system

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving content input by a user; means for analyzing the content; means for determining a possibility of flaming based on a result of the analysis; means for generating a correction proposal based on a result of the determination; means for presenting the correction proposal to the user; means for receiving the content corrected by the user again and performing a final confirmation; and means for posting the finally confirmed content to an SNS platform.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] With the spread of social networking sites (SNS), there has been an increase in cases where individuals and companies have posted content that has led to "flaming," resulting in reputational damage and financial losses. These types of flaming often stem from nuisance videos, inappropriate images, or offensive text. To prevent this, a system is needed that automatically checks content and prompts users to correct it before they accidentally post it. However, such a system is not widely available at present. Therefore, the present invention aims to provide an environment where users can safely use SNS by scanning content before it is posted, predicting the possibility of flaming, and suggesting specific corrections. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems using the following means. A system is provided that includes: means for receiving content entered by a user; means for analyzing the content; means for determining the possibility of a controversy based on the results of the analysis; means for generating revision suggestions based on the determination results; means for presenting the revision suggestions to the user; means for re-receiving the content revised by the user and performing a final check; and means for posting the finally checked content to a social networking platform. The content analysis means includes video analysis, image analysis, and text analysis, and the means for presenting the revision suggestions is configured to display the revision suggestions via a user interface. This allows users to correct inappropriate parts of their posted content in advance and prevent controversy from occurring.

[0006] "User" means an individual or organization that wishes to post content on a social networking site.

[0007] "Content" is a general term for digital data such as videos, images, and text that users wish to post.

[0008] The "means for receiving" is a function that uses information and communication technology to transmit the content entered by the user to the server.

[0009] An "analytical means" is an algorithm or program that analyzes information within the content and evaluates it based on specific criteria.

[0010] The "means for determining the possibility of a controversy" is a function that predicts the possibility that content will be viewed as socially problematic based on the analysis results.

[0011] The "means for generating correction suggestions" is a function that provides specific improvement measures for content that may cause controversy.

[0012] The "means for presenting revision suggestions" is a function that includes a user interface for displaying the generated revision suggestions to the user.

[0013] The "means for receiving again and making a final check" is a function that allows the user to receive the corrected content again and check whether the problem has been resolved.

[0014] "Means of posting to SNS platform" means the means of communication for posting the final verified content to the designated SNS service.

[0015] "Video analysis" is a processing technique that breaks down video frames and extracts and analyzes detailed information from each frame.

[0016] "Image analysis" is a processing technology that analyzes image data and extracts specific objects or features.

[0017] "Text analysis" is a processing technology that uses natural language processing to analyze text data and interpret emotions and meanings.

[0018] A "user interface" is an interface that includes a screen and input devices that allow a user to interact with a system. [Brief explanation of the drawings]

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

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

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

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

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

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

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

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

[0027] [First embodiment]

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

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

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

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

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

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

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

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

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

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

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

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

[0040] The following is a description of the preferred embodiment of the invention based on the claims.

[0041] ---

[0042] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[0043] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0044] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, including facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0045] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[0046] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[0047] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[0048] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0049] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By striving to analyze content quickly and accurately and suggest corrections, it is expected to reduce the risk of online outrage.

[0050] ---

[0051] The above is a specific example of the "mode for carrying out the invention."

[0052] The processing flow will be explained below.

[0053] Step 1:

[0054] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[0055] Step 2:

[0056] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[0057] Step 3:

[0058] The server stores the received content and starts the generative AI model to begin analysis. The server separates the received data by content and prepares it for analysis.

[0059] Step 4:

[0060] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[0061] Step 5:

[0062] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for inflammatory content.

[0063] Step 6:

[0064] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[0065] Step 7:

[0066] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[0067] Step 8:

[0068] To generate the suggested corrections, the server lists specific improvements, such as "suggesting blurring to protect the privacy of a specific person in the image" or "suggesting changes to the text to tone down offensive language."

[0069] Step 9:

[0070] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[0071] Step 10:

[0072] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[0073] Step 11:

[0074] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[0075] Step 12:

[0076] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[0077] Step 13:

[0078] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[0079] The above is a specific program processing flow based on the claims.

[0080] Example 1

[0081] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0082] In today's social networking environment, there is an increasing risk that users may unintentionally post inappropriate content, sparking a social media firestorm. This issue can have serious consequences for users, as it can include violations of personal privacy and offensive language. Conventional methods require users to manually review and correct posts, which not only takes time and effort but also carries the risk of false positives. There is a need for a system that can solve these issues and enable users to post appropriate content safely and quickly.

[0083] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0084] In this invention, the server includes means for receiving content entered by a user, means for transmitting the content to the server, means for saving the content, means for analyzing the content by activating a generative AI model, means for determining the possibility of a controversy, means for generating revision suggestions, means for transmitting the revision suggestions to a terminal, means for presenting the revision suggestions to the user via a user interface, means for re-receiving and re-analyzing the content revised by the user, and means for posting the final confirmed content to an SNS platform. This allows users to automatically determine the risk of content causing a controversy and quickly receive revision suggestions, thereby realizing safe posting.

[0085] A "generative AI model" is an artificial intelligence model trained using machine learning algorithms to analyze and classify input content.

[0086] "Content" refers to all digital data posted on social media, including videos, images, and text.

[0087] A "server" is a part of a computer system that stores, processes, and distributes data over a network.

[0088] "Terminal" refers to a device that is directly operated by a user, such as a smartphone or a personal computer.

[0089] "Analysis" is the process of breaking down the information contained in content and evaluating it for a specific purpose.

[0090] "Potential for a social media outrage" refers to the risk that the content posted will be deemed inappropriate and cause negative reactions or uproar on social media.

[0091] "Revision suggestions" is a means of generating specific revision suggestions for the posted content based on the analysis results.

[0092] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[0093] "Re-analysis" is the process of re-analyzing content that has been modified by a user to check for new problems.

[0094] "Post Request" means a request to post Final Reviewed Content to a Social Media Platform.

[0095] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[0096] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0097] The server stores the received content and starts analyzing it by activating a generative AI model, which is an artificial intelligence model trained using machine learning algorithms to analyze and classify the input content.

[0098] For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial, object, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, facial expression analysis, and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0099] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[0100] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[0101] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[0102] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0103] This system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By quickly and accurately analyzing content and suggesting corrections, it is expected to reduce the risk of online outrage.

[0104] An example of a prompt sentence is, "I want to post on social media about yesterday's drinking party. For example, it should include the following content: Photo: A photo of the drinking party (faces are clearly visible), Text: 'We drank too much last night and got rowdy!' Please check whether the content of this post poses a risk of causing a backlash and let me know any suggestions for corrections."

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Step 1:

[0107] A user inputs content into a multi-function posting app.

[0108] Specifically, a user uploads a video, image, or text to the app to post on a social network. The input is the data (video, image, text) that is stored within the application. The output is the content that is transformed into a structured format within the application to be sent to the next step.

[0109] Step 2:

[0110] The terminal receives the content and transmits it to the server.

[0111] The device sends the video, image, or text received from the multi-function posting app to the server. Specifically, the device securely transmits data using encryption protocols such as SSL / TLS. The input is the content uploaded by the user, and the output is the encrypted data sent to the server.

[0112] Step 3:

[0113] The server stores the received content.

[0114] The server receives the content sent from the terminal and temporarily stores it in a database. Specifically, the server checks the integrity of the received data and creates an entry in the database. The input is the encrypted data received, and the output is the content data stored in the database.

[0115] Step 4:

[0116] The server launches a generative AI model to analyze the content.

[0117] The server retrieves the stored content and launches the generative AI model to begin analysis. Specifically, in the case of video, the server extracts each frame and performs facial, object, and text recognition. In the case of images, it performs facial expression analysis and background checks, and in the case of text, it uses natural language processing (NLP) algorithms to perform sentiment analysis and detect discriminatory language. The input is the content stored on the server, and the output is a list of inappropriate content as the analysis result.

[0118] Step 5:

[0119] The server generates correction suggestions based on the analysis results.

[0120] The server determines the risk of outrage based on the analysis results obtained by the generative AI model and automatically generates specific correction suggestions. For example, it suggests blurring video frames containing inappropriate content or suggesting text changes to tone down offensive language. The input is the analysis results, and the output is a document containing specific correction suggestions.

[0121] Step 6:

[0122] The server sends the revision suggestions to the device.

[0123] The server sends the generated revision suggestions to the user's device. Specifically, the server formats the data and prepares it for transmission to the user's device. The input is the generated revision suggestions, and the output is data in a format that can be displayed in a user interface.

[0124] Step 7:

[0125] The terminal presents correction suggestions to the user.

[0126] The terminal presents the received revision suggestions to the user via a user interface. Specifically, the terminal receives data and displays it on the screen in a format that is easy for the user to understand. The input is the revision suggestion data sent from the server, and the output is the revision suggestions displayed on the user interface.

[0127] Step 8:

[0128] The user modifies the content based on the suggestions.

[0129] The user then modifies the content according to the suggested modifications. For example, they may use an image editing tool to apply a pixelated effect, or a text editor to change the wording. The input is the modification suggestions, and the output is the modified content.

[0130] Step 9:

[0131] The terminal sends the modified content back to the server.

[0132] The device then sends the modified content back to the server. Specifically, the data is sent securely again using an encryption protocol such as SSL / TLS. The input is the modified content, and the output is the encrypted data sent to the server.

[0133] Step 10:

[0134] The server reparses the modified content.

[0135] The server then re-analyzes the revised content to check for any new problems. Specifically, it launches the generative AI model again and performs the same analysis procedure. The input is the revised content, and the output is a report of the results of the re-analysis.

[0136] Step 11:

[0137] The server completes the final check and sends a post request to the social networking site.

[0138] If the server finds no new issues after re-analysis, it sends a post request to the social media platform. Specifically, it sends post data to the social media platform via API. The input is the final verified content, and the output is a post request sent to the social media platform.

[0139] Step 12:

[0140] The server sends a notification to the user that submission is complete.

[0141] The server notifies the user that the post to the SNS was successful. Specifically, it sends a message to the user's device using a notification API. The input is the status of the post completion, and the output is a notification message sent to the user's device.

[0142] (Application example 1)

[0143] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0144] Advertising content posted on social media and video platforms carries the risk of causing outrage and damaging brand image. Expressions that unconsciously offend consumers and content that may violate privacy are particularly problematic. The purpose of this invention is to detect these risks in advance and prompt appropriate corrections before posting, thereby enabling safe advertising while protecting brand image.

[0145] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0146] In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the results of the analysis, means for generating revision suggestions based on the determination results, means for presenting the revision suggestions to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to a social media platform or a video platform, means for analyzing the possibility that the content may have a negative impact on a brand image or consumer sentiment, and means for suggesting changes to inappropriate expressions. This allows advertising content to be posted safely without damaging the brand image.

[0147] "Means for receiving content entered by a user" refers to devices or software for receiving digital content such as videos, images, and text that a user wishes to post.

[0148] The "means for analyzing the content" refers to devices or software for analyzing the details of received content. Specifically, this includes video analysis, image analysis, and text analysis.

[0149] A "provocation risk assessment tool" is any device or software that assesses the risk that content will provoke public outrage or negative reaction based on the analyzed content.

[0150] The "means for generating correction suggestions" refers to a device or software that generates suggestions on how and which parts of content should be corrected in order to avoid backlash or damage to the brand image.

[0151] The "means for presenting revision suggestions to the user" refers to a device or software for notifying the user of the generated revision suggestions and visually displaying them.

[0152] The "means for receiving the content modified by the user again and performing a final check" refers to a device or software that receives the content after the user has made modifications and checks whether there are any new problems with the content.

[0153] "Means for posting final verified content to a social media platform or video platform" refers to the equipment or software used to actually post content that has been verified as having no problems to a social media platform or video platform.

[0154] "Means for analyzing the possibility of adversely affecting brand image or consumer sentiment" refers to devices or software for assessing the possibility that content may damage a company's brand image or cause negative emotions among consumers.

[0155] A "means for suggesting changes to inappropriate language" is a device or software that detects inappropriate language contained in content and suggests changing it to appropriate language.

[0156] This invention relates to a system that analyzes advertising content before it is posted on social media or video platforms to ensure that the advertisement does not have a negative impact on brand image or consumer sentiment, and makes suggestions for corrections as necessary.

[0157] System Configuration

[0158] The server has the following means:

[0159] 1. A means of receiving user-entered content

[0160] 2. Means for analyzing said content

[0161] 3. A means for determining the possibility of a controversy based on the results of the above analysis

[0162] 4. A means for generating a correction suggestion based on the result of the determination.

[0163] 5. Means for presenting said correction suggestions to the user

[0164] 6. A means for users to receive revised content again and perform a final review

[0165] 7. Posting the finalized content to a social media platform or video platform

[0166] 8. Means for analyzing the possibility that said content may have a negative impact on brand image or consumer sentiment

[0167] 9. How to suggest changes to inappropriate language

[0168] System Operation

[0169] 1. A means of receiving user-entered content

[0170] Users can input content to be posted through a smartphone application, and the input content is sent to the server by the smartphone.

[0171] 2. Means of content analysis

[0172] The server analyzes the received content using a generative AI model. The analysis results are as follows:

[0173] Video analysis: Extraction of video frames and frame-by-frame image analysis (e.g., face recognition, object recognition)

[0174] Image analysis: facial expression analysis and background check of images

[0175] Text analysis: Sentiment analysis and discriminatory language detection using natural language processing (NLP)

[0176] 3. How to determine the possibility of a controversy

[0177] The server determines whether there is a risk of a controversy based on the analysis results. For example, content containing offensive language or excessively negative emotions is identified as a risk of a controversy.

[0178] 4. A means of generating revision suggestions

[0179] If there is a possibility of a controversy, the server will generate specific suggestions for correction, such as "the image should be blurred" or "the offensive language should be toned down."

[0180] 5. A way to present suggested modifications to the user

[0181] The generated correction suggestions are presented to the user through the user interface of the smartphone application, where the user can review the suggestions and make any necessary corrections.

[0182] 6. A means for users to receive revised content again and perform a final review

[0183] The corrected content is sent back to the server, which re-analyzes it to check for any new problems.

[0184] 7. A means of posting the finalized content to a social media or video platform

[0185] Once the content has undergone final verification, it is posted from the server to a social media platform or video platform.

[0186] 8. A means of analyzing potential negative impacts on brand image and consumer sentiment

[0187] Specifically, for advertising content, analytical algorithms are used to determine whether it may have a negative impact on brand image or consumer sentiment, for example by detecting expressions that evoke negative emotions.

[0188] 9. How to suggest changes to inappropriate language

[0189] Based on the analysis results, the system suggests to the user how to change inappropriate language to appropriate ones, for example, by suggesting specific changes to the text to tone down offensive language.

[0190] Specific examples

[0191] For example, consider a case where a company uses AdSafeGuard to check an advertisement for a new product before posting it on social media. If the post contains offensive language such as "It's far better than other products," the app will detect this through sentiment analysis and offer suggestions to "tone down the offensive language." If the text entered is "It's foolish to use other companies' products. This product is better than any other!", the system will detect this and offer correction suggestions to the user.

[0192] In this way, advertising content can be posted safely without damaging the brand image.

[0193] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0194] Step 1:

[0195] A means of receiving user-entered content

[0196] Users input the advertising content (video, image, text) they wish to post through a smartphone application. The device sends this input data to the server, which then temporarily stores the received content. The data based on the input is stored on the server in its original format.

[0197] Step 2:

[0198] A means for analyzing the content

[0199] The server analyzes the received content using a generative AI model. For video, video frames are extracted and facial and object recognition is performed on each frame. For images, facial expression analysis and background checks are performed. For text, natural language processing (NLP) is used to perform sentiment analysis and discriminatory language detection. The input data (video, image, text) is analyzed and various features (face position, type of expression, emotion score, etc.) are extracted.

[0200] Step 3:

[0201] How to determine the possibility of a fire

[0202] The server determines the possibility of a controversy based on the analysis results. For example, it evaluates the results of facial recognition within video frames and the emotion score of the text. Based on the analysis results, if there is a high level of negative emotion (aggressive expressions, etc.), it determines that there is a possibility of a controversy and sets a risk level.

[0203] Step 4:

[0204] A means of generating revision suggestions

[0205] The server generates specific revision suggestions based on the level of risk of outrage, such as "images should be blurred" or "offensive language should be toned down." The server then generates a concrete action plan to mitigate the risk.

[0206] Step 5:

[0207] A means of presenting suggested revisions to the user

[0208] The server presents the generated correction suggestions to the user through the device's user interface. The user checks the displayed suggestions and makes necessary changes by operating their smartphone. Based on the presented suggestions, the user can then select an action to take.

[0209] Step 6:

[0210] A means for users to receive revised content again and perform a final review

[0211] After the user makes corrections, they send the corrected content back to the server. The server again uses the generative AI model to analyze the corrected content. A final check is made to see if there are any new problems with the corrections. A final check is made based on the results of the re-analysis after the corrections.

[0212] Step 7:

[0213] A means of posting the finalized content to a social media or video platform

[0214] Once the server has completed its final check, if the content is deemed to be safe, it will send a posting request to the social media platform or video platform. The content will then be posted, ensuring that the final checked data is safe.

[0215] Step 8:

[0216] A means of analyzing potential negative impacts on brand image and consumer sentiment

[0217] The server evaluates the content's potential to negatively impact brand image and consumer sentiment. Using sentiment analysis and facial expression recognition, the server identifies effective advertising language by determining whether the content is positive or negative. The server then analyzes the data to maintain brand image.

[0218] Step 9:

[0219] A way to suggest changes to inappropriate language

[0220] Based on the analysis results, the server will suggest changes if inappropriate language is found. For example, it will suggest specific text to tone down offensive language or suggest areas to modify images. Users can make corrections based on the suggestions, enabling them to create safe content that does not damage brand value.

[0221] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0222] Below is a description of the "Mode for Carrying Out the Invention" of the specification based on the claims of the invention combining emotion engines.

[0223] ---

[0224] This invention combines an emotion engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[0225] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0226] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, such as facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0227] In addition, an emotion engine is activated to recognize the user's emotions from the user's input. The server uses the emotion engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[0228] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[0229] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[0230] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[0231] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions for corrections such as "you should blur the image" and "tone down the offensive language." The user makes corrections based on these suggestions, and the post is finally completed successfully.

[0232] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[0233] ---

[0234] The above is a specific embodiment for carrying out the invention in which an emotion engine is combined.

[0235] The processing flow will be explained below.

[0236] MODE FOR CARRYING OUT THE INVENTION

[0237] This invention relates to a system that analyzes content posted by users on social networking sites, determines the possibility of the content becoming a controversy, and then proposes appropriate revisions by combining it with an emotion engine that recognizes the user's emotions. The specific processing flow is explained below, broken down into steps.

[0238] Step 1:

[0239] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[0240] Step 2:

[0241] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[0242] Step 3:

[0243] The server stores the received content and starts the generative AI model and emotion engine to begin analysis. The server separates the received data into various types of content and prepares them for analysis.

[0244] Step 4:

[0245] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[0246] Step 5:

[0247] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for controversy.

[0248] Step 6:

[0249] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[0250] Step 7:

[0251] The server uses a generative AI model and an emotion engine to determine the user's emotions (e.g., anger, joy, sadness) by analyzing the text and voice data entered by the user.

[0252] Step 8:

[0253] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[0254] Step 9:

[0255] To generate the correction suggestions, the server lists specific improvement measures, such as "proposing pixelation to protect the privacy of a specific person in the image" or "proposing changes to the text to tone down offensive language." Furthermore, the server adjusts the correction suggestions according to the user's emotions. For example, if the user is feeling angry, the server will suggest changes to the expression to tone down the emotion.

[0256] Step 10:

[0257] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[0258] Step 11:

[0259] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[0260] Step 12:

[0261] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[0262] Step 13:

[0263] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[0264] Step 14:

[0265] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[0266] ---

[0267] The above is a specific processing flow for implementing the invention that combines an emotion engine. Specific operations are clearly indicated at each step, allowing users to post to SNS more safely.

[0268] Example 2

[0269] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0270] The challenge is to prevent the risk of posts on SNS becoming controversial and provide a safe environment for use. In particular, there is a need for a system that can reduce users' psychological stress and allow them to use SNS with greater peace of mind by providing appropriate revision suggestions while taking users' emotions into consideration.

[0271] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the result of the analysis, means for generating a revision suggestion based on the determination result, means for presenting the revision suggestion to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to the SNS platform, and an emotion recognition engine for recognizing the user's emotions and reflecting them in the revision suggestion. This prevents the risk of a user's posted content causing a controversy before it happens, enabling safe and secure use of SNS.

[0272] Key Word Definitions

[0273] "User" refers to a person or entity who intends to use this system to post to the SNS and who inputs content.

[0274] "Content" is a general term for digital information such as videos, images, and text that users wish to post.

[0275] "Analysis" is the process performed by the server on the content it receives, evaluating and verifying the content to determine the possibility of it causing an uproar.

[0276] "Potential for a social media outcry" refers to the risk that a post will provoke a large number of negative reactions on social media, potentially having a harmful effect on users and third parties.

[0277] "Modification Suggestions" are specific changes or improvements generated by the server based on the analysis of the content to reduce the likelihood of a controversy.

[0278] An "emotion recognition engine" is an algorithm or technology that recognizes a user's emotions from the content or input data they are about to post, and adjusts suggested edits based on that emotional information.

[0279] "Generative AI Model" means an artificial intelligence model used by the server in the analysis process, including machine learning algorithms for evaluating and analyzing content.

[0280] "User interface" refers to the application or screen that a user operates to input content, suggest corrections, and perform final confirmation.

[0281] The "final check" is a process in which the server reanalyzes the content that has been corrected based on the user's suggested corrections and verifies that there are no problems.

[0282] "SNS platform" is a general term for social networking services that users use to share content.

[0283] MODE FOR CARRYING OUT THE INVENTION

[0284] This invention combines an emotion recognition engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[0285] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0286] The server stores the received content in a database and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame, specifically facial recognition, object recognition, and text recognition to check for inappropriate content. For images, the server analyzes the images, for example, performing facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0287] In addition, an emotion recognition engine is activated to recognize the user's emotions from the user's input. The server uses the emotion recognition engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[0288] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[0289] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[0290] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[0291] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion recognition engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0292] Example prompt sentence:

[0293] Prompt: Please explain the SNS post analysis system combined with the emotion engine. Please include the specific steps a user takes when trying to post content to SNS, the analysis method, the role of the emotion engine, the process of generating and presenting correction suggestions, and the steps to reanalyze and complete the post.

[0294] In this way, by using this system, users can prevent inappropriate posts on SNS and use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[0295] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0296] Program processing flow

[0297] Below, the program processing of this system will be explained in detail step by step.

[0298] Step 1:

[0299] The user inputs the content (video, image, text) they wish to post to SNS into a multi-function posting app.

[0300] Input: Digital information such as video, images, or text entered by a user.

[0301] Output: The multi-functional posting app stores the content data.

[0302] Step 2:

[0303] The terminal receives the content input by the user and transmits it to the server.

[0304] Input: Content entered into the user's device.

[0305] Data processing: The terminal converts the content data into a transfer format.

[0306] Output: The content data sent to the server.

[0307] Step 3:

[0308] The server stores the received content in a database and launches a generative AI model to begin analysis.

[0309] Input: Content data sent from the device.

[0310] Data processing: The server stores the content data in a database.

[0311] Output: Saved content data and activation signals for the generative AI model.

[0312] Step 4:

[0313] The server extracts each frame of the video and performs facial recognition, object recognition, and text recognition (in the case of video analysis).

[0314] Input: Stored video data.

[0315] Data processing / calculation: Extracting video frames and performing face, object, and text recognition on each frame.

[0316] Output: A list of frames containing inappropriate content as a result of execution.

[0317] Step 5:

[0318] The server analyzes the image and performs facial expression analysis and background checks (in the case of image analysis).

[0319] Input: Stored image data.

[0320] Data processing / computation: Facial expression analysis, background checks.

[0321] Output: Analysis results on facial expressions and background.

[0322] Step 6:

[0323] The server uses natural language processing (NLP) algorithms on the text data to perform sentiment analysis and detect discriminatory expressions (in the case of text analysis).

[0324] Input: Saved text data.

[0325] Data processing / calculation: Sentiment analysis using NLP algorithms, detection of discriminatory expressions.

[0326] Output: Sentiment analysis results and discriminatory expression detection results.

[0327] Step 7:

[0328] The server operates an emotion recognition engine based on the user's input data to determine the user's emotion.

[0329] Input: Text or voice data entered by the user.

[0330] Data processing / calculation: Emotion determination using an emotion recognition engine.

[0331] Output: User's emotion judgment result.

[0332] Step 8:

[0333] Based on the analysis results and the user's emotional assessment, the server identifies content that may cause controversy and generates specific suggestions for correction.

[0334] Input: Analysis results, user emotion judgment results.

[0335] Data processing / calculation: Listing problems and generating suggested fixes.

[0336] Output: A list of suggested fixes.

[0337] Step 9:

[0338] The server transmits the generated revision suggestions to the terminal and presents them to the user via the terminal's user interface.

[0339] Input: A list of correction suggestions.

[0340] Data processing / calculation: Sends suggested correction data to the terminal.

[0341] Output: Suggested fixes displayed on the terminal.

[0342] Step 10:

[0343] The user then corrects the posted content as necessary based on the suggested corrections.

[0344] Input: The suggested corrections displayed on the terminal.

[0345] Data manipulation / calculation: Modifying posted content using image editing tools or text editors.

[0346] Output: The modified content data.

[0347] Step 11:

[0348] The terminal transmits the corrected content to the server again.

[0349] Input: The modified content data.

[0350] Data processing / calculation: The corrected data is sent to the server.

[0351] Output: The modified content data sent to the server.

[0352] Step 12:

[0353] The server re-parses the modified content for any additional issues and performs a final check.

[0354] Input: The modified content data.

[0355] Data processing / calculation: Perform re-analysis process and re-check problems.

[0356] Output: The final verification result.

[0357] Step 13:

[0358] Once the server completes the final verification, it sends a request to post the secure content to the social media platform and actually completes the post.

[0359] Input: Final verification result.

[0360] Data processing / calculation: Generating and sending posting requests to social media platforms.

[0361] Output: The actual social media post completed.

[0362] (Application example 2)

[0363] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0364] Flaming and inappropriate posts on social media and other online platforms have become serious problems for companies and individuals. In addition, posts that strongly reflect the poster's emotions can easily lead to misunderstandings, which can cause serious problems, especially for official corporate accounts. Furthermore, conventional systems generate revision suggestions without taking the user's emotions into account, often resulting in specific and appropriate revision suggestions that do not reflect the user's intentions. For this reason, there is a growing demand for systems that can recognize user emotions, predict potential flame wars in advance, and make appropriate revision suggestions.

[0365] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy, means for generating revision suggestions, means for presenting the content to the user, means for receiving the content revised by the user again and performing a final check, means for recognizing the user's emotions and generating revision suggestions according to the emotions, means for analyzing the user's emotions using an emotion engine, and means for posting the finally checked content to an SNS platform. This makes it possible to make appropriate revision suggestions taking the user's emotions into consideration, and as a result, it is possible to reduce the risk of controversy and trouble on SNS.

[0366] "User" means a person who uses the system to create or post content.

[0367] "Content" is a general term for digital data such as text, images, and videos that users wish to post on social media platforms.

[0368] A "means for analyzing content" is a system that includes processes and algorithms for analyzing received content and identifying problems and characteristics of the content.

[0369] The "means for determining the possibility of a controversy" is a system that evaluates and determines the risk of a post causing a controversy on social media or in online communities based on the analysis results.

[0370] The "means for generating correction suggestions" refers to a process or system for suggesting appropriate corrections to the user based on the determined risk of a social media outrage.

[0371] The "means for presenting revision suggestions to the user" refers to an interface or display means for displaying the generated revision suggestions on the user's device so that the user can confirm them.

[0372] The "means for final confirmation" is a system that allows the user to recheck the content that has been revised to ensure there are no problems before final posting.

[0373] The "means for recognizing emotions and generating correction suggestions according to those emotions" is a system that analyzes the user's emotions and generates specific correction suggestions that will alleviate the emotions based on the results.

[0374] An "emotion engine" refers to an algorithm or model that analyzes a user's emotional state from input text or voice and detects specific emotions.

[0375] "Means of posting to social media platforms" refers to the system for actually posting the final, verified, safe content to social media or online platforms.

[0376] This invention is a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes revision suggestions taking into account the user's emotions. This system includes a series of processes, from receiving content entered by a user, analyzing it, determining the possibility of a controversy, analyzing the user's emotions using an emotion engine, generating revision suggestions based on those emotions, to finally posting the confirmed content to the social networking site platform.

[0377] The system has a means to receive content from the user's device. This content consists of digital data such as text, images, and video. The device then sends this content to the server, which stores it and starts analyzing it by launching a generative AI model.

[0378] Specifically, the server performs the following processes: For video, the server first extracts video frames and performs face, object, and text recognition for each frame; for images, it performs image analysis, facial expression analysis, and background checks; and for text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0379] In addition, the server uses an emotion engine to determine the user's emotions from the user's input. This emotion engine utilizes emotion analysis tools such as Hugging Face transformers and TextBlob to analyze the user's emotions from text and voice data. For example, if the user has strong emotions such as anger or anxiety, the server generates correction suggestions based on those emotions. These correction suggestions may include changes to the expression that will soften the user's emotions or more objective ways of expressing them.

[0380] Once the analysis identifies a potential flaming incident, the server generates specific suggestions for correction. The suggestions are tailored based on the user's emotions. For example, if the user is determined to be "angry," the server will provide suggestions for modifying the expression to tone down that emotion. The suggestions are sent to the user's device and presented via a user interface.

[0381] The user then corrects the content based on the suggested corrections and sends it back to the server via their device. The server then analyzes the corrected content again to check for any new issues. After the final check is complete, the server sends a request to the social media platform to post the safe content, and the post is then completed.

[0382] As a specific example, consider the case where a user tries to post "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends them to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0383] An example of a prompt for a generative AI model is the following text:

[0384] This mobile app analyzes the text of social media posts in real time and generates revision suggestions based on the user's emotional state. If the post is offensive, it will make suggestions to tone down the expression. For example, if someone posts "This product is completely useless and pointless!", the emotion engine will identify this as "anger" and suggest a more neutral and constructive revision such as "This product may not have worked for me, but it may be useful for others."

[0385] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0386] Step 1:

[0387] The device receives the content (text, images, videos) entered by the user and sends it to the server. The entered content may be a specific sentence for text, a JPEG file for images, or an MP4 file for videos. This data processing involves serializing the content so that it can be sent to the server in the appropriate format.

[0388] Step 2:

[0389] The server saves the received content and starts the generative AI model to begin analysis. The content data mentioned above is used as input. In the case of video, the server extracts video frames and saves them as images. In the case of images or text, the server directly begins analyzing them. This data processing involves extracting and saving video frames.

[0390] Step 3:

[0391] The server performs face recognition, object recognition, and text recognition for each video frame. Each frame of video data is used as input. The output is the location of recognized faces and objects, as well as text content. This data calculation involves running image analysis algorithms and calculating the features of faces and objects.

[0392] Step 4:

[0393] The server performs image analysis, including facial expression analysis and background checks. Image data and analysis algorithms are used as input. Analysis results (e.g., a determination of the risk of privacy violations or whether the background contains inappropriate content) are obtained as output. This data processing includes calculating facial characteristics using an expression analysis algorithm and evaluating the content using a background check algorithm.

[0394] Step 5:

[0395] The server performs text analysis and uses natural language processing (NLP) algorithms to perform sentiment analysis and discriminatory expression detection. The input is text data, and the output is a sentiment score and a list of detected discriminatory expressions. This data operation uses an NLP model to analyze the text and identify sentiment and discriminatory expressions.

[0396] Step 6:

[0397] The server uses an emotion engine to determine the user's emotions. The input is the user's input data (text, voice, etc.), and the output is the result of the user's emotional state (e.g., "anger," "anxiety," etc.). This data calculation is performed by analyzing the user's emotions using an emotion analysis model.

[0398] Step 7:

[0399] The server determines the possibility of a controversy based on the analysis results. The input is the various analysis data obtained in the previous steps (face recognition results, object recognition results, emotion scores, etc.), and the output is the evaluation result of the controversy risk. This data calculation integrates each acquired data and performs a risk assessment.

[0400] Step 8:

[0401] Based on the results of the assessment, the server generates revision suggestions that take the user's emotions into account. The input is the flame risk assessment and the user's emotional data, and the output is a list of specific revision suggestions. This data calculation adjusts the content according to the user's emotions.

[0402] Step 9:

[0403] The server sends the generated revision suggestions to the terminal and presents them to the user via a user interface. The input is the generated revision suggestion data, and the output is the revision suggestions displayed on the user interface. This data processing involves sending the suggestion content to the terminal in an appropriate format.

[0404] Step 10:

[0405] The user modifies the content based on the suggested modifications and then sends the modified content back to the server from the device. The input is the modifications made by the user, and the output is the modified content. This data processing is a serialization process to ensure that the modifications are sent appropriately to the server.

[0406] Step 11:

[0407] The server re-analyzes the modified content to check for any new issues. The input is the modified content data, and the output is the result of the final analysis. This data calculation applies the re-analysis algorithm to check for any new issues.

[0408] Step 12:

[0409] The server posts the content that has been finalized to the SNS platform. The input is the finalized content data, and the output is the URL of the content posted to the SNS platform. This data processing involves posting the content to the SNS platform in the appropriate format.

[0410] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0411] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0412] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0413] [Second embodiment]

[0414] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0415] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0416] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0417] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0418] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0419] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0420] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0421] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0422] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0423] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0424] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0425] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0426] The following is a description of the preferred embodiment of the invention based on the claims.

[0427] ---

[0428] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[0429] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0430] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, including facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0431] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[0432] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[0433] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[0434] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0435] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By striving to analyze content quickly and accurately and suggest corrections, it is expected to reduce the risk of online outrage.

[0436] ---

[0437] The above is a specific example of the "mode for carrying out the invention."

[0438] The processing flow will be explained below.

[0439] Step 1:

[0440] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[0441] Step 2:

[0442] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[0443] Step 3:

[0444] The server stores the received content and starts the generative AI model to begin analysis. The server separates the received data by content and prepares it for analysis.

[0445] Step 4:

[0446] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[0447] Step 5:

[0448] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for inflammatory content.

[0449] Step 6:

[0450] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[0451] Step 7:

[0452] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[0453] Step 8:

[0454] To generate the suggested corrections, the server lists specific improvements, such as "suggesting blurring to protect the privacy of a specific person in the image" or "suggesting changes to the text to tone down offensive language."

[0455] Step 9:

[0456] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[0457] Step 10:

[0458] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[0459] Step 11:

[0460] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[0461] Step 12:

[0462] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[0463] Step 13:

[0464] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[0465] The above is a specific program processing flow based on the claims.

[0466] Example 1

[0467] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0468] In today's social networking environment, there is an increasing risk that users may unintentionally post inappropriate content, sparking a social media firestorm. This issue can have serious consequences for users, as it can include violations of personal privacy and offensive language. Conventional methods require users to manually review and correct posts, which not only takes time and effort but also carries the risk of false positives. There is a need for a system that can solve these issues and enable users to post appropriate content safely and quickly.

[0469] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0470] In this invention, the server includes means for receiving content entered by a user, means for transmitting the content to the server, means for saving the content, means for analyzing the content by activating a generative AI model, means for determining the possibility of a controversy, means for generating revision suggestions, means for transmitting the revision suggestions to a terminal, means for presenting the revision suggestions to the user via a user interface, means for re-receiving and re-analyzing the content revised by the user, and means for posting the final confirmed content to an SNS platform. This allows users to automatically determine the risk of content causing a controversy and quickly receive revision suggestions, thereby realizing safe posting.

[0471] A "generative AI model" is an artificial intelligence model trained using machine learning algorithms to analyze and classify input content.

[0472] "Content" refers to all digital data posted on social media, including videos, images, and text.

[0473] A "server" is a part of a computer system that stores, processes, and distributes data over a network.

[0474] "Terminal" refers to a device that is directly operated by a user, such as a smartphone or a personal computer.

[0475] "Analysis" is the process of breaking down the information contained in content and evaluating it for a specific purpose.

[0476] "Potential for a social media outrage" refers to the risk that the content posted will be deemed inappropriate and cause negative reactions or uproar on social media.

[0477] "Revision suggestions" is a means of generating specific revision suggestions for the posted content based on the analysis results.

[0478] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[0479] "Re-analysis" is the process of re-analyzing content that has been modified by a user to check for new problems.

[0480] "Post Request" means a request to post Final Reviewed Content to a Social Media Platform.

[0481] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[0482] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0483] The server stores the received content and starts analyzing it by activating a generative AI model, which is an artificial intelligence model trained using machine learning algorithms to analyze and classify the input content.

[0484] For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial, object, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, facial expression analysis, and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0485] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[0486] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[0487] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[0488] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0489] This system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By quickly and accurately analyzing content and suggesting corrections, it is expected to reduce the risk of online outrage.

[0490] An example of a prompt sentence is, "I want to post on social media about yesterday's drinking party. For example, it should include the following content: Photo: A photo of the drinking party (faces are clearly visible), Text: 'We drank too much last night and got rowdy!' Please check whether the content of this post poses a risk of causing a backlash and let me know any suggestions for corrections."

[0491] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0492] Step 1:

[0493] A user inputs content into a multi-function posting app.

[0494] Specifically, a user uploads a video, image, or text to the app to post on a social network. The input is the data (video, image, text) that is stored within the application. The output is the content that is transformed into a structured format within the application to be sent to the next step.

[0495] Step 2:

[0496] The terminal receives the content and transmits it to the server.

[0497] The device sends the video, image, or text received from the multi-function posting app to the server. Specifically, the device securely transmits data using encryption protocols such as SSL / TLS. The input is the content uploaded by the user, and the output is the encrypted data sent to the server.

[0498] Step 3:

[0499] The server stores the received content.

[0500] The server receives the content sent from the terminal and temporarily stores it in a database. Specifically, the server checks the integrity of the received data and creates an entry in the database. The input is the encrypted data received, and the output is the content data stored in the database.

[0501] Step 4:

[0502] The server launches a generative AI model to analyze the content.

[0503] The server retrieves the stored content and launches the generative AI model to begin analysis. Specifically, in the case of video, the server extracts each frame and performs facial, object, and text recognition. In the case of images, it performs facial expression analysis and background checks, and in the case of text, it uses natural language processing (NLP) algorithms to perform sentiment analysis and detect discriminatory language. The input is the content stored on the server, and the output is a list of inappropriate content as the analysis result.

[0504] Step 5:

[0505] The server generates correction suggestions based on the analysis results.

[0506] The server determines the risk of outrage based on the analysis results obtained by the generative AI model and automatically generates specific correction suggestions. For example, it suggests blurring video frames containing inappropriate content or suggesting text changes to tone down offensive language. The input is the analysis results, and the output is a document containing specific correction suggestions.

[0507] Step 6:

[0508] The server sends the revision suggestions to the device.

[0509] The server sends the generated revision suggestions to the user's device. Specifically, the server formats the data and prepares it for transmission to the user's device. The input is the generated revision suggestions, and the output is data in a format that can be displayed in a user interface.

[0510] Step 7:

[0511] The terminal presents correction suggestions to the user.

[0512] The terminal presents the received revision suggestions to the user via a user interface. Specifically, the terminal receives data and displays it on the screen in a format that is easy for the user to understand. The input is the revision suggestion data sent from the server, and the output is the revision suggestions displayed on the user interface.

[0513] Step 8:

[0514] The user modifies the content based on the suggestions.

[0515] The user then modifies the content according to the suggested modifications. For example, they may use an image editing tool to apply a pixelated effect, or a text editor to change the wording. The input is the modification suggestions, and the output is the modified content.

[0516] Step 9:

[0517] The terminal sends the modified content back to the server.

[0518] The device then sends the modified content back to the server. Specifically, the data is sent securely again using an encryption protocol such as SSL / TLS. The input is the modified content, and the output is the encrypted data sent to the server.

[0519] Step 10:

[0520] The server reparses the modified content.

[0521] The server then re-analyzes the revised content to check for any new problems. Specifically, it launches the generative AI model again and performs the same analysis procedure. The input is the revised content, and the output is a report of the results of the re-analysis.

[0522] Step 11:

[0523] The server completes the final check and sends a post request to the social networking site.

[0524] If the server finds no new issues after re-analysis, it sends a post request to the social media platform. Specifically, it sends post data to the social media platform via API. The input is the final verified content, and the output is a post request sent to the social media platform.

[0525] Step 12:

[0526] The server sends a notification to the user that submission is complete.

[0527] The server notifies the user that the post to the SNS was successful. Specifically, it sends a message to the user's device using a notification API. The input is the status of the post completion, and the output is a notification message sent to the user's device.

[0528] (Application example 1)

[0529] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0530] Advertising content posted on social media and video platforms carries the risk of causing outrage and damaging brand image. Expressions that unconsciously offend consumers and content that may violate privacy are particularly problematic. The purpose of this invention is to detect these risks in advance and prompt appropriate corrections before posting, thereby enabling safe advertising while protecting brand image.

[0531] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0532] In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the results of the analysis, means for generating revision suggestions based on the determination results, means for presenting the revision suggestions to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to a social media platform or a video platform, means for analyzing the possibility that the content may have a negative impact on a brand image or consumer sentiment, and means for suggesting changes to inappropriate expressions. This allows advertising content to be posted safely without damaging the brand image.

[0533] "Means for receiving content entered by a user" refers to devices or software for receiving digital content such as videos, images, and text that a user wishes to post.

[0534] The "means for analyzing the content" refers to devices or software for analyzing the details of received content. Specifically, this includes video analysis, image analysis, and text analysis.

[0535] A "provocation risk assessment tool" is any device or software that assesses the risk that content will provoke public outrage or negative reaction based on the analyzed content.

[0536] The "means for generating correction suggestions" refers to a device or software that generates suggestions on how and which parts of content should be corrected in order to avoid backlash or damage to the brand image.

[0537] The "means for presenting revision suggestions to the user" refers to a device or software for notifying the user of the generated revision suggestions and visually displaying them.

[0538] The "means for receiving the content modified by the user again and performing a final check" refers to a device or software that receives the content after the user has made modifications and checks whether there are any new problems with the content.

[0539] "Means for posting final verified content to a social media platform or video platform" refers to the equipment or software used to actually post content that has been verified as having no problems to a social media platform or video platform.

[0540] "Means for analyzing the possibility of adversely affecting brand image or consumer sentiment" refers to devices or software for assessing the possibility that content may damage a company's brand image or cause negative emotions among consumers.

[0541] A "means for suggesting changes to inappropriate language" is a device or software that detects inappropriate language contained in content and suggests changing it to appropriate language.

[0542] This invention relates to a system that analyzes advertising content before it is posted on social media or video platforms to ensure that the advertisement does not have a negative impact on brand image or consumer sentiment, and makes suggestions for corrections as necessary.

[0543] System Configuration

[0544] The server has the following means:

[0545] 1. A means of receiving user-entered content

[0546] 2. Means for analyzing said content

[0547] 3. A means for determining the possibility of a controversy based on the results of the above analysis

[0548] 4. A means for generating a correction suggestion based on the result of the determination.

[0549] 5. Means for presenting said correction suggestions to the user

[0550] 6. A means for users to receive revised content again and perform a final review

[0551] 7. Posting the finalized content to a social media platform or video platform

[0552] 8. Means for analyzing the possibility that said content may have a negative impact on brand image or consumer sentiment

[0553] 9. How to suggest changes to inappropriate language

[0554] System Operation

[0555] 1. A means of receiving user-entered content

[0556] Users can input content to be posted through a smartphone application, and the input content is sent to the server by the smartphone.

[0557] 2. Means of content analysis

[0558] The server analyzes the received content using a generative AI model. The analysis results are as follows:

[0559] Video analysis: Extraction of video frames and frame-by-frame image analysis (e.g., face recognition, object recognition)

[0560] Image analysis: facial expression analysis and background check of images

[0561] Text analysis: Sentiment analysis and discriminatory language detection using natural language processing (NLP)

[0562] 3. How to determine the possibility of a controversy

[0563] The server determines whether there is a risk of a controversy based on the analysis results. For example, content containing offensive language or excessively negative emotions is identified as a risk of a controversy.

[0564] 4. A means of generating revision suggestions

[0565] If there is a possibility of a controversy, the server will generate specific suggestions for correction, such as "the image should be blurred" or "the offensive language should be toned down."

[0566] 5. A way to present suggested modifications to the user

[0567] The generated correction suggestions are presented to the user through the user interface of the smartphone application, where the user can review the suggestions and make any necessary corrections.

[0568] 6. A means for users to receive revised content again and perform a final review

[0569] The corrected content is sent back to the server, which re-analyzes it to check for any new problems.

[0570] 7. A means of posting the finalized content to a social media or video platform

[0571] Once the content has undergone final verification, it is posted from the server to a social media platform or video platform.

[0572] 8. A means of analyzing potential negative impacts on brand image and consumer sentiment

[0573] Specifically, for advertising content, analytical algorithms are used to determine whether it may have a negative impact on brand image or consumer sentiment, for example by detecting expressions that evoke negative emotions.

[0574] 9. How to suggest changes to inappropriate language

[0575] Based on the analysis results, the system suggests to the user how to change inappropriate language to appropriate ones, for example, by suggesting specific changes to the text to tone down offensive language.

[0576] Specific examples

[0577] For example, consider a case where a company uses AdSafeGuard to check an advertisement for a new product before posting it on social media. If the post contains offensive language such as "It's far better than other products," the app will detect this through sentiment analysis and offer suggestions to "tone down the offensive language." If the text entered is "It's foolish to use other companies' products. This product is better than any other!", the system will detect this and offer correction suggestions to the user.

[0578] In this way, advertising content can be posted safely without damaging the brand image.

[0579] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0580] Step 1:

[0581] A means of receiving user-entered content

[0582] Users input the advertising content (video, image, text) they wish to post through a smartphone application. The device sends this input data to the server, which then temporarily stores the received content. The data based on the input is stored on the server in its original format.

[0583] Step 2:

[0584] A means for analyzing the content

[0585] The server analyzes the received content using a generative AI model. For video, video frames are extracted and facial and object recognition is performed on each frame. For images, facial expression analysis and background checks are performed. For text, natural language processing (NLP) is used to perform sentiment analysis and discriminatory language detection. The input data (video, image, text) is analyzed and various features (face position, type of expression, emotion score, etc.) are extracted.

[0586] Step 3:

[0587] How to determine the possibility of a fire

[0588] The server determines the possibility of a controversy based on the analysis results. For example, it evaluates the results of facial recognition within video frames and the emotion score of the text. Based on the analysis results, if there is a high level of negative emotion (aggressive expressions, etc.), it determines that there is a possibility of a controversy and sets a risk level.

[0589] Step 4:

[0590] A means of generating revision suggestions

[0591] The server generates specific revision suggestions based on the level of risk of outrage, such as "images should be blurred" or "offensive language should be toned down." The server then generates a concrete action plan to mitigate the risk.

[0592] Step 5:

[0593] A means of presenting suggested revisions to the user

[0594] The server presents the generated correction suggestions to the user through the device's user interface. The user checks the displayed suggestions and makes necessary changes by operating their smartphone. Based on the presented suggestions, the user can then select an action to take.

[0595] Step 6:

[0596] A means for users to receive revised content again and perform a final review

[0597] After the user makes corrections, they send the corrected content back to the server. The server again uses the generative AI model to analyze the corrected content. A final check is made to see if there are any new problems with the corrections. A final check is made based on the results of the re-analysis after the corrections.

[0598] Step 7:

[0599] A means of posting the finalized content to a social media or video platform

[0600] Once the server has completed its final check, if the content is deemed to be safe, it will send a posting request to the social media platform or video platform. The content will then be posted, ensuring that the final checked data is safe.

[0601] Step 8:

[0602] A means of analyzing potential negative impacts on brand image and consumer sentiment

[0603] The server evaluates the content's potential to negatively impact brand image and consumer sentiment. Using sentiment analysis and facial expression recognition, the server identifies effective advertising language by determining whether the content is positive or negative. The server then analyzes the data to maintain brand image.

[0604] Step 9:

[0605] A way to suggest changes to inappropriate language

[0606] Based on the analysis results, the server will suggest changes if inappropriate language is found. For example, it will suggest specific text to tone down offensive language or suggest areas to modify images. Users can make corrections based on the suggestions, enabling them to create safe content that does not damage brand value.

[0607] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0608] Below is a description of the "Mode for Carrying Out the Invention" of the specification based on the claims of the invention combining emotion engines.

[0609] ---

[0610] This invention combines an emotion engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[0611] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0612] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, such as facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0613] In addition, an emotion engine is activated to recognize the user's emotions from the user's input. The server uses the emotion engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[0614] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[0615] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[0616] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[0617] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions for corrections such as "you should blur the image" and "tone down the offensive language." The user makes corrections based on these suggestions, and the post is finally completed successfully.

[0618] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[0619] ---

[0620] The above is a specific embodiment for carrying out the invention in which an emotion engine is combined.

[0621] The processing flow will be explained below.

[0622] MODE FOR CARRYING OUT THE INVENTION

[0623] This invention relates to a system that analyzes content posted by users on social networking sites, determines the possibility of the content becoming a controversy, and then proposes appropriate revisions by combining it with an emotion engine that recognizes the user's emotions. The specific processing flow is explained below, broken down into steps.

[0624] Step 1:

[0625] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[0626] Step 2:

[0627] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[0628] Step 3:

[0629] The server stores the received content and starts the generative AI model and emotion engine to begin analysis. The server separates the received data into various types of content and prepares them for analysis.

[0630] Step 4:

[0631] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[0632] Step 5:

[0633] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for controversy.

[0634] Step 6:

[0635] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[0636] Step 7:

[0637] The server uses a generative AI model and an emotion engine to determine the user's emotions (e.g., anger, joy, sadness) by analyzing the text and voice data entered by the user.

[0638] Step 8:

[0639] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[0640] Step 9:

[0641] To generate the correction suggestions, the server lists specific improvement measures, such as "proposing pixelation to protect the privacy of a specific person in the image" or "proposing changes to the text to tone down offensive language." Furthermore, the server adjusts the correction suggestions according to the user's emotions. For example, if the user is feeling angry, the server will suggest changes to the expression to tone down the emotion.

[0642] Step 10:

[0643] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[0644] Step 11:

[0645] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[0646] Step 12:

[0647] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[0648] Step 13:

[0649] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[0650] Step 14:

[0651] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[0652] ---

[0653] The above is a specific processing flow for implementing the invention that combines an emotion engine. Specific operations are clearly indicated at each step, allowing users to post to SNS more safely.

[0654] Example 2

[0655] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0656] The challenge is to prevent the risk of posts on SNS becoming controversial and provide a safe environment for use. In particular, there is a need for a system that can reduce users' psychological stress and allow them to use SNS with greater peace of mind by providing appropriate revision suggestions while taking users' emotions into consideration.

[0657] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the result of the analysis, means for generating a revision suggestion based on the determination result, means for presenting the revision suggestion to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to the SNS platform, and an emotion recognition engine for recognizing the user's emotions and reflecting them in the revision suggestion. This prevents the risk of a user's posted content causing a controversy before it happens, enabling safe and secure use of SNS.

[0658] Key Word Definitions

[0659] "User" refers to a person or entity who intends to use this system to post to the SNS and who inputs content.

[0660] "Content" is a general term for digital information such as videos, images, and text that users wish to post.

[0661] "Analysis" is the process performed by the server on the content it receives, evaluating and verifying the content to determine the possibility of it causing an uproar.

[0662] "Potential for a social media outcry" refers to the risk that a post will provoke a large number of negative reactions on social media, potentially having a harmful effect on users and third parties.

[0663] "Modification Suggestions" are specific changes or improvements generated by the server based on the analysis of the content to reduce the likelihood of a controversy.

[0664] An "emotion recognition engine" is an algorithm or technology that recognizes a user's emotions from the content or input data they are about to post, and adjusts suggested edits based on that emotional information.

[0665] "Generative AI Model" means an artificial intelligence model used by the server in the analysis process, including machine learning algorithms for evaluating and analyzing content.

[0666] "User interface" refers to the application or screen that a user operates to input content, suggest corrections, and perform final confirmation.

[0667] The "final check" is a process in which the server reanalyzes the content that has been corrected based on the user's suggested corrections and verifies that there are no problems.

[0668] "SNS platform" is a general term for social networking services that users use to share content.

[0669] MODE FOR CARRYING OUT THE INVENTION

[0670] This invention combines an emotion recognition engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[0671] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0672] The server stores the received content in a database and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame, specifically facial recognition, object recognition, and text recognition to check for inappropriate content. For images, the server analyzes the images, for example, performing facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0673] In addition, an emotion recognition engine is activated to recognize the user's emotions from the user's input. The server uses the emotion recognition engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[0674] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[0675] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[0676] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[0677] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion recognition engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0678] Example prompt sentence:

[0679] Prompt: Please explain the SNS post analysis system combined with the emotion engine. Please include the specific steps a user takes when trying to post content to SNS, the analysis method, the role of the emotion engine, the process of generating and presenting correction suggestions, and the steps to reanalyze and complete the post.

[0680] In this way, by using this system, users can prevent inappropriate posts on SNS and use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[0681] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0682] Program processing flow

[0683] Below, the program processing of this system will be explained in detail step by step.

[0684] Step 1:

[0685] The user inputs the content (video, image, text) they wish to post to SNS into a multi-function posting app.

[0686] Input: Digital information such as video, images, or text entered by a user.

[0687] Output: The multi-functional posting app stores the content data.

[0688] Step 2:

[0689] The terminal receives the content input by the user and transmits it to the server.

[0690] Input: Content entered into the user's device.

[0691] Data processing: The terminal converts the content data into a transfer format.

[0692] Output: The content data sent to the server.

[0693] Step 3:

[0694] The server stores the received content in a database and launches a generative AI model to begin analysis.

[0695] Input: Content data sent from the device.

[0696] Data processing: The server stores the content data in a database.

[0697] Output: Saved content data and activation signals for the generative AI model.

[0698] Step 4:

[0699] The server extracts each frame of the video and performs facial recognition, object recognition, and text recognition (in the case of video analysis).

[0700] Input: Stored video data.

[0701] Data processing / calculation: Extracting video frames and performing face, object, and text recognition on each frame.

[0702] Output: A list of frames containing inappropriate content as a result of execution.

[0703] Step 5:

[0704] The server analyzes the image and performs facial expression analysis and background checks (in the case of image analysis).

[0705] Input: Stored image data.

[0706] Data processing / computation: Facial expression analysis, background checks.

[0707] Output: Analysis results on facial expressions and background.

[0708] Step 6:

[0709] The server uses natural language processing (NLP) algorithms on the text data to perform sentiment analysis and detect discriminatory expressions (in the case of text analysis).

[0710] Input: Saved text data.

[0711] Data processing / calculation: Sentiment analysis using NLP algorithms, detection of discriminatory expressions.

[0712] Output: Sentiment analysis results and discriminatory expression detection results.

[0713] Step 7:

[0714] The server operates an emotion recognition engine based on the user's input data to determine the user's emotion.

[0715] Input: Text or voice data entered by the user.

[0716] Data processing / calculation: Emotion determination using an emotion recognition engine.

[0717] Output: User's emotion judgment result.

[0718] Step 8:

[0719] Based on the analysis results and the user's emotional assessment, the server identifies content that may cause controversy and generates specific suggestions for correction.

[0720] Input: Analysis results, user emotion judgment results.

[0721] Data processing / calculation: Listing problems and generating suggested fixes.

[0722] Output: A list of suggested fixes.

[0723] Step 9:

[0724] The server transmits the generated revision suggestions to the terminal and presents them to the user via the terminal's user interface.

[0725] Input: A list of correction suggestions.

[0726] Data processing / calculation: Sends suggested correction data to the terminal.

[0727] Output: Suggested fixes displayed on the terminal.

[0728] Step 10:

[0729] The user then corrects the posted content as necessary based on the suggested corrections.

[0730] Input: The suggested corrections displayed on the terminal.

[0731] Data manipulation / calculation: Modifying posted content using image editing tools or text editors.

[0732] Output: The modified content data.

[0733] Step 11:

[0734] The terminal transmits the corrected content to the server again.

[0735] Input: The modified content data.

[0736] Data processing / calculation: The corrected data is sent to the server.

[0737] Output: The modified content data sent to the server.

[0738] Step 12:

[0739] The server re-parses the modified content for any additional issues and performs a final check.

[0740] Input: The modified content data.

[0741] Data processing / calculation: Perform re-analysis process and re-check problems.

[0742] Output: The final verification result.

[0743] Step 13:

[0744] Once the server completes the final verification, it sends a request to post the secure content to the social media platform and actually completes the post.

[0745] Input: Final verification result.

[0746] Data processing / calculation: Generating and sending posting requests to social media platforms.

[0747] Output: The actual social media post completed.

[0748] (Application example 2)

[0749] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0750] Flaming and inappropriate posts on social media and other online platforms have become serious problems for companies and individuals. In addition, posts that strongly reflect the poster's emotions can easily lead to misunderstandings, which can cause serious problems, especially for official corporate accounts. Furthermore, conventional systems generate revision suggestions without taking the user's emotions into account, often resulting in specific and appropriate revision suggestions that do not reflect the user's intentions. For this reason, there is a growing demand for systems that can recognize user emotions, predict potential flame wars in advance, and make appropriate revision suggestions.

[0751] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy, means for generating revision suggestions, means for presenting the content to the user, means for receiving the content revised by the user again and performing a final check, means for recognizing the user's emotions and generating revision suggestions according to the emotions, means for analyzing the user's emotions using an emotion engine, and means for posting the finally checked content to an SNS platform. This makes it possible to make appropriate revision suggestions taking the user's emotions into consideration, and as a result, it is possible to reduce the risk of controversy and trouble on SNS.

[0752] "User" means a person who uses the system to create or post content.

[0753] "Content" is a general term for digital data such as text, images, and videos that users wish to post on social media platforms.

[0754] A "means for analyzing content" is a system that includes processes and algorithms for analyzing received content and identifying problems and characteristics of the content.

[0755] The "means for determining the possibility of a controversy" is a system that evaluates and determines the risk of a post causing a controversy on social media or in online communities based on the analysis results.

[0756] The "means for generating correction suggestions" refers to a process or system for suggesting appropriate corrections to the user based on the determined risk of a social media outrage.

[0757] The "means for presenting revision suggestions to the user" refers to an interface or display means for displaying the generated revision suggestions on the user's device so that the user can confirm them.

[0758] The "means for final confirmation" is a system that allows the user to recheck the content that has been revised to ensure there are no problems before final posting.

[0759] The "means for recognizing emotions and generating correction suggestions according to those emotions" is a system that analyzes the user's emotions and generates specific correction suggestions that will alleviate the emotions based on the results.

[0760] An "emotion engine" refers to an algorithm or model that analyzes a user's emotional state from input text or voice and detects specific emotions.

[0761] "Means of posting to social media platforms" refers to the system for actually posting the final, verified, safe content to social media or online platforms.

[0762] This invention is a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes revision suggestions taking into account the user's emotions. This system includes a series of processes, from receiving content entered by a user, analyzing it, determining the possibility of a controversy, analyzing the user's emotions using an emotion engine, generating revision suggestions based on those emotions, to finally posting the confirmed content to the social networking site platform.

[0763] The system has a means to receive content from the user's device. This content consists of digital data such as text, images, and video. The device then sends this content to the server, which stores it and starts analyzing it by launching a generative AI model.

[0764] Specifically, the server performs the following processes: For video, the server first extracts video frames and performs face, object, and text recognition for each frame; for images, it performs image analysis, facial expression analysis, and background checks; and for text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0765] In addition, the server uses an emotion engine to determine the user's emotions from the user's input. This emotion engine utilizes emotion analysis tools such as Hugging Face transformers and TextBlob to analyze the user's emotions from text and voice data. For example, if the user has strong emotions such as anger or anxiety, the server generates correction suggestions based on those emotions. These correction suggestions may include changes to the expression that will soften the user's emotions or more objective ways of expressing them.

[0766] Once the analysis identifies a potential flaming incident, the server generates specific suggestions for correction. The suggestions are tailored based on the user's emotions. For example, if the user is determined to be "angry," the server will provide suggestions for modifying the expression to tone down that emotion. The suggestions are sent to the user's device and presented via a user interface.

[0767] The user then corrects the content based on the suggested corrections and sends it back to the server via their device. The server then analyzes the corrected content again to check for any new issues. After the final check is complete, the server sends a request to the social media platform to post the safe content, and the post is then completed.

[0768] As a specific example, consider the case where a user tries to post "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends them to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0769] An example of a prompt for a generative AI model is the following text:

[0770] This mobile app analyzes the text of social media posts in real time and generates revision suggestions based on the user's emotional state. If the post is offensive, it will make suggestions to tone down the expression. For example, if someone posts "This product is completely useless and pointless!", the emotion engine will identify this as "anger" and suggest a more neutral and constructive revision such as "This product may not have worked for me, but it may be useful for others."

[0771] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0772] Step 1:

[0773] The device receives the content (text, images, videos) entered by the user and sends it to the server. The entered content may be a specific sentence for text, a JPEG file for images, or an MP4 file for videos. This data processing involves serializing the content so that it can be sent to the server in the appropriate format.

[0774] Step 2:

[0775] The server saves the received content and starts the generative AI model to begin analysis. The content data mentioned above is used as input. In the case of video, the server extracts video frames and saves them as images. In the case of images or text, the server directly begins analyzing them. This data processing involves extracting and saving video frames.

[0776] Step 3:

[0777] The server performs face recognition, object recognition, and text recognition for each video frame. Each frame of video data is used as input. The output is the location of recognized faces and objects, as well as text content. This data calculation involves running image analysis algorithms and calculating the features of faces and objects.

[0778] Step 4:

[0779] The server performs image analysis, including facial expression analysis and background checks. Image data and analysis algorithms are used as input. Analysis results (e.g., a determination of the risk of privacy violations or whether the background contains inappropriate content) are obtained as output. This data processing includes calculating facial characteristics using an expression analysis algorithm and evaluating the content using a background check algorithm.

[0780] Step 5:

[0781] The server performs text analysis and uses natural language processing (NLP) algorithms to perform sentiment analysis and discriminatory expression detection. The input is text data, and the output is a sentiment score and a list of detected discriminatory expressions. This data operation uses an NLP model to analyze the text and identify sentiment and discriminatory expressions.

[0782] Step 6:

[0783] The server uses an emotion engine to determine the user's emotions. The input is the user's input data (text, voice, etc.), and the output is the result of the user's emotional state (e.g., "anger," "anxiety," etc.). This data calculation is performed by analyzing the user's emotions using an emotion analysis model.

[0784] Step 7:

[0785] The server determines the possibility of a controversy based on the analysis results. The input is the various analysis data obtained in the previous steps (face recognition results, object recognition results, emotion scores, etc.), and the output is the evaluation result of the controversy risk. This data calculation integrates each acquired data and performs a risk assessment.

[0786] Step 8:

[0787] Based on the results of the assessment, the server generates revision suggestions that take the user's emotions into account. The input is the flame risk assessment and the user's emotional data, and the output is a list of specific revision suggestions. This data calculation adjusts the content according to the user's emotions.

[0788] Step 9:

[0789] The server sends the generated revision suggestions to the terminal and presents them to the user via a user interface. The input is the generated revision suggestion data, and the output is the revision suggestions displayed on the user interface. This data processing involves sending the suggestion content to the terminal in an appropriate format.

[0790] Step 10:

[0791] The user modifies the content based on the suggested modifications and then sends the modified content back to the server from the device. The input is the modifications made by the user, and the output is the modified content. This data processing is a serialization process to ensure that the modifications are sent appropriately to the server.

[0792] Step 11:

[0793] The server re-analyzes the modified content to check for any new issues. The input is the modified content data, and the output is the result of the final analysis. This data calculation applies the re-analysis algorithm to check for any new issues.

[0794] Step 12:

[0795] The server posts the content that has been finalized to the SNS platform. The input is the finalized content data, and the output is the URL of the content posted to the SNS platform. This data processing involves posting the content to the SNS platform in the appropriate format.

[0796] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0797] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0798] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0799] [Third embodiment]

[0800] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0801] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0802] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0803] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0804] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0805] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0806] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0807] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0808] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0809] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0810] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0811] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0812] The following is a description of the preferred embodiment of the invention based on the claims.

[0813] ---

[0814] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[0815] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0816] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, including facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0817] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[0818] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[0819] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[0820] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0821] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By striving to analyze content quickly and accurately and suggest corrections, it is expected to reduce the risk of online outrage.

[0822] ---

[0823] The above is a specific example of the "mode for carrying out the invention."

[0824] The processing flow will be explained below.

[0825] Step 1:

[0826] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[0827] Step 2:

[0828] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[0829] Step 3:

[0830] The server stores the received content and starts the generative AI model to begin analysis. The server separates the received data by content and prepares it for analysis.

[0831] Step 4:

[0832] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[0833] Step 5:

[0834] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for inflammatory content.

[0835] Step 6:

[0836] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[0837] Step 7:

[0838] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[0839] Step 8:

[0840] To generate the suggested corrections, the server lists specific improvements, such as "suggesting blurring to protect the privacy of a specific person in the image" or "suggesting changes to the text to tone down offensive language."

[0841] Step 9:

[0842] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[0843] Step 10:

[0844] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[0845] Step 11:

[0846] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[0847] Step 12:

[0848] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[0849] Step 13:

[0850] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[0851] The above is a specific program processing flow based on the claims.

[0852] Example 1

[0853] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0854] In today's social networking environment, there is an increasing risk that users may unintentionally post inappropriate content, sparking a social media firestorm. This issue can have serious consequences for users, as it can include violations of personal privacy and offensive language. Conventional methods require users to manually review and correct posts, which not only takes time and effort but also carries the risk of false positives. There is a need for a system that can solve these issues and enable users to post appropriate content safely and quickly.

[0855] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0856] In this invention, the server includes means for receiving content entered by a user, means for transmitting the content to the server, means for saving the content, means for analyzing the content by activating a generative AI model, means for determining the possibility of a controversy, means for generating revision suggestions, means for transmitting the revision suggestions to a terminal, means for presenting the revision suggestions to the user via a user interface, means for re-receiving and re-analyzing the content revised by the user, and means for posting the final confirmed content to an SNS platform. This allows users to automatically determine the risk of content causing a controversy and quickly receive revision suggestions, thereby realizing safe posting.

[0857] A "generative AI model" is an artificial intelligence model trained using machine learning algorithms to analyze and classify input content.

[0858] "Content" refers to all digital data posted on social media, including videos, images, and text.

[0859] A "server" is a part of a computer system that stores, processes, and distributes data over a network.

[0860] "Terminal" refers to a device that is directly operated by a user, such as a smartphone or a personal computer.

[0861] "Analysis" is the process of breaking down the information contained in content and evaluating it for a specific purpose.

[0862] "Potential for a social media outrage" refers to the risk that the content posted will be deemed inappropriate and cause negative reactions or uproar on social media.

[0863] "Revision suggestions" is a means of generating specific revision suggestions for the posted content based on the analysis results.

[0864] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[0865] "Re-analysis" is the process of re-analyzing content that has been modified by a user to check for new problems.

[0866] "Post Request" means a request to post Final Reviewed Content to a Social Media Platform.

[0867] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[0868] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0869] The server stores the received content and starts analyzing it by activating a generative AI model, which is an artificial intelligence model trained using machine learning algorithms to analyze and classify the input content.

[0870] For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial, object, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, facial expression analysis, and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0871] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[0872] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[0873] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[0874] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[0875] This system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By quickly and accurately analyzing content and suggesting corrections, it is expected to reduce the risk of online outrage.

[0876] An example of a prompt sentence is, "I want to post on social media about yesterday's drinking party. For example, it should include the following content: Photo: A photo of the drinking party (faces are clearly visible), Text: 'We drank too much last night and got rowdy!' Please check whether the content of this post poses a risk of causing a backlash and let me know any suggestions for corrections."

[0877] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0878] Step 1:

[0879] A user inputs content into a multi-function posting app.

[0880] Specifically, a user uploads a video, image, or text to the app to post on a social network. The input is the data (video, image, text) that is stored within the application. The output is the content that is transformed into a structured format within the application to be sent to the next step.

[0881] Step 2:

[0882] The terminal receives the content and transmits it to the server.

[0883] The device sends the video, image, or text received from the multi-function posting app to the server. Specifically, the device securely transmits data using encryption protocols such as SSL / TLS. The input is the content uploaded by the user, and the output is the encrypted data sent to the server.

[0884] Step 3:

[0885] The server stores the received content.

[0886] The server receives the content sent from the terminal and temporarily stores it in a database. Specifically, the server checks the integrity of the received data and creates an entry in the database. The input is the encrypted data received, and the output is the content data stored in the database.

[0887] Step 4:

[0888] The server launches a generative AI model to analyze the content.

[0889] The server retrieves the stored content and launches the generative AI model to begin analysis. Specifically, in the case of video, the server extracts each frame and performs facial, object, and text recognition. In the case of images, it performs facial expression analysis and background checks, and in the case of text, it uses natural language processing (NLP) algorithms to perform sentiment analysis and detect discriminatory language. The input is the content stored on the server, and the output is a list of inappropriate content as the analysis result.

[0890] Step 5:

[0891] The server generates correction suggestions based on the analysis results.

[0892] The server determines the risk of outrage based on the analysis results obtained by the generative AI model and automatically generates specific correction suggestions. For example, it suggests blurring video frames containing inappropriate content or suggesting text changes to tone down offensive language. The input is the analysis results, and the output is a document containing specific correction suggestions.

[0893] Step 6:

[0894] The server sends the revision suggestions to the device.

[0895] The server sends the generated revision suggestions to the user's device. Specifically, the server formats the data and prepares it for transmission to the user's device. The input is the generated revision suggestions, and the output is data in a format that can be displayed in a user interface.

[0896] Step 7:

[0897] The terminal presents correction suggestions to the user.

[0898] The terminal presents the received revision suggestions to the user via a user interface. Specifically, the terminal receives data and displays it on the screen in a format that is easy for the user to understand. The input is the revision suggestion data sent from the server, and the output is the revision suggestions displayed on the user interface.

[0899] Step 8:

[0900] The user modifies the content based on the suggestions.

[0901] The user then modifies the content according to the suggested modifications. For example, they may use an image editing tool to apply a pixelated effect, or a text editor to change the wording. The input is the modification suggestions, and the output is the modified content.

[0902] Step 9:

[0903] The terminal sends the modified content back to the server.

[0904] The device then sends the modified content back to the server. Specifically, the data is sent securely again using an encryption protocol such as SSL / TLS. The input is the modified content, and the output is the encrypted data sent to the server.

[0905] Step 10:

[0906] The server reparses the modified content.

[0907] The server then re-analyzes the revised content to check for any new problems. Specifically, it launches the generative AI model again and performs the same analysis procedure. The input is the revised content, and the output is a report of the results of the re-analysis.

[0908] Step 11:

[0909] The server completes the final check and sends a post request to the social networking site.

[0910] If the server finds no new issues after re-analysis, it sends a post request to the social media platform. Specifically, it sends post data to the social media platform via API. The input is the final verified content, and the output is a post request sent to the social media platform.

[0911] Step 12:

[0912] The server sends a notification to the user that submission is complete.

[0913] The server notifies the user that the post to the SNS was successful. Specifically, it sends a message to the user's device using a notification API. The input is the status of the post completion, and the output is a notification message sent to the user's device.

[0914] (Application example 1)

[0915] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0916] Advertising content posted on social media and video platforms carries the risk of causing outrage and damaging brand image. Expressions that unconsciously offend consumers and content that may violate privacy are particularly problematic. The purpose of this invention is to detect these risks in advance and prompt appropriate corrections before posting, thereby enabling safe advertising while protecting brand image.

[0917] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0918] In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the results of the analysis, means for generating revision suggestions based on the determination results, means for presenting the revision suggestions to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to a social media platform or a video platform, means for analyzing the possibility that the content may have a negative impact on a brand image or consumer sentiment, and means for suggesting changes to inappropriate expressions. This allows advertising content to be posted safely without damaging the brand image.

[0919] "Means for receiving content entered by a user" refers to devices or software for receiving digital content such as videos, images, and text that a user wishes to post.

[0920] The "means for analyzing the content" refers to devices or software for analyzing the details of received content. Specifically, this includes video analysis, image analysis, and text analysis.

[0921] A "provocation risk assessment tool" is any device or software that assesses the risk that content will provoke public outrage or negative reaction based on the analyzed content.

[0922] The "means for generating correction suggestions" refers to a device or software that generates suggestions on how and which parts of content should be corrected in order to avoid backlash or damage to the brand image.

[0923] The "means for presenting revision suggestions to the user" refers to a device or software for notifying the user of the generated revision suggestions and visually displaying them.

[0924] The "means for receiving the content modified by the user again and performing a final check" refers to a device or software that receives the content after the user has made modifications and checks whether there are any new problems with the content.

[0925] "Means for posting final verified content to a social media platform or video platform" refers to the equipment or software used to actually post content that has been verified as having no problems to a social media platform or video platform.

[0926] "Means for analyzing the possibility of adversely affecting brand image or consumer sentiment" refers to devices or software for assessing the possibility that content may damage a company's brand image or cause negative emotions among consumers.

[0927] A "means for suggesting changes to inappropriate language" is a device or software that detects inappropriate language contained in content and suggests changing it to appropriate language.

[0928] This invention relates to a system that analyzes advertising content before it is posted on social media or video platforms to ensure that the advertisement does not have a negative impact on brand image or consumer sentiment, and makes suggestions for corrections as necessary.

[0929] System Configuration

[0930] The server has the following means:

[0931] 1. A means of receiving user-entered content

[0932] 2. Means for analyzing said content

[0933] 3. A means for determining the possibility of a controversy based on the results of the above analysis

[0934] 4. A means for generating a correction suggestion based on the result of the determination.

[0935] 5. Means for presenting said correction suggestions to the user

[0936] 6. A means for users to receive revised content again and perform a final review

[0937] 7. Posting the finalized content to a social media platform or video platform

[0938] 8. Means for analyzing the possibility that said content may have a negative impact on brand image or consumer sentiment

[0939] 9. How to suggest changes to inappropriate language

[0940] System Operation

[0941] 1. A means of receiving user-entered content

[0942] Users can input content to be posted through a smartphone application, and the input content is sent to the server by the smartphone.

[0943] 2. Means of content analysis

[0944] The server analyzes the received content using a generative AI model. The analysis results are as follows:

[0945] Video analysis: Extraction of video frames and frame-by-frame image analysis (e.g., face recognition, object recognition)

[0946] Image analysis: facial expression analysis and background check of images

[0947] Text analysis: Sentiment analysis and discriminatory language detection using natural language processing (NLP)

[0948] 3. How to determine the possibility of a controversy

[0949] The server determines whether there is a risk of a controversy based on the analysis results. For example, content containing offensive language or excessively negative emotions is identified as a risk of a controversy.

[0950] 4. A means of generating revision suggestions

[0951] If there is a possibility of a controversy, the server will generate specific suggestions for correction, such as "the image should be blurred" or "the offensive language should be toned down."

[0952] 5. A way to present suggested modifications to the user

[0953] The generated correction suggestions are presented to the user through the user interface of the smartphone application, where the user can review the suggestions and make any necessary corrections.

[0954] 6. A means for users to receive revised content again and perform a final review

[0955] The corrected content is sent back to the server, which re-analyzes it to check for any new problems.

[0956] 7. A means of posting the finalized content to a social media or video platform

[0957] Once the content has undergone final verification, it is posted from the server to a social media platform or video platform.

[0958] 8. A means of analyzing potential negative impacts on brand image and consumer sentiment

[0959] Specifically, for advertising content, analytical algorithms are used to determine whether it may have a negative impact on brand image or consumer sentiment, for example by detecting expressions that evoke negative emotions.

[0960] 9. How to suggest changes to inappropriate language

[0961] Based on the analysis results, the system suggests to the user how to change inappropriate language to appropriate ones, for example, by suggesting specific changes to the text to tone down offensive language.

[0962] Specific examples

[0963] For example, consider a case where a company uses AdSafeGuard to check an advertisement for a new product before posting it on social media. If the post contains offensive language such as "It's far better than other products," the app will detect this through sentiment analysis and offer suggestions to "tone down the offensive language." If the text entered is "It's foolish to use other companies' products. This product is better than any other!", the system will detect this and offer correction suggestions to the user.

[0964] In this way, advertising content can be posted safely without damaging the brand image.

[0965] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0966] Step 1:

[0967] A means of receiving user-entered content

[0968] Users input the advertising content (video, image, text) they wish to post through a smartphone application. The device sends this input data to the server, which then temporarily stores the received content. The data based on the input is stored on the server in its original format.

[0969] Step 2:

[0970] A means for analyzing the content

[0971] The server analyzes the received content using a generative AI model. For video, video frames are extracted and facial and object recognition is performed on each frame. For images, facial expression analysis and background checks are performed. For text, natural language processing (NLP) is used to perform sentiment analysis and discriminatory language detection. The input data (video, image, text) is analyzed and various features (face position, type of expression, emotion score, etc.) are extracted.

[0972] Step 3:

[0973] How to determine the possibility of a fire

[0974] The server determines the possibility of a controversy based on the analysis results. For example, it evaluates the results of facial recognition within video frames and the emotion score of the text. Based on the analysis results, if there is a high level of negative emotion (aggressive expressions, etc.), it determines that there is a possibility of a controversy and sets a risk level.

[0975] Step 4:

[0976] A means of generating revision suggestions

[0977] The server generates specific revision suggestions based on the level of risk of outrage, such as "images should be blurred" or "offensive language should be toned down." The server then generates a concrete action plan to mitigate the risk.

[0978] Step 5:

[0979] A means of presenting suggested revisions to the user

[0980] The server presents the generated correction suggestions to the user through the device's user interface. The user checks the displayed suggestions and makes necessary changes by operating their smartphone. Based on the presented suggestions, the user can then select an action to take.

[0981] Step 6:

[0982] A means for users to receive revised content again and perform a final review

[0983] After the user makes corrections, they send the corrected content back to the server. The server again uses the generative AI model to analyze the corrected content. A final check is made to see if there are any new problems with the corrections. A final check is made based on the results of the re-analysis after the corrections.

[0984] Step 7:

[0985] A means of posting the finalized content to a social media or video platform

[0986] Once the server has completed its final check, if the content is deemed to be safe, it will send a posting request to the social media platform or video platform. The content will then be posted, ensuring that the final checked data is safe.

[0987] Step 8:

[0988] A means of analyzing potential negative impacts on brand image and consumer sentiment

[0989] The server evaluates the content's potential to negatively impact brand image and consumer sentiment. Using sentiment analysis and facial expression recognition, the server identifies effective advertising language by determining whether the content is positive or negative. The server then analyzes the data to maintain brand image.

[0990] Step 9:

[0991] A way to suggest changes to inappropriate language

[0992] Based on the analysis results, the server will suggest changes if inappropriate language is found. For example, it will suggest specific text to tone down offensive language or suggest areas to modify images. Users can make corrections based on the suggestions, enabling them to create safe content that does not damage brand value.

[0993] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0994] Below is a description of the "Mode for Carrying Out the Invention" of the specification based on the claims of the invention combining emotion engines.

[0995] ---

[0996] This invention combines an emotion engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[0997] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[0998] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, such as facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[0999] In addition, an emotion engine is activated to recognize the user's emotions from the user's input. The server uses the emotion engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[1000] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[1001] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[1002] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[1003] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions for corrections such as "you should blur the image" and "tone down the offensive language." The user makes corrections based on these suggestions, and the post is finally completed successfully.

[1004] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[1005] ---

[1006] The above is a specific embodiment for carrying out the invention in which an emotion engine is combined.

[1007] The processing flow will be explained below.

[1008] MODE FOR CARRYING OUT THE INVENTION

[1009] This invention relates to a system that analyzes content posted by users on social networking sites, determines the possibility of the content becoming a controversy, and then proposes appropriate revisions by combining it with an emotion engine that recognizes the user's emotions. The specific processing flow is explained below, broken down into steps.

[1010] Step 1:

[1011] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[1012] Step 2:

[1013] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[1014] Step 3:

[1015] The server stores the received content and starts the generative AI model and emotion engine to begin analysis. The server separates the received data into various types of content and prepares them for analysis.

[1016] Step 4:

[1017] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[1018] Step 5:

[1019] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for controversy.

[1020] Step 6:

[1021] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[1022] Step 7:

[1023] The server uses a generative AI model and an emotion engine to determine the user's emotions (e.g., anger, joy, sadness) by analyzing the text and voice data entered by the user.

[1024] Step 8:

[1025] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[1026] Step 9:

[1027] To generate the correction suggestions, the server lists specific improvement measures, such as "proposing pixelation to protect the privacy of a specific person in the image" or "proposing changes to the text to tone down offensive language." Furthermore, the server adjusts the correction suggestions according to the user's emotions. For example, if the user is feeling angry, the server will suggest changes to the expression to tone down the emotion.

[1028] Step 10:

[1029] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[1030] Step 11:

[1031] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[1032] Step 12:

[1033] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[1034] Step 13:

[1035] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[1036] Step 14:

[1037] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[1038] ---

[1039] The above is a specific processing flow for implementing the invention that combines an emotion engine. Specific operations are clearly indicated at each step, allowing users to post to SNS more safely.

[1040] Example 2

[1041] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1042] The challenge is to prevent the risk of posts on SNS becoming controversial and provide a safe environment for use. In particular, there is a need for a system that can reduce users' psychological stress and allow them to use SNS with greater peace of mind by providing appropriate revision suggestions while taking users' emotions into consideration.

[1043] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the result of the analysis, means for generating a revision suggestion based on the determination result, means for presenting the revision suggestion to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to the SNS platform, and an emotion recognition engine for recognizing the user's emotions and reflecting them in the revision suggestion. This prevents the risk of a user's posted content causing a controversy before it happens, enabling safe and secure use of SNS.

[1044] Key Word Definitions

[1045] "User" refers to a person or entity who intends to use this system to post to the SNS and who inputs content.

[1046] "Content" is a general term for digital information such as videos, images, and text that users wish to post.

[1047] "Analysis" is the process performed by the server on the content it receives, evaluating and verifying the content to determine the possibility of it causing an uproar.

[1048] "Potential for a social media outcry" refers to the risk that a post will provoke a large number of negative reactions on social media, potentially having a harmful effect on users and third parties.

[1049] "Modification Suggestions" are specific changes or improvements generated by the server based on the analysis of the content to reduce the likelihood of a controversy.

[1050] An "emotion recognition engine" is an algorithm or technology that recognizes a user's emotions from the content or input data they are about to post, and adjusts suggested edits based on that emotional information.

[1051] "Generative AI Model" means an artificial intelligence model used by the server in the analysis process, including machine learning algorithms for evaluating and analyzing content.

[1052] "User interface" refers to the application or screen that a user operates to input content, suggest corrections, and perform final confirmation.

[1053] The "final check" is a process in which the server reanalyzes the content that has been corrected based on the user's suggested corrections and verifies that there are no problems.

[1054] "SNS platform" is a general term for social networking services that users use to share content.

[1055] MODE FOR CARRYING OUT THE INVENTION

[1056] This invention combines an emotion recognition engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[1057] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[1058] The server stores the received content in a database and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame, specifically facial recognition, object recognition, and text recognition to check for inappropriate content. For images, the server analyzes the images, for example, performing facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1059] In addition, an emotion recognition engine is activated to recognize the user's emotions from the user's input. The server uses the emotion recognition engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[1060] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[1061] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[1062] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[1063] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion recognition engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[1064] Example prompt sentence:

[1065] Prompt: Please explain the SNS post analysis system combined with the emotion engine. Please include the specific steps a user takes when trying to post content to SNS, the analysis method, the role of the emotion engine, the process of generating and presenting correction suggestions, and the steps to reanalyze and complete the post.

[1066] In this way, by using this system, users can prevent inappropriate posts on SNS and use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[1067] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1068] Program processing flow

[1069] Below, the program processing of this system will be explained in detail step by step.

[1070] Step 1:

[1071] The user inputs the content (video, image, text) they wish to post to SNS into a multi-function posting app.

[1072] Input: Digital information such as video, images, or text entered by a user.

[1073] Output: The multi-functional posting app stores the content data.

[1074] Step 2:

[1075] The terminal receives the content input by the user and transmits it to the server.

[1076] Input: Content entered into the user's device.

[1077] Data processing: The terminal converts the content data into a transfer format.

[1078] Output: The content data sent to the server.

[1079] Step 3:

[1080] The server stores the received content in a database and launches a generative AI model to begin analysis.

[1081] Input: Content data sent from the device.

[1082] Data processing: The server stores the content data in a database.

[1083] Output: Saved content data and activation signals for the generative AI model.

[1084] Step 4:

[1085] The server extracts each frame of the video and performs facial recognition, object recognition, and text recognition (in the case of video analysis).

[1086] Input: Stored video data.

[1087] Data processing / calculation: Extracting video frames and performing face, object, and text recognition on each frame.

[1088] Output: A list of frames containing inappropriate content as a result of execution.

[1089] Step 5:

[1090] The server analyzes the image and performs facial expression analysis and background checks (in the case of image analysis).

[1091] Input: Stored image data.

[1092] Data processing / computation: Facial expression analysis, background checks.

[1093] Output: Analysis results on facial expressions and background.

[1094] Step 6:

[1095] The server uses natural language processing (NLP) algorithms on the text data to perform sentiment analysis and detect discriminatory expressions (in the case of text analysis).

[1096] Input: Saved text data.

[1097] Data processing / calculation: Sentiment analysis using NLP algorithms, detection of discriminatory expressions.

[1098] Output: Sentiment analysis results and discriminatory expression detection results.

[1099] Step 7:

[1100] The server operates an emotion recognition engine based on the user's input data to determine the user's emotion.

[1101] Input: Text or voice data entered by the user.

[1102] Data processing / calculation: Emotion determination using an emotion recognition engine.

[1103] Output: User's emotion judgment result.

[1104] Step 8:

[1105] Based on the analysis results and the user's emotional assessment, the server identifies content that may cause controversy and generates specific suggestions for correction.

[1106] Input: Analysis results, user emotion judgment results.

[1107] Data processing / calculation: Listing problems and generating suggested fixes.

[1108] Output: A list of suggested fixes.

[1109] Step 9:

[1110] The server transmits the generated revision suggestions to the terminal and presents them to the user via the terminal's user interface.

[1111] Input: A list of correction suggestions.

[1112] Data processing / calculation: Sends suggested correction data to the terminal.

[1113] Output: Suggested fixes displayed on the terminal.

[1114] Step 10:

[1115] The user then corrects the posted content as necessary based on the suggested corrections.

[1116] Input: The suggested corrections displayed on the terminal.

[1117] Data manipulation / calculation: Modifying posted content using image editing tools or text editors.

[1118] Output: The modified content data.

[1119] Step 11:

[1120] The terminal transmits the corrected content to the server again.

[1121] Input: The modified content data.

[1122] Data processing / calculation: The corrected data is sent to the server.

[1123] Output: The modified content data sent to the server.

[1124] Step 12:

[1125] The server re-parses the modified content for any additional issues and performs a final check.

[1126] Input: The modified content data.

[1127] Data processing / calculation: Perform re-analysis process and re-check problems.

[1128] Output: The final verification result.

[1129] Step 13:

[1130] Once the server completes the final verification, it sends a request to post the secure content to the social media platform and actually completes the post.

[1131] Input: Final verification result.

[1132] Data processing / calculation: Generating and sending posting requests to social media platforms.

[1133] Output: The actual social media post completed.

[1134] (Application example 2)

[1135] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1136] Flaming and inappropriate posts on social media and other online platforms have become serious problems for companies and individuals. In addition, posts that strongly reflect the poster's emotions can easily lead to misunderstandings, which can cause serious problems, especially for official corporate accounts. Furthermore, conventional systems generate revision suggestions without taking the user's emotions into account, often resulting in specific and appropriate revision suggestions that do not reflect the user's intentions. For this reason, there is a growing demand for systems that can recognize user emotions, predict potential flame wars in advance, and make appropriate revision suggestions.

[1137] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy, means for generating revision suggestions, means for presenting the content to the user, means for receiving the content revised by the user again and performing a final check, means for recognizing the user's emotions and generating revision suggestions according to the emotions, means for analyzing the user's emotions using an emotion engine, and means for posting the finally checked content to an SNS platform. This makes it possible to make appropriate revision suggestions taking the user's emotions into consideration, and as a result, it is possible to reduce the risk of controversy and trouble on SNS.

[1138] "User" means a person who uses the system to create or post content.

[1139] "Content" is a general term for digital data such as text, images, and videos that users wish to post on social media platforms.

[1140] A "means for analyzing content" is a system that includes processes and algorithms for analyzing received content and identifying problems and characteristics of the content.

[1141] The "means for determining the possibility of a controversy" is a system that evaluates and determines the risk of a post causing a controversy on social media or in online communities based on the analysis results.

[1142] The "means for generating correction suggestions" refers to a process or system for suggesting appropriate corrections to the user based on the determined risk of a social media outrage.

[1143] The "means for presenting revision suggestions to the user" refers to an interface or display means for displaying the generated revision suggestions on the user's device so that the user can confirm them.

[1144] The "means for final confirmation" is a system that allows the user to recheck the content that has been revised to ensure there are no problems before final posting.

[1145] The "means for recognizing emotions and generating correction suggestions according to those emotions" is a system that analyzes the user's emotions and generates specific correction suggestions that will alleviate the emotions based on the results.

[1146] An "emotion engine" refers to an algorithm or model that analyzes a user's emotional state from input text or voice and detects specific emotions.

[1147] "Means of posting to social media platforms" refers to the system for actually posting the final, verified, safe content to social media or online platforms.

[1148] This invention is a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes revision suggestions taking into account the user's emotions. This system includes a series of processes, from receiving content entered by a user, analyzing it, determining the possibility of a controversy, analyzing the user's emotions using an emotion engine, generating revision suggestions based on those emotions, to finally posting the confirmed content to the social networking site platform.

[1149] The system has a means to receive content from the user's device. This content consists of digital data such as text, images, and video. The device then sends this content to the server, which stores it and starts analyzing it by launching a generative AI model.

[1150] Specifically, the server performs the following processes: For video, the server first extracts video frames and performs face, object, and text recognition for each frame; for images, it performs image analysis, facial expression analysis, and background checks; and for text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1151] In addition, the server uses an emotion engine to determine the user's emotions from the user's input. This emotion engine utilizes emotion analysis tools such as Hugging Face transformers and TextBlob to analyze the user's emotions from text and voice data. For example, if the user has strong emotions such as anger or anxiety, the server generates correction suggestions based on those emotions. These correction suggestions may include changes to the expression that will soften the user's emotions or more objective ways of expressing them.

[1152] Once the analysis identifies a potential flaming incident, the server generates specific suggestions for correction. The suggestions are tailored based on the user's emotions. For example, if the user is determined to be "angry," the server will provide suggestions for modifying the expression to tone down that emotion. The suggestions are sent to the user's device and presented via a user interface.

[1153] The user then corrects the content based on the suggested corrections and sends it back to the server via their device. The server then analyzes the corrected content again to check for any new issues. After the final check is complete, the server sends a request to the social media platform to post the safe content, and the post is then completed.

[1154] As a specific example, consider the case where a user tries to post "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends them to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[1155] An example of a prompt for a generative AI model is the following text:

[1156] This mobile app analyzes the text of social media posts in real time and generates revision suggestions based on the user's emotional state. If the post is offensive, it will make suggestions to tone down the expression. For example, if someone posts "This product is completely useless and pointless!", the emotion engine will identify this as "anger" and suggest a more neutral and constructive revision such as "This product may not have worked for me, but it may be useful for others."

[1157] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1158] Step 1:

[1159] The device receives the content (text, images, videos) entered by the user and sends it to the server. The entered content may be a specific sentence for text, a JPEG file for images, or an MP4 file for videos. This data processing involves serializing the content so that it can be sent to the server in the appropriate format.

[1160] Step 2:

[1161] The server saves the received content and starts the generative AI model to begin analysis. The content data mentioned above is used as input. In the case of video, the server extracts video frames and saves them as images. In the case of images or text, the server directly begins analyzing them. This data processing involves extracting and saving video frames.

[1162] Step 3:

[1163] The server performs face recognition, object recognition, and text recognition for each video frame. Each frame of video data is used as input. The output is the location of recognized faces and objects, as well as text content. This data calculation involves running image analysis algorithms and calculating the features of faces and objects.

[1164] Step 4:

[1165] The server performs image analysis, including facial expression analysis and background checks. Image data and analysis algorithms are used as input. Analysis results (e.g., a determination of the risk of privacy violations or whether the background contains inappropriate content) are obtained as output. This data processing includes calculating facial characteristics using an expression analysis algorithm and evaluating the content using a background check algorithm.

[1166] Step 5:

[1167] The server performs text analysis and uses natural language processing (NLP) algorithms to perform sentiment analysis and discriminatory expression detection. The input is text data, and the output is a sentiment score and a list of detected discriminatory expressions. This data operation uses an NLP model to analyze the text and identify sentiment and discriminatory expressions.

[1168] Step 6:

[1169] The server uses an emotion engine to determine the user's emotions. The input is the user's input data (text, voice, etc.), and the output is the result of the user's emotional state (e.g., "anger," "anxiety," etc.). This data calculation is performed by analyzing the user's emotions using an emotion analysis model.

[1170] Step 7:

[1171] The server determines the possibility of a controversy based on the analysis results. The input is the various analysis data obtained in the previous steps (face recognition results, object recognition results, emotion scores, etc.), and the output is the evaluation result of the controversy risk. This data calculation integrates each acquired data and performs a risk assessment.

[1172] Step 8:

[1173] Based on the results of the assessment, the server generates revision suggestions that take the user's emotions into account. The input is the flame risk assessment and the user's emotional data, and the output is a list of specific revision suggestions. This data calculation adjusts the content according to the user's emotions.

[1174] Step 9:

[1175] The server sends the generated revision suggestions to the terminal and presents them to the user via a user interface. The input is the generated revision suggestion data, and the output is the revision suggestions displayed on the user interface. This data processing involves sending the suggestion content to the terminal in an appropriate format.

[1176] Step 10:

[1177] The user modifies the content based on the suggested modifications and then sends the modified content back to the server from the device. The input is the modifications made by the user, and the output is the modified content. This data processing is a serialization process to ensure that the modifications are sent appropriately to the server.

[1178] Step 11:

[1179] The server re-analyzes the modified content to check for any new issues. The input is the modified content data, and the output is the result of the final analysis. This data calculation applies the re-analysis algorithm to check for any new issues.

[1180] Step 12:

[1181] The server posts the content that has been finalized to the SNS platform. The input is the finalized content data, and the output is the URL of the content posted to the SNS platform. This data processing involves posting the content to the SNS platform in the appropriate format.

[1182] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1183] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1184] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1185] [Fourth embodiment]

[1186] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1187] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1188] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1189] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1190] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1191] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1192] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1193] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1194] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1195] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1196] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1197] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1198] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1199] The following is a description of the preferred embodiment of the invention based on the claims.

[1200] ---

[1201] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[1202] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[1203] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, including facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1204] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[1205] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[1206] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[1207] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[1208] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By striving to analyze content quickly and accurately and suggest corrections, it is expected to reduce the risk of online outrage.

[1209] ---

[1210] The above is a specific example of the "mode for carrying out the invention."

[1211] The processing flow will be explained below.

[1212] Step 1:

[1213] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[1214] Step 2:

[1215] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[1216] Step 3:

[1217] The server stores the received content and starts the generative AI model to begin analysis. The server separates the received data by content and prepares it for analysis.

[1218] Step 4:

[1219] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[1220] Step 5:

[1221] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for inflammatory content.

[1222] Step 6:

[1223] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[1224] Step 7:

[1225] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[1226] Step 8:

[1227] To generate the suggested corrections, the server lists specific improvements, such as "suggesting blurring to protect the privacy of a specific person in the image" or "suggesting changes to the text to tone down offensive language."

[1228] Step 9:

[1229] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[1230] Step 10:

[1231] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[1232] Step 11:

[1233] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[1234] Step 12:

[1235] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[1236] Step 13:

[1237] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[1238] The above is a specific program processing flow based on the claims.

[1239] Example 1

[1240] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1241] In today's social networking environment, there is an increasing risk that users may unintentionally post inappropriate content, sparking a social media firestorm. This issue can have serious consequences for users, as it can include violations of personal privacy and offensive language. Conventional methods require users to manually review and correct posts, which not only takes time and effort but also carries the risk of false positives. There is a need for a system that can solve these issues and enable users to post appropriate content safely and quickly.

[1242] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1243] In this invention, the server includes means for receiving content entered by a user, means for transmitting the content to the server, means for saving the content, means for analyzing the content by activating a generative AI model, means for determining the possibility of a controversy, means for generating revision suggestions, means for transmitting the revision suggestions to a terminal, means for presenting the revision suggestions to the user via a user interface, means for re-receiving and re-analyzing the content revised by the user, and means for posting the final confirmed content to an SNS platform. This allows users to automatically determine the risk of content causing a controversy and quickly receive revision suggestions, thereby realizing safe posting.

[1244] A "generative AI model" is an artificial intelligence model trained using machine learning algorithms to analyze and classify input content.

[1245] "Content" refers to all digital data posted on social media, including videos, images, and text.

[1246] A "server" is a part of a computer system that stores, processes, and distributes data over a network.

[1247] "Terminal" refers to a device that is directly operated by a user, such as a smartphone or a personal computer.

[1248] "Analysis" is the process of breaking down the information contained in content and evaluating it for a specific purpose.

[1249] "Potential for a social media outrage" refers to the risk that the content posted will be deemed inappropriate and cause negative reactions or uproar on social media.

[1250] "Revision suggestions" is a means of generating specific revision suggestions for the posted content based on the analysis results.

[1251] "User interface" refers to the screens and input devices that allow a user to interact with a system.

[1252] "Re-analysis" is the process of re-analyzing content that has been modified by a user to check for new problems.

[1253] "Post Request" means a request to post Final Reviewed Content to a Social Media Platform.

[1254] The present invention relates to a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and proposes corrections. A specific embodiment of this system is described below.

[1255] First, the user inputs the content (video, image, text) they want to post to the SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[1256] The server stores the received content and starts analyzing it by activating a generative AI model, which is an artificial intelligence model trained using machine learning algorithms to analyze and classify the input content.

[1257] For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial, object, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, facial expression analysis, and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1258] If the analysis identifies potentially controversial content, the server lists the issues and generates specific suggestions for correction, such as blurring an image to protect the privacy of a specific person in the image, or suggesting changes to the text to tone down offensive language.

[1259] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user can then review the suggested corrections and make any necessary corrections to the post, for example, by using an image editing tool to blur the image or by changing the wording in a text editor.

[1260] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a posting request to the social media platform and complete the actual posting.

[1261] As a concrete example, consider the case where a user posts a message titled "Yesterday's drinking party." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Based on this, the server presents the user with suggestions for revisions, such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[1262] This system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. By quickly and accurately analyzing content and suggesting corrections, it is expected to reduce the risk of online outrage.

[1263] An example of a prompt sentence is, "I want to post on social media about yesterday's drinking party. For example, it should include the following content: Photo: A photo of the drinking party (faces are clearly visible), Text: 'We drank too much last night and got rowdy!' Please check whether the content of this post poses a risk of causing a backlash and let me know any suggestions for corrections."

[1264] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1265] Step 1:

[1266] A user inputs content into a multi-function posting app.

[1267] Specifically, a user uploads a video, image, or text to the app to post on a social network. The input is the data (video, image, text) that is stored within the application. The output is the content that is transformed into a structured format within the application to be sent to the next step.

[1268] Step 2:

[1269] The terminal receives the content and transmits it to the server.

[1270] The device sends the video, image, or text received from the multi-function posting app to the server. Specifically, the device securely transmits data using encryption protocols such as SSL / TLS. The input is the content uploaded by the user, and the output is the encrypted data sent to the server.

[1271] Step 3:

[1272] The server stores the received content.

[1273] The server receives the content sent from the terminal and temporarily stores it in a database. Specifically, the server checks the integrity of the received data and creates an entry in the database. The input is the encrypted data received, and the output is the content data stored in the database.

[1274] Step 4:

[1275] The server launches a generative AI model to analyze the content.

[1276] The server retrieves the stored content and launches the generative AI model to begin analysis. Specifically, in the case of video, the server extracts each frame and performs facial, object, and text recognition. In the case of images, it performs facial expression analysis and background checks, and in the case of text, it uses natural language processing (NLP) algorithms to perform sentiment analysis and detect discriminatory language. The input is the content stored on the server, and the output is a list of inappropriate content as the analysis result.

[1277] Step 5:

[1278] The server generates correction suggestions based on the analysis results.

[1279] The server determines the risk of outrage based on the analysis results obtained by the generative AI model and automatically generates specific correction suggestions. For example, it suggests blurring video frames containing inappropriate content or suggesting text changes to tone down offensive language. The input is the analysis results, and the output is a document containing specific correction suggestions.

[1280] Step 6:

[1281] The server sends the revision suggestions to the device.

[1282] The server sends the generated revision suggestions to the user's device. Specifically, the server formats the data and prepares it for transmission to the user's device. The input is the generated revision suggestions, and the output is data in a format that can be displayed in a user interface.

[1283] Step 7:

[1284] The terminal presents correction suggestions to the user.

[1285] The terminal presents the received revision suggestions to the user via a user interface. Specifically, the terminal receives data and displays it on the screen in a format that is easy for the user to understand. The input is the revision suggestion data sent from the server, and the output is the revision suggestions displayed on the user interface.

[1286] Step 8:

[1287] The user modifies the content based on the suggestions.

[1288] The user then modifies the content according to the suggested modifications. For example, they may use an image editing tool to apply a pixelated effect, or a text editor to change the wording. The input is the modification suggestions, and the output is the modified content.

[1289] Step 9:

[1290] The terminal sends the modified content back to the server.

[1291] The device then sends the modified content back to the server. Specifically, the data is sent securely again using an encryption protocol such as SSL / TLS. The input is the modified content, and the output is the encrypted data sent to the server.

[1292] Step 10:

[1293] The server reparses the modified content.

[1294] The server then re-analyzes the revised content to check for any new problems. Specifically, it launches the generative AI model again and performs the same analysis procedure. The input is the revised content, and the output is a report of the results of the re-analysis.

[1295] Step 11:

[1296] The server completes the final check and sends a post request to the social networking site.

[1297] If the server finds no new issues after re-analysis, it sends a post request to the social media platform. Specifically, it sends post data to the social media platform via API. The input is the final verified content, and the output is a post request sent to the social media platform.

[1298] Step 12:

[1299] The server sends a notification to the user that submission is complete.

[1300] The server notifies the user that the post to the SNS was successful. Specifically, it sends a message to the user's device using a notification API. The input is the status of the post completion, and the output is a notification message sent to the user's device.

[1301] (Application example 1)

[1302] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1303] Advertising content posted on social media and video platforms carries the risk of causing outrage and damaging brand image. Expressions that unconsciously offend consumers and content that may violate privacy are particularly problematic. The purpose of this invention is to detect these risks in advance and prompt appropriate corrections before posting, thereby enabling safe advertising while protecting brand image.

[1304] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1305] In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the results of the analysis, means for generating revision suggestions based on the determination results, means for presenting the revision suggestions to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to a social media platform or a video platform, means for analyzing the possibility that the content may have a negative impact on a brand image or consumer sentiment, and means for suggesting changes to inappropriate expressions. This allows advertising content to be posted safely without damaging the brand image.

[1306] "Means for receiving content entered by a user" refers to devices or software for receiving digital content such as videos, images, and text that a user wishes to post.

[1307] The "means for analyzing the content" refers to devices or software for analyzing the details of received content. Specifically, this includes video analysis, image analysis, and text analysis.

[1308] A "provocation risk assessment tool" is any device or software that assesses the risk that content will provoke public outrage or negative reaction based on the analyzed content.

[1309] The "means for generating correction suggestions" refers to a device or software that generates suggestions on how and which parts of content should be corrected in order to avoid backlash or damage to the brand image.

[1310] The "means for presenting revision suggestions to the user" refers to a device or software for notifying the user of the generated revision suggestions and visually displaying them.

[1311] The "means for receiving the content modified by the user again and performing a final check" refers to a device or software that receives the content after the user has made modifications and checks whether there are any new problems with the content.

[1312] "Means for posting final verified content to a social media platform or video platform" refers to the equipment or software used to actually post content that has been verified as having no problems to a social media platform or video platform.

[1313] "Means for analyzing the possibility of adversely affecting brand image or consumer sentiment" refers to devices or software for assessing the possibility that content may damage a company's brand image or cause negative emotions among consumers.

[1314] A "means for suggesting changes to inappropriate language" is a device or software that detects inappropriate language contained in content and suggests changing it to appropriate language.

[1315] This invention relates to a system that analyzes advertising content before it is posted on social media or video platforms to ensure that the advertisement does not have a negative impact on brand image or consumer sentiment, and makes suggestions for corrections as necessary.

[1316] System Configuration

[1317] The server has the following means:

[1318] 1. A means of receiving user-entered content

[1319] 2. Means for analyzing said content

[1320] 3. A means for determining the possibility of a controversy based on the results of the above analysis

[1321] 4. A means for generating a correction suggestion based on the result of the determination.

[1322] 5. Means for presenting said correction suggestions to the user

[1323] 6. A means for users to receive revised content again and perform a final review

[1324] 7. Posting the finalized content to a social media platform or video platform

[1325] 8. Means for analyzing the possibility that said content may have a negative impact on brand image or consumer sentiment

[1326] 9. How to suggest changes to inappropriate language

[1327] System Operation

[1328] 1. A means of receiving user-entered content

[1329] Users can input content to be posted through a smartphone application, and the input content is sent to the server by the smartphone.

[1330] 2. Means of content analysis

[1331] The server analyzes the received content using a generative AI model. The analysis results are as follows:

[1332] Video analysis: Extraction of video frames and frame-by-frame image analysis (e.g., face recognition, object recognition)

[1333] Image analysis: facial expression analysis and background check of images

[1334] Text analysis: Sentiment analysis and discriminatory language detection using natural language processing (NLP)

[1335] 3. How to determine the possibility of a controversy

[1336] The server determines whether there is a risk of a controversy based on the analysis results. For example, content containing offensive language or excessively negative emotions is identified as a risk of a controversy.

[1337] 4. A means of generating revision suggestions

[1338] If there is a possibility of a controversy, the server will generate specific suggestions for correction, such as "the image should be blurred" or "the offensive language should be toned down."

[1339] 5. A way to present suggested modifications to the user

[1340] The generated correction suggestions are presented to the user through the user interface of the smartphone application, where the user can review the suggestions and make any necessary corrections.

[1341] 6. A means for users to receive revised content again and perform a final review

[1342] The corrected content is sent back to the server, which re-analyzes it to check for any new problems.

[1343] 7. A means of posting the finalized content to a social media or video platform

[1344] Once the content has undergone final verification, it is posted from the server to a social media platform or video platform.

[1345] 8. A means of analyzing potential negative impacts on brand image and consumer sentiment

[1346] Specifically, for advertising content, analytical algorithms are used to determine whether it may have a negative impact on brand image or consumer sentiment, for example by detecting expressions that evoke negative emotions.

[1347] 9. How to suggest changes to inappropriate language

[1348] Based on the analysis results, the system suggests to the user how to change inappropriate language to appropriate ones, for example, by suggesting specific changes to the text to tone down offensive language.

[1349] Specific examples

[1350] For example, consider a case where a company uses AdSafeGuard to check an advertisement for a new product before posting it on social media. If the post contains offensive language such as "It's far better than other products," the app will detect this through sentiment analysis and offer suggestions to "tone down the offensive language." If the text entered is "It's foolish to use other companies' products. This product is better than any other!", the system will detect this and offer correction suggestions to the user.

[1351] In this way, advertising content can be posted safely without damaging the brand image.

[1352] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1353] Step 1:

[1354] A means of receiving user-entered content

[1355] Users input the advertising content (video, image, text) they wish to post through a smartphone application. The device sends this input data to the server, which then temporarily stores the received content. The data based on the input is stored on the server in its original format.

[1356] Step 2:

[1357] A means for analyzing the content

[1358] The server analyzes the received content using a generative AI model. For video, video frames are extracted and facial and object recognition is performed on each frame. For images, facial expression analysis and background checks are performed. For text, natural language processing (NLP) is used to perform sentiment analysis and discriminatory language detection. The input data (video, image, text) is analyzed and various features (face position, type of expression, emotion score, etc.) are extracted.

[1359] Step 3:

[1360] How to determine the possibility of a fire

[1361] The server determines the possibility of a controversy based on the analysis results. For example, it evaluates the results of facial recognition within video frames and the emotion score of the text. Based on the analysis results, if there is a high level of negative emotion (aggressive expressions, etc.), it determines that there is a possibility of a controversy and sets a risk level.

[1362] Step 4:

[1363] A means of generating revision suggestions

[1364] The server generates specific revision suggestions based on the level of risk of outrage, such as "images should be blurred" or "offensive language should be toned down." The server then generates a concrete action plan to mitigate the risk.

[1365] Step 5:

[1366] A means of presenting suggested revisions to the user

[1367] The server presents the generated correction suggestions to the user through the device's user interface. The user checks the displayed suggestions and makes necessary changes by operating their smartphone. Based on the presented suggestions, the user can then select an action to take.

[1368] Step 6:

[1369] A means for users to receive revised content again and perform a final review

[1370] After the user makes corrections, they send the corrected content back to the server. The server again uses the generative AI model to analyze the corrected content. A final check is made to see if there are any new problems with the corrections. A final check is made based on the results of the re-analysis after the corrections.

[1371] Step 7:

[1372] A means of posting the finalized content to a social media or video platform

[1373] Once the server has completed its final check, if the content is deemed to be safe, it will send a posting request to the social media platform or video platform. The content will then be posted, ensuring that the final checked data is safe.

[1374] Step 8:

[1375] A means of analyzing potential negative impacts on brand image and consumer sentiment

[1376] The server evaluates the content's potential to negatively impact brand image and consumer sentiment. Using sentiment analysis and facial expression recognition, the server identifies effective advertising language by determining whether the content is positive or negative. The server then analyzes the data to maintain brand image.

[1377] Step 9:

[1378] A way to suggest changes to inappropriate language

[1379] Based on the analysis results, the server will suggest changes if inappropriate language is found. For example, it will suggest specific text to tone down offensive language or suggest areas to modify images. Users can make corrections based on the suggestions, enabling them to create safe content that does not damage brand value.

[1380] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1381] Below is a description of the "Mode for Carrying Out the Invention" of the specification based on the claims of the invention combining emotion engines.

[1382] ---

[1383] This invention combines an emotion engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[1384] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[1385] The server stores the received content and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame. Specifically, it performs facial recognition, object recognition, and text recognition on each frame to check for inappropriate content. For images, the server performs image analysis, such as facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1386] In addition, an emotion engine is activated to recognize the user's emotions from the user's input. The server uses the emotion engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[1387] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[1388] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[1389] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[1390] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions for corrections such as "you should blur the image" and "tone down the offensive language." The user makes corrections based on these suggestions, and the post is finally completed successfully.

[1391] In this way, this system will prevent users from posting inappropriate content on social media, allowing them to use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[1392] ---

[1393] The above is a specific embodiment for carrying out the invention in which an emotion engine is combined.

[1394] The processing flow will be explained below.

[1395] MODE FOR CARRYING OUT THE INVENTION

[1396] This invention relates to a system that analyzes content posted by users on social networking sites, determines the possibility of the content becoming a controversy, and then proposes appropriate revisions by combining it with an emotion engine that recognizes the user's emotions. The specific processing flow is explained below, broken down into steps.

[1397] Step 1:

[1398] Users input the content (videos, images, text) they wish to post to SNS into a multi-function posting app. Users enter text into the app on their smartphone or PC and upload video and image files.

[1399] Step 2:

[1400] The device sends the input content to the server via an HTTP POST request. Specifically, the device sends a JSON-formatted request containing text, image, and video data to the server's analysis endpoint.

[1401] Step 3:

[1402] The server stores the received content and starts the generative AI model and emotion engine to begin analysis. The server separates the received data into various types of content and prepares them for analysis.

[1403] Step 4:

[1404] In the case of video, the server extracts video frames and performs image analysis on each frame, including facial recognition, object recognition, and text recognition, to check for inappropriate content.

[1405] Step 5:

[1406] For images, the server performs image analysis, such as facial expression analysis and background checks, to detect specific elements or text within the image and assess its potential for controversy.

[1407] Step 6:

[1408] For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment and detect discriminatory language in the input text, specifically by performing word-level analysis and contextual analysis to identify offensive or problematic content.

[1409] Step 7:

[1410] The server uses a generative AI model and an emotion engine to determine the user's emotions (e.g., anger, joy, sadness) by analyzing the text and voice data entered by the user.

[1411] Step 8:

[1412] The server identifies content that may cause controversy based on the analysis results provided by the generative AI model, creates a list of problems from the analysis results, and generates correction suggestions based on the content.

[1413] Step 9:

[1414] To generate the correction suggestions, the server lists specific improvement measures, such as "proposing pixelation to protect the privacy of a specific person in the image" or "proposing changes to the text to tone down offensive language." Furthermore, the server adjusts the correction suggestions according to the user's emotions. For example, if the user is feeling angry, the server will suggest changes to the expression to tone down the emotion.

[1415] Step 10:

[1416] The server converts the generated revision suggestions into JSON format and sends them to the terminal as an HTTP response. The terminal displays the revision suggestions received from the server on a user interface (UI).

[1417] Step 11:

[1418] The device notifies the user of the suggested revisions, and the user can review the revisions. For example, the user can use an image editing tool to blur the image or change the wording in a text editor.

[1419] Step 12:

[1420] The user checks the revised content and finally decides what to post. The revised content that the user has checked is then sent back to the server from the device.

[1421] Step 13:

[1422] The server then re-parses the modified content to check for any new issues, and then makes a request to the social media platform to post the safe content.

[1423] Step 14:

[1424] Once the posting to the SNS platform is complete, the device will display a posting completion notification to the user, who will then be able to confirm that the corrected and safe content has been posted.

[1425] ---

[1426] The above is a specific processing flow for implementing the invention that combines an emotion engine. Specific operations are clearly indicated at each step, allowing users to post to SNS more safely.

[1427] Example 2

[1428] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1429] The challenge is to prevent the risk of posts on SNS becoming controversial and provide a safe environment for use. In particular, there is a need for a system that can reduce users' psychological stress and allow them to use SNS with greater peace of mind by providing appropriate revision suggestions while taking users' emotions into consideration.

[1430] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy based on the result of the analysis, means for generating a revision suggestion based on the determination result, means for presenting the revision suggestion to the user, means for receiving the content revised by the user again and performing a final check, means for posting the finally checked content to the SNS platform, and an emotion recognition engine for recognizing the user's emotions and reflecting them in the revision suggestion. This prevents the risk of a user's posted content causing a controversy before it happens, enabling safe and secure use of SNS.

[1431] Key Word Definitions

[1432] "User" refers to a person or entity who intends to use this system to post to the SNS and who inputs content.

[1433] "Content" is a general term for digital information such as videos, images, and text that users wish to post.

[1434] "Analysis" is the process performed by the server on the content it receives, evaluating and verifying the content to determine the possibility of it causing an uproar.

[1435] "Potential for a social media outcry" refers to the risk that a post will provoke a large number of negative reactions on social media, potentially having a harmful effect on users and third parties.

[1436] "Modification Suggestions" are specific changes or improvements generated by the server based on the analysis of the content to reduce the likelihood of a controversy.

[1437] An "emotion recognition engine" is an algorithm or technology that recognizes a user's emotions from the content or input data they are about to post, and adjusts suggested edits based on that emotional information.

[1438] "Generative AI Model" means an artificial intelligence model used by the server in the analysis process, including machine learning algorithms for evaluating and analyzing content.

[1439] "User interface" refers to the application or screen that a user operates to input content, suggest corrections, and perform final confirmation.

[1440] The "final check" is a process in which the server reanalyzes the content that has been corrected based on the user's suggested corrections and verifies that there are no problems.

[1441] "SNS platform" is a general term for social networking services that users use to share content.

[1442] MODE FOR CARRYING OUT THE INVENTION

[1443] This invention combines an emotion recognition engine with a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes suggestions for corrections, thereby recognizing the user's emotions and providing appropriate suggestions for corrections. A specific embodiment of this system is described below.

[1444] Users input the content (videos, images, text) they want to post to SNS into the multi-function posting app. The content entered by the user is received by the device and sent to the server.

[1445] The server stores the received content in a database and launches a generative AI model to begin analysis. For video, the server extracts video frames and performs image analysis on each frame, specifically facial recognition, object recognition, and text recognition to check for inappropriate content. For images, the server analyzes the images, for example, performing facial expression analysis and background checks. For text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1446] In addition, an emotion recognition engine is activated to recognize the user's emotions from the user's input. The server uses the emotion recognition engine to determine the user's emotions from the text and voice data entered by the user. For example, if the user has strong emotions such as anger or anxiety, the server will suggest corrections according to those emotions.

[1447] If the analysis identifies content that could potentially spark controversy, the server lists the issues and generates specific suggestions for revising the content. The suggestions are tailored based on the user's emotions. For example, if the user is angry, the server suggests changing the wording to soften the emotion.

[1448] The generated correction suggestions are sent from the server to the device and presented to the user via the device's user interface. The user then checks the suggested corrections and makes any necessary corrections to the post, such as by using an image editing tool to apply pixelation or by changing the wording in a text editor.

[1449] Once the user has confirmed the revised content, the device will send the revised content back to the server. The server will then re-analyze the content to check for any new issues. Once the final confirmation is complete, the server will make a request to the social media platform to post the safe content, and the post will be completed.

[1450] As a concrete example, consider the case where a user posts a message titled "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends the content to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion recognition engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[1451] Example prompt sentence:

[1452] Prompt: Please explain the SNS post analysis system combined with the emotion engine. Please include the specific steps a user takes when trying to post content to SNS, the analysis method, the role of the emotion engine, the process of generating and presenting correction suggestions, and the steps to reanalyze and complete the post.

[1453] In this way, by using this system, users can prevent inappropriate posts on SNS and use the site safely. In particular, by providing appropriate suggestions for corrections that take into account the user's feelings, it is expected to further reduce the risk of online outrage.

[1454] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1455] Program processing flow

[1456] Below, the program processing of this system will be explained in detail step by step.

[1457] Step 1:

[1458] The user inputs the content (video, image, text) they wish to post to SNS into a multi-function posting app.

[1459] Input: Digital information such as video, images, or text entered by a user.

[1460] Output: The multi-functional posting app stores the content data.

[1461] Step 2:

[1462] The terminal receives the content input by the user and transmits it to the server.

[1463] Input: Content entered into the user's device.

[1464] Data processing: The terminal converts the content data into a transfer format.

[1465] Output: The content data sent to the server.

[1466] Step 3:

[1467] The server stores the received content in a database and launches a generative AI model to begin analysis.

[1468] Input: Content data sent from the device.

[1469] Data processing: The server stores the content data in a database.

[1470] Output: Saved content data and activation signals for the generative AI model.

[1471] Step 4:

[1472] The server extracts each frame of the video and performs facial recognition, object recognition, and text recognition (in the case of video analysis).

[1473] Input: Stored video data.

[1474] Data processing / calculation: Extracting video frames and performing face, object, and text recognition on each frame.

[1475] Output: A list of frames containing inappropriate content as a result of execution.

[1476] Step 5:

[1477] The server analyzes the image and performs facial expression analysis and background checks (in the case of image analysis).

[1478] Input: Stored image data.

[1479] Data processing / computation: Facial expression analysis, background checks.

[1480] Output: Analysis results on facial expressions and background.

[1481] Step 6:

[1482] The server uses natural language processing (NLP) algorithms on the text data to perform sentiment analysis and detect discriminatory expressions (in the case of text analysis).

[1483] Input: Saved text data.

[1484] Data processing / calculation: Sentiment analysis using NLP algorithms, detection of discriminatory expressions.

[1485] Output: Sentiment analysis results and discriminatory expression detection results.

[1486] Step 7:

[1487] The server operates an emotion recognition engine based on the user's input data to determine the user's emotion.

[1488] Input: Text or voice data entered by the user.

[1489] Data processing / calculation: Emotion determination using an emotion recognition engine.

[1490] Output: User's emotion judgment result.

[1491] Step 8:

[1492] Based on the analysis results and the user's emotional assessment, the server identifies content that may cause controversy and generates specific suggestions for correction.

[1493] Input: Analysis results, user emotion judgment results.

[1494] Data processing / calculation: Listing problems and generating suggested fixes.

[1495] Output: A list of suggested fixes.

[1496] Step 9:

[1497] The server transmits the generated revision suggestions to the terminal and presents them to the user via the terminal's user interface.

[1498] Input: A list of correction suggestions.

[1499] Data processing / calculation: Sends suggested correction data to the terminal.

[1500] Output: Suggested fixes displayed on the terminal.

[1501] Step 10:

[1502] The user then corrects the posted content as necessary based on the suggested corrections.

[1503] Input: The suggested corrections displayed on the terminal.

[1504] Data manipulation / calculation: Modifying posted content using image editing tools or text editors.

[1505] Output: The modified content data.

[1506] Step 11:

[1507] The terminal transmits the corrected content to the server again.

[1508] Input: The modified content data.

[1509] Data processing / calculation: The corrected data is sent to the server.

[1510] Output: The modified content data sent to the server.

[1511] Step 12:

[1512] The server re-parses the modified content for any additional issues and performs a final check.

[1513] Input: The modified content data.

[1514] Data processing / calculation: Perform re-analysis process and re-check problems.

[1515] Output: The final verification result.

[1516] Step 13:

[1517] Once the server completes the final verification, it sends a request to post the secure content to the social media platform and actually completes the post.

[1518] Input: Final verification result.

[1519] Data processing / calculation: Generating and sending posting requests to social media platforms.

[1520] Output: The actual social media post completed.

[1521] (Application example 2)

[1522] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1523] Flaming and inappropriate posts on social media and other online platforms have become serious problems for companies and individuals. In addition, posts that strongly reflect the poster's emotions can easily lead to misunderstandings, which can cause serious problems, especially for official corporate accounts. Furthermore, conventional systems generate revision suggestions without taking the user's emotions into account, often resulting in specific and appropriate revision suggestions that do not reflect the user's intentions. For this reason, there is a growing demand for systems that can recognize user emotions, predict potential flame wars in advance, and make appropriate revision suggestions.

[1524] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving content entered by a user, means for analyzing the content, means for determining the possibility of a controversy, means for generating revision suggestions, means for presenting the content to the user, means for receiving the content revised by the user again and performing a final check, means for recognizing the user's emotions and generating revision suggestions according to the emotions, means for analyzing the user's emotions using an emotion engine, and means for posting the finally checked content to an SNS platform. This makes it possible to make appropriate revision suggestions taking the user's emotions into consideration, and as a result, it is possible to reduce the risk of controversy and trouble on SNS.

[1525] "User" means a person who uses the system to create or post content.

[1526] "Content" is a general term for digital data such as text, images, and videos that users wish to post on social media platforms.

[1527] A "means for analyzing content" is a system that includes processes and algorithms for analyzing received content and identifying problems and characteristics of the content.

[1528] The "means for determining the possibility of a controversy" is a system that evaluates and determines the risk of a post causing a controversy on social media or in online communities based on the analysis results.

[1529] The "means for generating correction suggestions" refers to a process or system for suggesting appropriate corrections to the user based on the determined risk of a social media outrage.

[1530] The "means for presenting revision suggestions to the user" refers to an interface or display means for displaying the generated revision suggestions on the user's device so that the user can confirm them.

[1531] The "means for final confirmation" is a system that allows the user to recheck the content that has been revised to ensure there are no problems before final posting.

[1532] The "means for recognizing emotions and generating correction suggestions according to those emotions" is a system that analyzes the user's emotions and generates specific correction suggestions that will alleviate the emotions based on the results.

[1533] An "emotion engine" refers to an algorithm or model that analyzes a user's emotional state from input text or voice and detects specific emotions.

[1534] "Means of posting to social media platforms" refers to the system for actually posting the final, verified, safe content to social media or online platforms.

[1535] This invention is a system that analyzes the content of posts to social networking sites in advance, determines the possibility of a controversy, and makes revision suggestions taking into account the user's emotions. This system includes a series of processes, from receiving content entered by a user, analyzing it, determining the possibility of a controversy, analyzing the user's emotions using an emotion engine, generating revision suggestions based on those emotions, to finally posting the confirmed content to the social networking site platform.

[1536] The system has a means to receive content from the user's device. This content consists of digital data such as text, images, and video. The device then sends this content to the server, which stores it and starts analyzing it by launching a generative AI model.

[1537] Specifically, the server performs the following processes: For video, the server first extracts video frames and performs face, object, and text recognition for each frame; for images, it performs image analysis, facial expression analysis, and background checks; and for text, the server uses natural language processing (NLP) algorithms to analyze the sentiment of the input text and detect discriminatory language.

[1538] In addition, the server uses an emotion engine to determine the user's emotions from the user's input. This emotion engine utilizes emotion analysis tools such as Hugging Face transformers and TextBlob to analyze the user's emotions from text and voice data. For example, if the user has strong emotions such as anger or anxiety, the server generates correction suggestions based on those emotions. These correction suggestions may include changes to the expression that will soften the user's emotions or more objective ways of expressing them.

[1539] Once the analysis identifies a potential flaming incident, the server generates specific suggestions for correction. The suggestions are tailored based on the user's emotions. For example, if the user is determined to be "angry," the server will provide suggestions for modifying the expression to tone down that emotion. The suggestions are sent to the user's device and presented via a user interface.

[1540] The user then corrects the content based on the suggested corrections and sends it back to the server via their device. The server then analyzes the corrected content again to check for any new issues. After the final check is complete, the server sends a request to the social media platform to post the safe content, and the post is then completed.

[1541] As a specific example, consider the case where a user tries to post "How yesterday's drinking party went." The user inputs photos and text from the drinking party and sends them to the server via their device. The server identifies the possibility of violating the privacy of the people in the photo and determines that the text contains offensive language. Furthermore, if the emotion engine determines that the user's emotion is "anger," it makes suggestions such as "you should blur the image" and "tone down the offensive language." The user makes the corrections based on these suggestions, and the post is finally completed successfully.

[1542] An example of a prompt for a generative AI model is the following text:

[1543] This mobile app analyzes the text of social media posts in real time and generates revision suggestions based on the user's emotional state. If the post is offensive, it will make suggestions to tone down the expression. For example, if someone posts "This product is completely useless and pointless!", the emotion engine will identify this as "anger" and suggest a more neutral and constructive revision such as "This product may not have worked for me, but it may be useful for others."

[1544] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1545] Step 1:

[1546] The device receives the content (text, images, videos) entered by the user and sends it to the server. The entered content may be a specific sentence for text, a JPEG file for images, or an MP4 file for videos. This data processing involves serializing the content so that it can be sent to the server in the appropriate format.

[1547] Step 2:

[1548] The server saves the received content and starts the generative AI model to begin analysis. The content data mentioned above is used as input. In the case of video, the server extracts video frames and saves them as images. In the case of images or text, the server directly begins analyzing them. This data processing involves extracting and saving video frames.

[1549] Step 3:

[1550] The server performs face recognition, object recognition, and text recognition for each video frame. Each frame of video data is used as input. The output is the location of recognized faces and objects, as well as text content. This data calculation involves running image analysis algorithms and calculating the features of faces and objects.

[1551] Step 4:

[1552] The server performs image analysis, including facial expression analysis and background checks. Image data and analysis algorithms are used as input. Analysis results (e.g., a determination of the risk of privacy violations or whether the background contains inappropriate content) are obtained as output. This data processing includes calculating facial characteristics using an expression analysis algorithm and evaluating the content using a background check algorithm.

[1553] Step 5:

[1554] The server performs text analysis and uses natural language processing (NLP) algorithms to perform sentiment analysis and discriminatory expression detection. The input is text data, and the output is a sentiment score and a list of detected discriminatory expressions. This data operation uses an NLP model to analyze the text and identify sentiment and discriminatory expressions.

[1555] Step 6:

[1556] The server uses an emotion engine to determine the user's emotions. The input is the user's input data (text, voice, etc.), and the output is the result of the user's emotional state (e.g., "anger," "anxiety," etc.). This data calculation is performed by analyzing the user's emotions using an emotion analysis model.

[1557] Step 7:

[1558] The server determines the possibility of a controversy based on the analysis results. The input is the various analysis data obtained in the previous steps (face recognition results, object recognition results, emotion scores, etc.), and the output is the evaluation result of the controversy risk. This data calculation integrates each acquired data and performs a risk assessment.

[1559] Step 8:

[1560] Based on the results of the assessment, the server generates revision suggestions that take the user's emotions into account. The input is the flame risk assessment and the user's emotional data, and the output is a list of specific revision suggestions. This data calculation adjusts the content according to the user's emotions.

[1561] Step 9:

[1562] The server sends the generated revision suggestions to the terminal and presents them to the user via a user interface. The input is the generated revision suggestion data, and the output is the revision suggestions displayed on the user interface. This data processing involves sending the suggestion content to the terminal in an appropriate format.

[1563] Step 10:

[1564] The user modifies the content based on the suggested modifications and then sends the modified content back to the server from the device. The input is the modifications made by the user, and the output is the modified content. This data processing is a serialization process to ensure that the modifications are sent appropriately to the server.

[1565] Step 11:

[1566] The server re-analyzes the modified content to check for any new issues. The input is the modified content data, and the output is the result of the final analysis. This data calculation applies the re-analysis algorithm to check for any new issues.

[1567] Step 12:

[1568] The server posts the content that has been finalized to the SNS platform. The input is the finalized content data, and the output is the URL of the content posted to the SNS platform. This data processing involves posting the content to the SNS platform in the appropriate format.

[1569] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1570] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1571] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1572] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1573] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1574] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1575] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1576] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1577] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1578] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1579] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1580] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1581] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1582] 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.

[1583] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1584] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1585] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1586] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1587] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1588] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes wi...

Claims

1. means for receiving user-entered content; means for analyzing the content; A means for determining the possibility of a fire outbreak based on the results of the analysis; means for generating a revision suggestion based on the determination result; means for presenting said revision suggestions to a user; A means for the user to receive the revised content again and perform a final review; means for posting the finalized content to a social media platform; A system including:

2. The system of claim 1 , wherein the content analysis means includes video analysis, image analysis, and text analysis.

3. The system of claim 1 , wherein the means for presenting the revision suggestions displays the revision suggestions via a user interface.

Images