assessment. Using a generation AI model, it specifically points out areas for improvement in the posted content and encourages constructive feedback. The output is the proposed revisions.

[0686] Step 6:

[0687] The server sends the generated revised draft and response draft to the terminal. The revised draft specifically outlines how the information to be communicated should be improved. The terminal receives the output from the server and displays the suggestions to the user.

[0688] Step 7:

[0689] Users adjust and finalize their information based on the suggested revisions displayed on their devices. During this process, they refer to the prompt message, "Please enter your post content. In particular, please revise it to mitigate negative sentiment and ensure it is constructive and fair." The final revised content is then disseminated externally in a way that reduces risk.

[0690] 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.

[0691] 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.

[0692] 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.

[0693] [Fourth Embodiment]

[0694] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0695] 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.

[0696] 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).

[0697] 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.

[0698] 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.

[0699] 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).

[0700] 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.

[0701] 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.

[0702] 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.

[0703] 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.

[0704] 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.

[0705] 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.

[0706] 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".

[0707] This invention provides a system that checks information before posting it to social media and other platforms to prevent problems from occurring. This system mainly includes an information analysis device, a terminal, and a server.

[0708] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0709] When the server receives submitted data, it analyzes the data using natural language processing techniques. The purpose of the analysis is to check whether the submitted content poses any legal or ethical risks. For example, it identifies whether it contains racist or potentially misleading language. Based on these results, the server lists the potential risks present in the submission and generates specific suggestions for revision.

[0710] Next, the server sends the generated risk assessment and proposed corrections to the user's device. On the device, this information is displayed visually to the user, allowing them to modify their post considering the identified risks.

[0711] After a post is published, the server continuously monitors external reactions to that post. The server collects the reactions and determines if they are abnormal. For example, a sudden surge in critical comments or a topic becoming a major topic of discussion in a short period of time may be considered a sign of an impending online firestorm.

[0712] If a controversy is detected, the server quickly references similar past case studies and generates a draft apology. This draft apology is crafted with sincerity and specificity in mind and is promptly provided to the user.

[0713] Furthermore, the server analyzes the collected external reactions and generates a summary of user feedback. This summary shows the proportion of positive, negative, and neutral opinions and includes specific suggestions for improvement that will help users with future posts.

[0714] As a concrete example, consider a scenario where a public relations representative for a company posts an introduction to a new product. Suppose this post contains an inappropriate comparison to a competing product. The system detects this and suggests revisions from an ethical standpoint. The PR representative adjusts the post according to the suggestions, implements preventative measures, and then publishes it. If, after publication, criticism from the competing company increases, the server provides a suitable apology draft and also offers suggestions for improving future posts.

[0715] In this way, the system of the present invention supports information providers in communicating smoothly while minimizing risks.

[0716] The following describes the processing flow.

[0717] Step 1:

[0718] The user creates a post on their device. The device receives the user input as text data.

[0719] Step 2:

[0720] The device presses the "Confirm" button. Afterward, the device prepares to send the posted data to the server.

[0721] Step 3:

[0722] The server receives the submitted data. The server executes a natural language processing algorithm to analyze the information.

[0723] Step 4:

[0724] The server evaluates the content of posts based on legal and ethical standards. If there are any risk points, they are identified and highlighted.

[0725] Step 5:

[0726] The server generates revised plans along with the evaluation results. These plans include specific suggestions for mitigating risks.

[0727] Step 6:

[0728] The server sends the generated evaluation results and suggested corrections to the terminal. The results are then displayed to the user on the terminal.

[0729] Step 7:

[0730] Users edit their posts through their devices. They make corrections as needed based on the identified risks and then finalize the post.

[0731] Step 8:

[0732] The user publishes a post, and the device uploads that information to the social networking platform.

[0733] Step 9:

[0734] The server continuously monitors external reactions to posts after they are published, paying particular attention to criticism and sudden increases in the number of comments.

[0735] Step 10:

[0736] When the server detects an abnormal response, it automatically generates an appropriate response by referring to past case studies.

[0737] Step 11:

[0738] The server generates a draft response message and sends it to the terminal. The user reviews it and customizes it as needed.

[0739] Step 12:

[0740] The server analyzes external reaction comments and summarizes them as positive, negative, or neutral.

[0741] Step 13:

[0742] The server sends a summary and improvement suggestions to the user's device, which helps them in creating future posts.

[0743] (Example 1)

[0744] 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".

[0745] There is a lack of solutions to prevent legal and ethical problems when information providers publish information on social media and other online platforms, and to respond quickly and appropriately to reactions after publication. Furthermore, in a situation where unexpected external reactions and criticisms to published information are rapidly increasing, there is a need for a mechanism that efficiently presents countermeasures and reduces the burden on information providers.

[0746] 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.

[0747] In this invention, the server includes means for analyzing information generated by the information sender using natural language processing technology and evaluating risks based on legal and ethical standards; means for generating revised versions based on prompt sentences using a generation AI model and presenting them to the information sender; and means for continuously monitoring external reactions and detecting abnormal reactions. As a result, the information sender can grasp risks in advance, revise their expression to an appropriate one, and respond quickly to reactions after publication.

[0748] An "information analysis device" is a device that analyzes data generated by information providers and has the function of evaluating risks based on legal and ethical standards.

[0749] "Natural language processing technology" is a method that understands text data semantically and grammatically, and analyzes the ambiguity and complexity unique to human language.

[0750] A "generative AI model" is an artificial intelligence model that learns from past data and generates new text or suggestions.

[0751] A "prompt" is a sentence of instruction or question input to an AI model to obtain a specific response or output.

[0752] "Risk assessment" is the process of determining whether the content of information may cause legal or ethical problems before it is made public.

[0753] A "proposal for revision" refers to specific changes proposed to make the information being disseminated more appropriate and problem-free.

[0754] An "abnormal reaction" refers to a response that exhibits critical or emotional feedback from an external source that exceeds the normal range.

[0755] A "draft response" is a document provided by the information provider to enable them to respond quickly to detected abnormal reactions.

[0756] A "summary" is a shortened version of collected external responses, extracting the key points and providing them to the information provider.

[0757] This invention is a system that minimizes the risks for information providers when posting information on online platforms. The system mainly consists of a server, terminals, and users.

[0758] server

[0759] The server receives posted data sent from the user's terminal. This data is analyzed using natural language processing techniques. Specifically, Python natural language processing libraries (e.g., spaCy, NLTK) are used to evaluate the legal and ethical risks of the posted content. The server utilizes a generative AI model to generate suggested revisions based on the provided prompt text. These suggested revisions are notified to the user via their terminal. The server also monitors external reactions after the post is published and detects any abnormal reactions. If an abnormal reaction is detected, it generates a suggested response based on past data and provides it to the user promptly.

[0760] terminal

[0761] The device collects user-entered posting data and sends it to the server. It then receives suggested revisions and risk assessments from the server, which are displayed visually to the user. This allows the user to review their posts and make revisions as needed.

[0762] User

[0763] Users can use this system to create their posts, reviewing risk assessments and suggested revisions along the way. For example, suppose a corporate public relations representative creating a new product description includes ethically problematic language. In this case, the system will detect the language and suggest revisions. The user can then adjust the content based on the feedback and make an appropriate post.

[0764] For example, a prompt might include instructions such as, "Please revise this expression to be neutral." Based on these instructions, the AI ​​generates specific revisions to mitigate risk. This allows users to communicate information more safely and effectively.

[0765] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0766] Step 1:

[0767] Users create post data on their devices. This post data may include text, images, or videos. The information entered by the user is first temporarily stored on the device. At this time, a post text is generated based on the user's goals and intentions.

[0768] Step 2:

[0769] The terminal prepares to send the created post data to the server. The terminal converts the data to the appropriate format and sends it to the server using the encrypted HTTP protocol, ensuring data security over network communication.

[0770] Step 3:

[0771] The server receives posted data sent from the terminal. Based on the received data, it analyzes the data using natural language processing techniques. This analysis process breaks down the text, extracts keywords and phrases, and assesses the legal or ethical risks. For example, it calculates a sensitivity score for each sentence and detects inappropriate phrases.

[0772] Step 4:

[0773] Based on the analysis results, the server generates suggested corrections for areas where risks were detected. Specifically, it uses a generation AI model to receive the prompt "Please correct this expression to be neutral" and generates a new expression. During this process, data is input to the model and suggested corrections are output.

[0774] Step 5:

[0775] The server sends the generated revised proposals and risk assessments to the terminal. The transmitted data is formatted in a way that is easily understandable to the user.

[0776] Step 6:

[0777] The terminal displays the risk assessment and proposed corrections received from the server to the user. Using a visual GUI, the terminal highlights risk areas and displays proposed corrections in pop-ups or side panels. This allows users to easily review their feedback.

[0778] Step 7:

[0779] Users review their posts based on the suggested revisions and make corrections as needed. The revised posts are then reviewed and ready for publication.

[0780] Step 8:

[0781] The server continuously monitors external reactions to a post even after it has been published. The server collects data in real time and analyzes whether any abnormal reactions are occurring. This analysis includes comment frequency and sentiment analysis.

[0782] Step 9:

[0783] If the server detects an abnormal response, it will generate a suggested response based on similar past cases and provide it to the user. This suggested response will include apologies and suggestions for corrections to help the user respond quickly.

[0784] (Application Example 1)

[0785] 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".

[0786] In modern digital communication, information dissemination via social media is commonplace, but at the same time, inappropriate posts that carry legal and ethical risks have become a problem. Therefore, it is necessary to detect these risks in advance and address them appropriately. However, conventional systems have difficulty responding in real time, and there are challenges in making highly accurate risk assessments and corrective suggestions.

[0787] 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.

[0788] In this invention, the server includes means for analyzing information generated by information senders using an information analysis device and evaluating the risks based on legal and ethical standards; means for enhancing natural language processing using a generative AI model to analyze posted information and generate proposed revisions; and means for dynamically adjusting input prompt sentences to the generative AI model to improve analysis accuracy. This makes it possible to evaluate the risks in SNS posts in advance with high accuracy and to quickly present appropriate revision suggestions.

[0789] An "information analysis device" is a device that analyzes data generated by information providers and assesses risks based on legal and ethical standards.

[0790] A "generative AI model" is an artificial intelligence model designed for advanced natural language processing, and is used for analyzing posted information and generating suggested revisions.

[0791] A "prompt sentence" is a sentence used as input to a generative AI model, and it is dynamically adjusted to improve the accuracy of the analysis.

[0792] "External reactions" refer to comments and feedback received from third parties regarding information made public through information dissemination devices.

[0793] An "abnormal reaction" refers to a characteristic response that is not typical, such as a sudden surge in critical comments or rapid growth in public attention.

[0794] A "response draft" is a draft text, generated based on past data and successful cases, intended to be an apology or response, provided to the information provider when an abnormal reaction is detected.

[0795] A "revision proposal" is a specific suggestion for correction presented to the information provider when legal or ethical risks are detected in the content of their post.

[0796] A "summary" is information that shows the proportion of positive, negative, and neutral opinions, compiled to provide information providers with the results of an analysis of collected external reactions.

[0797] To implement this invention, the server must be equipped with an information analysis device, a generative AI model, and a prompt sentence adjustment system. The server receives posted data from information senders and performs analysis using natural language processing technology. This analysis utilizes a generative AI model to perform a highly accurate risk assessment of the text. At this time, the prompt sentences input to the generative AI model are dynamically adjusted to improve the accuracy of the analysis.

[0798] Based on the risk assessment generated from the analysis results, the server presents the user with suggested revisions on their device. The user can then adjust their post according to the suggested revisions displayed on their device and resubmit it. This feature enables the dissemination of information that complies with legal and ethical standards.

[0799] Furthermore, after the information is published, the server continuously monitors external reactions. If an abnormal reaction is detected, the server uses a generative AI model, based on past cases, to generate an appropriate response and provide it to the user. In addition, the server analyzes the collected external reactions and presents them to the information provider as a summary, which can be used to improve future posts.

[0800] For example, if a user is about to post an opinion about a new product, this system can pre-scan the content, identify potentially misleading parts, and suggest neutral wording.

[0801] An example of a prompt message might be: "Regarding a new social media post, it says: 'Our new product is the best on the market and overwhelmingly superior to the competition.' Please assess whether this post contains any ethical or legal risks and provide suggestions for revisions if necessary."

[0802] In this way, the entire system enables information providers to consistently manage risks both before and after the event, and to facilitate smooth communication.

[0803] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0804] Step 1:

[0805] Users create post data using their devices. This post data is prepared to be sent from the device to the server for sharing on the SNS platform. The input is free-form text data created by the user, and the output is text data ready to be sent to the server.

[0806] Step 2:

[0807] The server uses a generative AI model to perform natural language processing to process the received posted data. The server prepares prompt sentences and inputs these prompt sentences into the AI ​​model for analysis. The input consists of the user's posted text and prompt sentences, and the output is an assessment of legal and ethical risks.

[0808] Step 3:

[0809] Based on the generated risk assessment, the server creates proposed revisions to the posted content. These revisions include specific suggestions for improving particular parts of the information. The input is the result of the risk assessment, and the output is the proposed revisions.

[0810] Step 4:

[0811] The suggested revisions are displayed on the terminal's user interface and presented to the user. The user reviews them and modifies the post as needed. The input is the suggested revisions sent from the server, and the output is the updated post after the user has made the necessary corrections.

[0812] Step 5:

[0813] After the information is officially released, the server continuously monitors external reactions on the SNS platform. If an abnormal reaction is detected, the server refers to past data and generates a response message. The input is the external reaction data, and the output is the generated response message.

[0814] Step 6:

[0815] The server analyzes the collected external reactions, creates a summary, and provides it to the user. This summary shows the proportion of positive, negative, and neutral opinions, which can be used to help create future posts. The input is the external reaction data, and the output is a summary of the analysis results.

[0816] 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.

[0817] This invention provides a system that enables information providers on social media and other platforms to manage information more appropriately and reduce risks such as online firestorms. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0818] First, the information provider (user) creates an SNS post on their device. The device receives the post data from the user and prepares to send it to the server.

[0819] The server uses natural language processing techniques to analyze the received posted data. This analysis assesses the potential legal and ethical risks contained in the posted content. It also uses an emotion engine to recognize the user's emotional state, for example, whether the user is stressed or relaxed.

[0820] Once the risk assessment is complete, the server generates risk points along with proposed solutions based on them. Furthermore, these solutions are adjusted to best suit the user's emotional state. For example, if the user is stressed, the explanation might be presented more gently.

[0821] Once the proposed revisions are generated, the server sends them to the terminal. The terminal then presents the proposed revisions and related risk information to the user, who then modifies their post accordingly.

[0822] After a post is published, the server continuously monitors external reactions to it. The server analyzes this reaction data and detects negative reactions that exceed the normal range (abnormal reactions). If an abnormal reaction is detected, the server creates a draft response based on past data. In this process, the tone of the draft response may be adjusted to match the emotional state of the information provider.

[0823] As a concrete example, consider a scenario where a company posts about a new product, and that post receives criticism from some customers. The system, in order to alleviate the stress on the company's public relations representative (user), activates an emotion engine to generate and quickly deliver an appropriate response in the right tone. As a result, the public relations representative can respond appropriately, contributing to maintaining the company's image.

[0824] Thus, the system of the present invention functions effectively to minimize risks on social media while taking into account the emotions of information providers.

[0825] The following describes the processing flow.

[0826] Step 1:

[0827] The user creates a post on their device. The device collects the user's input data in real time, and the sentiment engine begins to recognize the user's emotional state.

[0828] Step 2:

[0829] The user presses the "Confirm" button. The device prepares to send the posted data and the user's emotional state information to the server.

[0830] Step 3:

[0831] The server receives the posted data. The server uses natural language processing algorithms to analyze the content of the post and assess the legal and ethical risks.

[0832] Step 4:

[0833] The server uses the results of the emotion engine's analysis to consider the user's emotional state. In particular, it examines whether the user is experiencing anxiety or stress.

[0834] Step 5:

[0835] The server identifies risk points in a post and generates suggested revisions as needed. These revisions are adjusted to reflect the user's emotional state. For example, if the user is feeling stressed, more detailed explanations will be provided.

[0836] Step 6:

[0837] The server sends the risk assessment results and the adjusted proposed modifications to the terminal.

[0838] Step 7:

[0839] The terminal displays information sent from the server to the user. The user reviews the posted content based on the displayed risk assessment and suggested revisions, and makes corrections as needed.

[0840] Step 8:

[0841] The user confirms the revised post and publishes it. The device then reviews the post and uploads it to the social media platform.

[0842] Step 9:

[0843] After a post is published, the server continuously monitors external reactions. Through analysis of these reactions, it detects abnormal reactions, such as an increase in negative comments.

[0844] Step 10:

[0845] When an abnormal response is detected, the server automatically generates an appropriate response by referring to past case studies. This response is also adjusted to take the user's emotional state into consideration.

[0846] Step 11:

[0847] The server generates a response message and sends it to the terminal. The user reviews it on the terminal and makes any necessary customizations.

[0848] Step 12:

[0849] The server performs a detailed analysis of external reactions, summarizes the results, and provides them to the user. This allows the user to obtain useful feedback for future activities.

[0850] (Example 2)

[0851] 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".

[0852] When information providers disseminate information on social media and other platforms, there is a lack of means to prevent legal and ethical risks, as well as problems arising from negative reactions from external parties. Furthermore, generating revised versions or responses without considering the emotional state of the information provider can actually cause stress and inappropriate responses. Therefore, it is necessary to resolve these issues and enable information providers to disseminate information with confidence.

[0853] 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.

[0854] In this invention, the server includes means for analyzing information generated by the information sender and evaluating the risks based on legal and ethical standards, means for suggesting revisions while considering the emotional state of the information sender, and means for continuously monitoring external reactions to the information and detecting abnormal reactions. This enables responses that are appropriate to the emotional state of the information sender, resulting in more appropriate and effective risk management.

[0855] An "information analysis device" is a device that analyzes information generated by information providers and has the function of evaluating the risks from a legal and ethical standpoint.

[0856] An "information provider" refers to an individual or organization that disseminates information on social media or other platforms.

[0857] "Risk assessment" is the process of identifying the legal and ethical risks inherent in the content of information and measuring their severity.

[0858] A "revised proposal" is a suggestion to improve or adjust the information content based on a risk assessment.

[0859] "Emotional state" is an indicator used to assess the psychological health and emotional arousal of an information provider, and includes states of stress and relaxation.

[0860] An "abnormal response" is data indicating a negative external reaction that exceeds the normal range.

[0861] A "response draft" is a draft of a statement generated to facilitate appropriate communication in response to an abnormal reaction.

[0862] "Collected external reactions" refer to comments and feedback received from external sources regarding the information that has been disseminated.

[0863] This invention provides a system for reducing legal and ethical risks and appropriately managing external reactions when information providers disseminate information on social media, etc. This system includes an information analysis device, a terminal, a server, and an emotion engine.

[0864] The user creates a social media post on their device. The device then prepares to send the generated post data to the server. The user can create the post content using a standard text editor or a dedicated application.

[0865] The server uses natural language processing techniques to analyze the content of posts received from terminals. Modern deep learning models (e.g., BERT and GPT) are used as generative AI models for this analysis. This process is necessary to assess the legal and ethical risks of the posts. Furthermore, an emotion engine is utilized to recognize the user's emotional state.

[0866] After the analysis is complete, the server generates a revised version based on the risk assessment. This revised version is optimized according to the user's emotional state. For example, if the user is stressed, the revised version can be more detailed and thoughtful. The generated revised version is sent from the server to the terminal and presented to the user. The user can then revise their post based on this revised version.

[0867] As a concrete example, consider a situation where a company posts about a new product on social media, and the post is met with criticism from some customers. In this situation, the system can take into account the emotional state of the public relations representative (user) and generate and present an appropriate response, enabling a swift and effective response.

[0868] An example of a prompt might be the text, "Generate a response to critical comments on the new product."

[0869] In this way, the system provides practical means for information providers to minimize risks on social media.

[0870] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0871] Step 1:

[0872] The user creates a social media post using their device. As input, the user enters the text for the post using a text editor or a dedicated app. This data is temporarily stored on the device as a draft for later transmission to the server. Once the user completes the post and issues a send command, the process moves to the next step.

[0873] Step 2:

[0874] The terminal prepares to send user-created post data to the server. Specifically, the terminal encrypts the post data in a secure manner and structures the data according to the transmission protocol. The input is the user-prepared post data, and the output is encrypted data ready for transmission.

[0875] Step 3:

[0876] The server analyzes the posted data received from the terminal. This analysis uses natural language processing techniques, particularly generative AI models, to syntactically parse the text data and assess legal and ethical risks. The received posted data is used as input, and a risk assessment report is generated as output. Furthermore, an emotion engine is used to analyze the user's emotional state and assess their stress and relaxation levels.

[0877] Step 4:

[0878] The server generates revised suggestions based on risk assessment. This revision generation utilizes a generative AI model, which suggests optimizing the posted content while considering the user's emotional state. The inputs used are the risk assessment report and emotional state data, and the output is the revised suggestions presented to the user.

[0879] Step 5:

[0880] The server sends the generated proposed corrections and risk information to the terminal. The terminal then displays the received information to the user. Specifically, the terminal screen displays the proposed corrections and risk points, and the interface is designed to make it easy for the user to review them. The input is the proposed correction data from the server, and the output is visual feedback provided to the user.

[0881] Step 6:

[0882] After a post is published, the server monitors external reactions. It collects new comments and feedback via SNS APIs and analyzes them. The input is unstructured external data, and the output is the detection of abnormal reactions. Based on these results, the server generates appropriate response drafts and adjusts the tone according to the emotional state of the information provider.

[0883] (Application Example 2)

[0884] 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".

[0885] In online information dissemination, negative reactions and online firestorms can have a significant impact on individuals and e-commerce, making it crucial to mitigate risks beforehand. Furthermore, emotionally driven statements by information providers can lead to trouble. Against this backdrop, a system is needed that provides appropriate revisions and response templates while considering the emotional state of the information provider.

[0886] 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.

[0887] In this invention, the server includes means for evaluating the emotional state of the information sender using an information analysis device, means for making constructive suggestions based on a risk assessment of the review or comment, and means for adjusting the content of the post using the emotion analysis device. This enables information senders to send information calmly and constructively while mitigating risks.

[0888] An "information analysis device" is a device that analyzes information generated by information providers and assesses the risks based on legal and ethical standards.

[0889] A "sentiment analysis device" is a device that evaluates the emotional state of information providers and helps them adjust the content of their posts.

[0890] A "risk assessment-based revision proposal" is a revised proposal presented to the information provider in order to mitigate the legal and ethical risks inherent in the posted content.

[0891] An "abnormal reaction" refers to a negative external response to information transmitted by an information transmission device that exceeds the normal range.

[0892] A "response draft" is a proposed response generated in response to negative reactions received by an information provider, often by referring to past successful examples.

[0893] A "constructive suggestion" is a specific proposal to make the information provider's review or comment more positive and fair.

[0894] The system of this invention reduces risks and promotes constructive communication by using natural language processing and sentiment analysis technologies when users transmit information. The entire system consists of a terminal, a server, and related analysis devices.

[0895] The terminal is responsible for receiving information entered by the user and sending that data to the server. The server uses an information analysis device to perform natural language processing (for example, using TensorFlow or the Transformers library from Hugging Face) and evaluate the legal and ethical risks of the entered information. In addition, an emotion analysis device determines the emotional state of the information sender and identifies emotions such as stress and anger.

[0896] Based on this analysis, the server generates revised proposals aligned with the risk assessment. These proposals aim to soften responses and include more positive and specific language. Draft response statements based on past cases are also created and presented to the information provider. This allows users to improve their information dissemination from a calm and objective perspective.

[0897] As a concrete example, suppose a user writes a review about a product. If the review contains emotional and negative content, the system assesses the risk and suggests, "Try to provide constructive feedback by specifically mentioning areas for improvement in the product." This process guides user submissions to be fair and easily accepted.

[0898] An example of a prompt message generated by the AI ​​model is, "Please enter your post content. In particular, please revise it to soften negative sentiments and make it constructive and fair." This allows users to follow the prompt and adjust their own expression.

[0899] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0900] Step 1:

[0901] Users create information posts using their devices. Here, users input the content of reviews and comments. The entered text data is temporarily stored on the device.

[0902] Step 2:

[0903] The terminal sends the entered text data to the server. The terminal transfers the data to the server using a specific protocol (e.g., HTTP or HTTPS). Text data is sent as input.

[0904] Step 3:

[0905] The server analyzes the received text data using a natural language processing model. Here, libraries such as TensorFlow and Transformers are used to analyze the text and evaluate the legal and ethical risks inherent in its content. The input text data is the target of the analysis, and a risk assessment is generated as output.

[0906] Step 4:

[0907] Simultaneously, the server uses an emotion analysis device to evaluate the user's emotional state. Based on the input text, it performs emotion analysis and detects the emotions the user may be experiencing (e.g., stress, anger, etc.). The output is the evaluation result of the emotional state.

[0908] Step 5:

[0909] The server generates suggested revisions for the user based on risk assessment and emotional state assessment. Using a generation AI model, it specifically points out areas for improvement in the posted content and encourages constructive feedback. The output is the proposed revisions.

[0910] Step 6:

[0911] The server sends the generated revised draft and response draft to the terminal. The revised draft specifically outlines how the information to be communicated should be improved. The terminal receives the output from the server and displays the suggestions to the user.

[0912] Step 7:

[0913] Users adjust and finalize their information based on the suggested revisions displayed on their devices. During this process, they refer to the prompt message, "Please enter your post content. In particular, please revise it to mitigate negative sentiment and ensure it is constructive and fair." The final revised content is then disseminated externally in a way that reduces risk.

[0914] 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.

[0915] 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.

[0916] 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 robot 414.

[0917] 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.

[0918] 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.

[0919] 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.

[0920] 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.

[0921] 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.

[0922] 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."

[0923] 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.

[0924] 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.

[0925] 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.

[0926] 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.

[0927] 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.

[0928] 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.

[0929] 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.

[0930] 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.

[0931] 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.

[0932] 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.

[0933] 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.

[0934] 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 to be incorporated by reference.

[0935] The following is further disclosed regarding the embodiments described above.

[0936] (Claim 1)

[0937] Information analysis equipment provides a means to analyze information generated by information providers and assess risks based on legal and ethical standards.

[0938] A means of presenting revised proposals to information providers based on the aforementioned risk assessment,

[0939] A means for continuously monitoring external reactions to information transmitted by an information transmission device and detecting abnormal reactions,

[0940] A means for generating a response draft based on past data and providing it to the information sender when the aforementioned abnormal reaction is detected,

[0941] A means of analyzing collected external reactions and providing summaries to information providers,

[0942] A system that includes this.

[0943] (Claim 2)

[0944] The system according to claim 1, wherein the proposed revisions displayed to the information provider are generated using natural language processing.

[0945] (Claim 3)

[0946] The system according to claim 1, wherein the response text displayed in the information transmission device is generated by referring to past successful cases.

[0947] "Example 1"

[0948] (Claim 1)

[0949] Information analysis equipment provides a means to analyze information generated by information providers and assess risks based on legal and ethical standards.

[0950] A means of presenting revised proposals to information providers based on the aforementioned risk assessment,

[0951] A means for continuously monitoring external reactions to information transmitted by an information dissemination means and detecting abnormal reactions,

[0952] A means for generating a proposed response based on past data and providing it to the information sender when the aforementioned abnormal reaction is detected,

[0953] A means for analyzing external reactions and providing a summary of those reactions to the information provider,

[0954] A means of information dissemination that uses a generative AI model to generate suggested revisions to a specific expression based on a prompt sentence,

[0955] A system that includes this.

[0956] (Claim 2)

[0957] The system according to claim 1, wherein the proposed revisions displayed to the information provider are generated using natural language processing technology.

[0958] (Claim 3)

[0959] The system according to claim 1, wherein a proposed response in the information dissemination means is generated by referring to past successful examples.

[0960] "Application Example 1"

[0961] (Claim 1)

[0962] Information analysis equipment provides a means to analyze information generated by information providers and assess risks based on legal and ethical standards.

[0963] A means of presenting revised proposals to information providers based on the aforementioned risk assessment,

[0964] A means for continuously monitoring external reactions to information transmitted by an information transmission device and detecting abnormal reactions,

[0965] A means for generating a response draft based on past data and providing it to the information sender when the aforementioned abnormal reaction is detected,

[0966] A means of analyzing collected external reactions and providing summaries to information providers,

[0967] A means to enhance natural language processing using a generative AI model for analyzing posted information and generating proposed revisions,

[0968] A means for dynamically adjusting the input prompt sentences to the aforementioned generated AI model and improving the accuracy of the analysis,

[0969] A system that includes this.

[0970] (Claim 2)

[0971] The system according to claim 1, wherein the proposed revisions displayed to the information provider are generated using natural language processing and their accuracy is improved using a generative AI model.

[0972] (Claim 3)

[0973] The system according to claim 1, wherein the response text installed in the information transmission device is generated using a generation AI model, with reference to past successful cases.

[0974] "Example 2 of combining an emotion engine"

[0975] (Claim 1)

[0976] Information analysis equipment provides a means to analyze information generated by information providers and assess risks based on legal and ethical standards.

[0977] A means of presenting revised proposals based on the aforementioned risk assessment, while taking into account the emotional state of the information provider,

[0978] A means for continuously monitoring external reactions to information transmitted by an information transmission device and detecting abnormal reactions,

[0979] A means for generating and providing a response draft based on past data, according to the emotional state of the information sender, when the aforementioned abnormal reaction is detected.

[0980] A means of analyzing collected external reactions and providing summaries to information providers,

[0981] A system that includes this.

[0982] (Claim 2)

[0983] The system according to claim 1, wherein the proposed revisions displayed to the information provider are generated using natural language processing and adjusted according to the emotional state of the information provider.

[0984] (Claim 3)

[0985] The system according to claim 1, wherein the response text installed in the information transmission device is generated by referring to past successful cases and according to the emotional state of the information sender.

[0986] "Application example 2 when combining with an emotional engine"

[0987] (Claim 1)

[0988] Information analysis equipment provides a means to analyze information generated by information providers and assess risks based on legal and ethical standards.

[0989] A means of presenting revised proposals to information providers based on the aforementioned risk assessment,

[0990] A means for continuously monitoring external reactions to information transmitted by an information transmission device and detecting abnormal reactions,

[0991] A means for generating a response draft based on past data and providing it to the information sender when the aforementioned abnormal reaction is detected,

[0992] A means of analyzing collected external reactions and providing summaries to information providers,

[0993] A means of evaluating the emotional state of information providers using an emotion analysis device and supporting the adjustment of post content,

[0994] A means of making constructive suggestions based on a risk assessment of reviews or comments,

[0995] A system that includes this.

[0996] (Claim 2)

[0997] The system according to claim 1, wherein the proposed revisions displayed to the information provider are generated using natural language processing.

[0998] (Claim 3)

[0999] The system according to claim 1, wherein the response text displayed in the information transmission device is generated by referring to past successful cases. [Explanation of Symbols]

[1000] 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. Information analysis equipment provides a means to analyze information generated by information providers and assess risks based on legal and ethical standards. A means of presenting revised proposals to information providers based on the aforementioned risk assessment, A means for continuously monitoring external reactions to information transmitted by an information transmission device and detecting abnormal reactions, A means for generating a response draft based on past data and providing it to the information sender when the aforementioned abnormal reaction is detected, A means of analyzing collected external reactions and providing summaries to information providers, A means to enhance natural language processing using a generative AI model for analyzing posted information and generating proposed revisions, A means for dynamically adjusting the input prompt sentences to the aforementioned generated AI model and improving the accuracy of the analysis, A system that includes this.

2. The system according to claim 1, wherein the proposed revisions displayed to the information provider are generated using natural language processing, and their accuracy is improved using a generative AI model.

3. The system according to claim 1, wherein the response text installed in the information transmission device is generated using a generation AI model, with reference to past successful cases.

Images