system

Abstract

Provide a system. 【Solution means】 Receiving means for receiving the input from the user from the information processing device, Classification means for classifying the received input into a plurality of formats, Analysis means for performing text analysis based on the classified format, Analysis means for performing video analysis based on the classified format, Analysis means for performing voice analysis based on the classified format, Storage means for storing records of past incident records, Evaluation means for evaluating the risk of incident by comparing the result by the analysis means with the record of the storage means, Notification means for outputting loss prediction and recommended content based on the evaluation result, A system including.

Description

Technical Field

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

Background Art

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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern society, problems frequently occur where the reputation of companies and individuals deteriorates significantly due to SNS posts. Such problems often lead directly to legal litigation and economic losses, so it is very important to prevent them. Also, it is difficult to pre-evaluate the potential for a firestorm caused by inappropriate posts, and no effective solutions have been provided so far. To address this issue, there is a need for technology that analyzes the content of SNS posts and predicts the risk in advance.

Means for Solving the Problems

[0005] This invention classifies SNS posts received from an information processing device into various formats and provides analysis means corresponding to each format. Specifically, it is a system that evaluates the risk of a post becoming a source of online harassment by using methods such as text analysis, video analysis, and audio analysis, and comparing them with records of past online harassment incidents. This system includes a receiving means, a classification means, an analysis means, a storage means, an evaluation means, and a notification means, and enables the provision of loss predictions and appropriate actions to the user based on the evaluation results. As a result, the user can recognize the risk before posting and make appropriate decisions. This is the means for solving the problem of this invention.

[0006] An "information processing device" is a computer system used for receiving, analyzing, and evaluating data.

[0007] "User" refers to an individual or legal entity that plans to post on social media.

[0008] "Receiving means" refers to the function of receiving data input by the user within the information processing device.

[0009] A "classification method" is a function that divides received data into text data, image data, audio data, and video data formats.

[0010] "Analysis means" refers to a function that analyzes the content of classified data and extracts inappropriate parts or characteristics.

[0011] A "memory device" is a database or storage device that preserves records of past online controversies.

[0012] The "evaluation method" is a function that uses the analysis results to calculate the risk of a post causing a social media firestorm.

[0013] "Notification means" refers to a display or audio output function that provides feedback to the user based on the evaluation results.

[0014] "The risk of going viral" is an indicator that shows the degree to which specific posted content attracts negative attention on SNS and may have harmful effects.

[0015] "Loss prediction" is a function of estimating potential economic losses and litigation risks based on the risk of going viral.

Brief Description of the Drawings

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

Mode for Carrying Out the Invention

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

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

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

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

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

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

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

[0024] [First Embodiment]

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

[0026] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

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

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

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

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

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

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

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

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0037] The system of this invention is designed to analyze the content of a post and predict its risks before the user posts it on social media. The user transmits text, images, audio, and video data they wish to post to the system via a terminal. This data is received by a server and classified according to its type. Based on the received data, the server applies natural language processing technology to the text, image recognition technology to the images, and speech recognition technology to the audio, performing analysis accordingly.

[0038] For example, if a user attempts to post offensive content, the server analyzes the post's text and compares it to a database of past online controversies. If similar problematic expressions are detected, the server determines that the risk of online backlash is high. Furthermore, if a posted image contains inappropriate symbols, the server analyzes them and issues a warning to the user.

[0039] Furthermore, the server integrates these analysis results and calculates the expected losses caused by the posted content. These losses include potential economic losses, including legal compensation. Based on the evaluation results, the server provides feedback to the user and suggests reviewing or canceling the post. This allows users to prevent inappropriate posts on social media. Through this process, the system provides an effective means of maintaining the credibility of companies and individuals and reducing risks.

[0040] The following describes the processing flow.

[0041] Step 1:

[0042] Users prepare text, images, audio, and video data they want to post to social media using their devices and upload them to the system. The devices then send this data to the server.

[0043] Step 2:

[0044] The server receives data sent from the terminal. The received data is classified into text data, image data, audio data, and video data.

[0045] Step 3:

[0046] The server applies natural language processing algorithms to the text data to analyze its content. This process identifies whether the data contains offensive language or discriminatory words.

[0047] Step 4:

[0048] The server uses image recognition technology to analyze the image data. It analyzes the poses of objects and people in the image and evaluates whether any inappropriate elements are included.

[0049] Step 5:

[0050] The server converts the audio data into text using speech recognition technology. The converted text is then checked again using natural language processing to ensure it does not contain any inappropriate content.

[0051] Step 6:

[0052] The server extracts frames from the video data and analyzes them using image recognition technology. Furthermore, audio within the video is analyzed separately.

[0053] Step 7:

[0054] The server integrates all analysis results and evaluates whether the information matches past online firestorm incidents. It calculates a firestorm risk score and estimates the expected amount of loss.

[0055] Step 8:

[0056] The server notifies the user of the evaluation results. This includes feedback detailing the risks and possible mitigation strategies. The user then uses this information to consider whether to revise their post.

[0057] (Example 1)

[0058] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0059] There is a growing need to prevent inappropriate posts on social networking services and other platforms. However, it is currently difficult for users to accurately judge the risk of their posts causing controversy. This invention aims to reduce inappropriate posts by automatically analyzing the content of posts, evaluating the risks, and providing feedback before users post them.

[0060] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0061] In this invention, the server includes means for receiving user information, means for classifying the received information into data formats, and means for processing based on the classified data formats. This makes it possible to analyze the risk of user posts becoming controversial in advance and provide appropriate feedback.

[0062] "Means for receiving user information" refers to the function by which the server receives text data, visual data, audio data, and video data sent from the terminal by the user.

[0063] "Means of classifying data into data formats" refers to the function of classifying received information into text data, visual data, audio data, and video data according to its content.

[0064] "Processing means for performing text analysis" refers to a function that analyzes received text data based on natural language processing technology to understand its content and meaning.

[0065] "Image analysis processing means" refers to a function that uses received visual data to detect specific patterns or inappropriate content.

[0066] "A processing means for performing audio analysis" refers to a function that converts received audio data into text and analyzes its meaning and emotion based on that text data.

[0067] "Video analysis processing means" refers to a function that analyzes received video data frame by frame and performs image and audio analysis necessary to understand the content.

[0068] A "storage means for retaining past data" refers to a function that stores past cases and data in a database for future analysis and evaluation.

[0069] A "means for assessing risk" is a function that quantifies the potential risks posed by a post by comparing the analysis results from the processing method with past data.

[0070] "Means for providing predictions and suggestions" refers to a function that generates and provides recommendations when conveying the results of a risk assessment to the user as feedback.

[0071] This system analyzes the content of posts made by users on social networking services before they are submitted, in order to predict the risk of controversy and inappropriateness. The specific configuration and operation of the system are described below.

[0072] Users first send the text, images, audio, or video data they wish to post to the system via their device. The device then transmits this data to the server over the internet. This communication utilizes the HTTPS protocol, ensuring secure data transfer.

[0073] The server has the ability to immediately check received data and classify it by type based on its format. For example, text data is analyzed using natural language processing techniques. Specific software used includes Python's NLTK library and spaCy. By utilizing these, aggressive language and negative sentiments in text can be identified.

[0074] Next, image recognition is performed on the image data using the machine learning framework TENSORFLOW®. For example, a pre-trained model is used to detect images containing inappropriate symbols or content.

[0075] The audio data is converted to text using the Google® Speech-to-Text API, and then sentiment analysis is performed on the text. This identifies aggressive remarks and emotional content contained in the audio.

[0076] The video data is analyzed frame by frame using OpenCV to detect inappropriate content from both image and audio elements.

[0077] The server uses these analysis results to compare with past database data and assess the potential risk of a post causing a backlash. This process calculates a risk score, which users can use to decide whether to revise or postpone their post.

[0078] For example, if a user tries to post a message saying, "This product is really terrible. You shouldn't use it!", the server can warn about this aggression through text analysis. Another example of a prompt for a generative AI model might be, "Please analyze whether the following content I plan to post on social media is offensive or at risk of causing controversy: 'This product is really terrible. You shouldn't use it!'"

[0079] This allows the system to provide users with appropriate feedback and prevent online controversies and misunderstandings. The system as a whole plays a crucial role in maintaining user trust and protecting the brand value of companies and individuals.

[0080] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0081] Step 1:

[0082] Users input text, images, audio, and video data through their devices and send that data. The input data is sent to the server as posted content. This process involves data transfer over the internet, and the HTTPS protocol is used to ensure security.

[0083] Step 2:

[0084] The server immediately analyzes received data by classifying it by type. Specifically, it identifies each data type as text data, visual data, audio data, or video data, and stores them in the appropriate directory or database. This classification allows subsequent processing steps to proceed more efficiently.

[0085] Step 3:

[0086] When the server receives text data, it performs text analysis using natural language processing techniques. The input is the received text data, and the output is the analysis results regarding the sentiment and aggression contained in the text. A Python library is used to segment the text and calculate the frequency of occurrence of specific keywords.

[0087] Step 4:

[0088] When the server receives visual data, it starts image analysis using an image recognition algorithm. The input is classified visual data, and the output is the analysis result identifying inappropriate content and symbols. Using TensorFlow or similar tools, a pre-trained model is applied to identify and evaluate objects within the image.

[0089] Step 5:

[0090] When the server receives audio data, it performs speech recognition processing to convert the audio data into text. The input is audio data, and the output is the converted text result. The Google Speech-to-Text API is used to convert the audio input into text data, and the result is then subjected to further text analysis.

[0091] Step 6:

[0092] When the server receives video data, it extracts all frames and applies image recognition technology. The input is video data, and the output is a report of inappropriateness for each frame obtained from the video. Using OpenCV, the video is decomposed into image frames, and the content of each frame is evaluated through image analysis.

[0093] Step 7:

[0094] Based on the analysis results, the server assesses the risk by comparing it with historical data stored in the database. Using the analyzed data as input, the server calculates the similarity to past online firestorm incidents and outputs the result as a risk score.

[0095] Step 8:

[0096] Based on the evaluation results, the server notifies the user of appropriate feedback. The input includes a risk score and recommended actions, and the output is a warning message the user receives. For example, if the risk is high, it will send a specific suggestion such as, "This post may be inappropriate. We recommend you make changes."

[0097] This series of steps creates a system that helps users mitigate risks before posting and supports safe and appropriate communication.

[0098] (Application Example 1)

[0099] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0100] In recent years, with the rapid dissemination of information in the real world, the spread of inappropriate content from individuals and businesses has increased. As a result, there have been numerous cases where the reputation of companies and individuals has been negatively affected. In particular, brick-and-mortar stores face the challenge of managing the risks associated with social media posts by staff and customers, as these posts spread instantly. It is necessary to prevent this from happening and make the flow of information in society safer and more reliable.

[0101] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0102] In this invention, the server includes a receiving means for receiving input from an information processing device, a classification means for classifying the received input into multiple formats, an analysis means for performing analysis based on the classified formats, a storage means for storing records of past risk cases, an evaluation means for comparing the results of the analysis means with the records in the storage means to evaluate the risk, a notification means for notifying the user in real time of loss predictions and recommendations based on the evaluation results, and a management means for managing information dissemination at real-world interaction locations and protecting reputation. This makes it possible to manage the risks of information dissemination to SNS even in physical stores, and to prevent problems caused by posted content.

[0103] An "information processing device" refers to a computer system that receives, transmits, and processes data, and is a device that receives user input in various formats.

[0104] A "receiving means" is a means that provides the function of receiving data from an external source to an information processing device, and plays the role of receiving various input data provided by the user.

[0105] A "classification method" is a means that provides a function to organize received data based on its attributes and format, and to divide it into appropriate categories.

[0106] "Analysis methods" refer to methods that use algorithms and techniques to apply advanced processing to classified data, understand its content, and analyze it.

[0107] A "memory device" is a means of storing past cases and data and providing a function to manage them so that they can be quickly searched and referenced.

[0108] An "evaluation tool" is a means that provides a function to predict and evaluate the likelihood of specific risks or problems by comparing analyzed data with stored data.

[0109] A "notification method" is a means that, based on the evaluation results, provides users with necessary information and communicates recommended actions and precautions.

[0110] "Management measures" refer to means for comprehensively monitoring and controlling the dissemination, distribution, and protection of information, and provide methods for appropriately managing information in the real world.

[0111] The server implements a system for managing information dissemination risks in physical stores. This system receives data transmitted from users' smart devices, classifies and analyzes it appropriately, compares it with past risk cases, and provides real-time feedback of the evaluation results.

[0112] On the smart device side, for example, a smartphone or tablet is used, and an application is installed on it. This application uses a program based on Python and TensorFlow as its execution environment and sends input such as text, images, and audio from the user to the server.

[0113] On the server, received data is automatically analyzed using natural language processing (NLP), image recognition, and speech recognition technologies. Specifically, for text, NLP is used to detect offensive or inappropriate expressions, and for images, a Convolutional Neural Network (CNN) is used to recognize inappropriate symbols and trademarks. Audio data is analyzed in a similar manner.

[0114] Next, the evaluation system compares the analysis results with a database of past risk cases and calculates a risk score. Based on this score, a notification is presented to the user's device. This allows the user to confirm the safety of immediate information dissemination within the store and to review or cancel the posted content as needed.

[0115] For example, if a customer who visited a cafe posts a review with a photo on social media, the app will detect if the photo contains a trademark and issue a warning. An example of a prompt to the generative AI model might be, "Please assess the risks of images containing specific brand names or trademarks."

[0116] In this way, this system can strengthen risk management for information dissemination in physical stores and provide a safe and secure communication environment for businesses and customers.

[0117] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0118] Step 1:

[0119] The device receives user input. During this process, the user inputs data such as text, images, and audio into the application via a smartphone or tablet. The input data is then prepared to be sent to the server.

[0120] Step 2:

[0121] The terminal sends the received data to the server. The data is transferred to the server via the network, where it is classified into an appropriate format for analysis. Input can include text, images, and audio.

[0122] Step 3:

[0123] The server receives the transmitted data and first organizes it into different formats using a classification mechanism. For example, text data is converted into string format, image data into pixel information, and audio data into a spectrum.

[0124] Step 4:

[0125] The server performs different analyses depending on the classified format. It uses NLP for text, CNN for images, and speech recognition technology for audio. For example, in text analysis, it performs grammatical analysis and keyword extraction to detect offensive expressions.

[0126] Step 5:

[0127] Based on the analysis results, the server compares them with a database of past risk cases. The evaluation system calculates a risk score, quantifying the level of risk. The risk score is determined based on the analysis results compared with the database.

[0128] Step 6:

[0129] The server sends the assessed risk score to the user's device via a notification system. The user receives the notification and decides whether to re-evaluate the posted content. An example of a prompt message generated by the AI ​​model is, "Please assess the risk of images containing specific brand names or trademarks."

[0130] Step 7:

[0131] Users will receive feedback from the server and, if necessary, revise their posts appropriately. If the re-evaluation reveals that a post contains problems, the user will edit the post to ensure safe information dissemination.

[0132] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0133] The present invention's system is designed to assess in advance the potential risk of online backlash caused by posts on social media, and further features analysis of the user's emotional state. When a user inputs the content they wish to post using a terminal, the terminal sends that data to a server. The server classifies the received data into text, audio, images, and videos, and applies an appropriate analysis process to each.

[0134] The server analyzes text data using natural language processing algorithms. Particular attention is paid to evaluating whether aggressive language or negative emotions are present. Image data is examined using image recognition technology to check for symbolic or inappropriate elements. Audio data is converted to text using speech recognition, and an emotion engine analyzes voice tone and intensity, allowing for a detailed examination of the emotional nuances of the posted content.

[0135] After these analyses are complete, the server integrates the results and calculates a crisis risk score by comparing them to past crisis cases. Sentiment analysis plays a significant role in this process, as the emotions the user is trying to convey have a major influence. The server then makes a loss prediction based on this and notifies the user as appropriate feedback. At this point, the user can obtain specific information to consider reviewing or deleting their post.

[0136] For example, if a user attempts to post a long text message containing many emotionally charged words, the server will perform a sentiment analysis of the text. If it determines that negative emotions are dominant, it will warn the user that there is a high risk of the message causing a social media firestorm. Similarly, in the analysis of audio data, if strong emotions such as anger or sadness are expressed, the server will alert the user to help them make appropriate decisions. In this way, the present invention provides a means to realize safer and healthier use of social media.

[0137] The following describes the processing flow.

[0138] Step 1:

[0139] Users input text, images, audio, and video data they want to post to social media into the system via their device and upload it. The device then sends this data to the server.

[0140] Step 2:

[0141] The server receives data from the terminal and classifies it into text, image, audio, and video formats. Based on this classification, the server initiates an analysis process appropriate to each format.

[0142] Step 3:

[0143] The server applies natural language processing algorithms to the text data. It extracts keywords and performs sentiment analysis to identify inappropriate expressions and negative emotions.

[0144] Step 4:

[0145] The server uses image recognition technology to detect inappropriate symbols and recognizable objects in the image data, paying particular attention to elements that may cause offense.

[0146] Step 5:

[0147] The server converts the audio data into text using speech recognition technology, and then analyzes that text using an emotion engine. It analyzes the tone and intensity of the voice and evaluates the underlying emotions.

[0148] Step 6:

[0149] For video data, the server extracts frames and performs image recognition on them. It also performs audio analysis, including the video's sound. This allows for the evaluation of the emotional impact of the combined visual and auditory elements.

[0150] Step 7:

[0151] After all analyses are complete, the server integrates the results. This allows for comparison and analysis with past online controversies, and calculates a controversy risk score for the posted content.

[0152] Step 8:

[0153] Based on the analysis results and risk assessment obtained by the server, the server sends specific feedback to the user. The user reviews the posted content based on the information provided and considers modifying or canceling the post if necessary.

[0154] (Example 2)

[0155] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0156] The challenge lies in proactively evaluating the potential risks of online backlash and negative emotional impacts of content posted by users on social media, thereby promoting safe and healthy communication. In today's information society, new methods are needed to mitigate the risks caused by inappropriate or emotionally charged posts.

[0157] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0158] In this invention, the server includes means for receiving user input from an information processing device, means for classifying the received input into multiple data formats, and means for performing text analysis using a natural language processing algorithm based on the classified text data. This enables appropriate analysis according to each data format posted by the user, allows for appropriate feedback after understanding the risks in advance, and supports safe communication.

[0159] An "information processing device" is an electronic device that has the functions of receiving, processing, and outputting data.

[0160] A "user" refers to a person or entity that operates the system and enters data.

[0161] "Receiving means" refers to the protocol or interface for receiving data transmitted from an information processing device.

[0162] "Data format" refers to the structure or type of information that is represented, and includes text, audio, images, and video.

[0163] A "classification method" refers to the process or mechanism of sorting received data according to its format.

[0164] "Means of performing text analysis" refers to the process of analyzing the content of text data using natural language processing techniques.

[0165] "Image recognition technology" refers to algorithms and systems used to extract and analyze specific information from image data.

[0166] "Speech recognition technology" is a technology that converts speech data into text data and analyzes its characteristics and emotions.

[0167] "Emotion inference" is the process of evaluating the type and intensity of emotions from analyzed data.

[0168] A "storage medium" is a device or medium that stores data and keeps it accessible at a later date.

[0169] "Means of risk assessment" refers to the process of quantifying or qualitatively evaluating the degree of potential risk based on the analysis results.

[0170] "Means of generating notifications" refers to a system that creates and presents messages and warnings to convey information to users.

[0171] This invention is a system that assesses the risks of content posted by users to social networking services (SNS) via an information processing device, thereby promoting safe communication. First, the user inputs text, audio, image, or video data they wish to post using a terminal. The terminal then transmits this input data to a server.

[0172] The server classifies the received data according to its format and applies the appropriate analysis process to each. Text data is analyzed using natural language processing libraries. Specifically, techniques such as extracting emotions and aggressive expressions from text are used, employing Python's NLTK and spaCy.

[0173] Audio data is converted to text on the server using speech recognition technology. The Google Speech-to-Text API is used, and then IBM Watson® is utilized to analyze voice tone and emotion. This allows for the evaluation of the emotional nuances of the audio content.

[0174] Image data is analyzed using TensorFlow and common image recognition techniques to check for symbolic or inappropriate elements. For example, provocative gestures or inappropriate objects can be automatically detected.

[0175] The analysis results are integrated, and the server generates a crisis risk score by comparing it to a database of past cases. Finally, based on this, a loss prediction is made and communicated to the user as specific feedback. This allows users to understand the risks of their posts in advance and take appropriate action.

[0176] For example, if a user attempts to post emotional text such as "This product is completely useless!", the server will analyze the text and issue a warning if negative sentiment is detected. Another example of a prompt message is, "Evaluate the degree of negative sentiment based on the content the user is attempting to post and generate an appropriate risk warning."

[0177] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0178] Step 1:

[0179] The user enters the content they want to post to social media using their device. The entered data can be in text, audio, image, or video format. Specifically, the user enters the content to be posted into an input form on their smartphone or computer and presses the submit button. The entered data is then ready to be sent to the server.

[0180] Step 2:

[0181] The terminal sends input data to the server. The data is in the form of an HTTP request sent to the server over the internet. Once the input data is transferred to the server, a series of analysis preparation processes begin.

[0182] Step 3:

[0183] The server classifies the received data. Specifically, it analyzes metadata to identify the data format and categorizes it into text, audio, image, and video. The classified data is then passed to the appropriate analysis process according to its format.

[0184] Step 4:

[0185] The server analyzes text data using natural language processing algorithms. The input is text data, and the output is the analysis result showing its emotional characteristics. Specifically, the server uses Python's natural language processing library to analyze and evaluate aggressive expressions and negative emotions from the text data.

[0186] Step 5:

[0187] The server converts audio data into text and then performs sentiment analysis. The input is audio data, and the output is the result of evaluating the emotions contained in the audio. Specifically, it uses speech recognition technology to convert the audio into text, and then uses an emotion engine to analyze the tone and emotions of the voice.

[0188] Step 6:

[0189] The server analyzes image data using image recognition technology. The input is image data, and the output is an evaluation based on the image content. Specifically, it uses an image recognition algorithm to detect symbolic or inappropriate elements within the image and generates results.

[0190] Step 7:

[0191] The server integrates the results of each analysis and calculates a crisis risk score. The input is the results of each analysis, and the output is an overall risk assessment. Specifically, the server compares the analysis results with past case data and objectively evaluates the risks involved.

[0192] Step 8:

[0193] The server generates notifications based on the evaluation results and provides feedback to the user. The input is the risk score, and the output is the feedback notification. Specifically, the server generates a warning message according to the risk level and sends it to the user. This gives the user an opportunity to review their posts.

[0194] (Application Example 2)

[0195] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0196] In online advertising, deploying ads without predicting consumer emotions and reactions can lead to backlash and negative responses, potentially harming a company's brand image and sales. Therefore, it is essential to evaluate how the content will be received by consumers and the level of backlash risk before the ad is delivered, and to take preventative measures.

[0197] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0198] In this invention, the server includes receiving means for receiving user input from an information processing device, classification means for classifying the received input into multiple formats, and advertising analysis means for evaluating in advance how advertising content will affect the user's emotions and providing feedback. This makes it possible to predict consumer reactions before advertising is delivered and reduce the risk of backlash.

[0199] An "information processing device" is a computer or server used to receive and analyze data.

[0200] A "receiving means" is an element that performs the function of collecting data provided by the user and incorporating it into the system.

[0201] A "classification means" is an element that has the function of identifying and sorting received data into specific formats such as text, images, audio, and video.

[0202] An "analysis tool" is an element that has the function of analyzing data of each classified format and evaluating its content.

[0203] A "memory device" is a medium or device for accumulating and storing past online firestorm incidents that are compared with the analyzed data.

[0204] "Evaluation methods" refer to a mechanism that assesses the likelihood of a social media firestorm by comparing the analysis results with past data.

[0205] A "notification method" is an element that has a method or technique for providing feedback on evaluation results to the user.

[0206] "Advertising analysis tools" are elements that have the function of evaluating the emotional impact that advertising content has on consumers and predicting the risk of backlash.

[0207] To realize this invention, a server, an information processing device, a user terminal, and a communication network connecting them are necessary. The server uses software such as Python, TensorFlow, NLTK, and spaCy to analyze the received data. Specifically, the server acquires advertising content received from the user terminal via a receiving means and classifies it into different formats. Using a classification means, it separates the content into text, image, audio, and video formats, and an analysis means performs an analysis appropriate to each format.

[0208] For text data, natural language processing algorithms are applied to extract offensive words and negative expressions. Image data is analyzed using image recognition technology to attempt to detect inappropriate elements and symbolic content. Audio data is converted to text using speech recognition technology, and then an emotion engine evaluates the emotional nuances. Similar processing is performed on video data.

[0209] The analysis results are compared with past online controversies stored in memory, and an evaluation tool calculates the risk of a controversy. This score is then provided to the user visually or audibly through a notification tool. Additionally, an advertising analysis tool evaluates the emotional impact of the advertisement on consumers and provides feedback. In this process, a generative AI model assists by generating or suggesting revisions to the advertisement based on prompt text.

[0210] As a concrete example, a user uploads an ad video to a server before launching a new advertising campaign. The prompt might be something like, "Analyze the new campaign ad video and predict its social impact." The server analyzes this and returns feedback such as, "There is a high probability of a negative reaction," allowing the user to gain specific guidance for improving the campaign.

[0211] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0212] Step 1:

[0213] The user inputs advertising content using their device and sends it to the server. The input data can take the form of text, images, audio, or video in digital format. The server receives this input and begins processing it within the system.

[0214] Step 2:

[0215] The server classifies received content into four categories—text, images, audio, and video—using classification mechanisms. This classification is necessary to apply an analysis process appropriate to the data format. For example, the data format is identified by determining the file extension.

[0216] Step 3:

[0217] The server uses natural language processing algorithms to extract aggressive expressions and negative emotions from text data. The input is raw text data, and the output is a set of analyzed emotion information. This process is performed using a natural language processing library.

[0218] Step 4:

[0219] Image data is analyzed using image recognition technology to detect symbolic or inappropriate elements. The input is image data, and the output is the analysis result as metadata. An image processing library is used for this process.

[0220] Step 5:

[0221] The audio data is converted to text by a speech recognition engine, and then sentiment analysis is performed, similar to that for text. The input is an audio file, and the output is the corresponding text along with its sentiment information. Speech recognition is performed using speech analysis software.

[0222] Step 6:

[0223] The video data is analyzed using a video analysis tool to extract still images from each frame, and then the image recognition process from step 4 is applied to these still images. The input is a video file, and the output is the analysis results for each frame.

[0224] Step 7:

[0225] The server compares the analysis results with past online firestorm incidents stored in a memory device and calculates a firestorm risk score using an evaluation device. The input is the analysis results for each data format, and the output is the firestorm risk score. A historical database is used for comparison.

[0226] Step 8:

[0227] Through a notification system, users are provided with a risk score for online controversy and feedback based on that score. The input is the score, and the output is text and warning messages displayed on the user's device. This allows users to review and modify their advertising content.

[0228] This series of processes provides users with guidelines for generating or modifying ad content based on prompts from the generation AI model.

[0229] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0230] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0231] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0232] [Second Embodiment]

[0233] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0234] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0235] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0236] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0237] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0238] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0239] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0240] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0241] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0242] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0243] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0244] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0245] The system of this invention is designed to analyze the content of a post and predict its risks before the user posts it on social media. The user transmits text, images, audio, and video data they wish to post to the system via a terminal. This data is received by a server and classified according to its type. Based on the received data, the server applies natural language processing technology to the text, image recognition technology to the images, and speech recognition technology to the audio, performing analysis accordingly.

[0246] For example, if a user attempts to post offensive content, the server analyzes the post's text and compares it to a database of past online controversies. If similar problematic expressions are detected, the server determines that the risk of online backlash is high. Furthermore, if a posted image contains inappropriate symbols, the server analyzes them and issues a warning to the user.

[0247] Furthermore, the server integrates these analysis results and calculates the expected losses caused by the posted content. These losses include potential economic losses, including legal compensation. Based on the evaluation results, the server provides feedback to the user and suggests reviewing or canceling the post. This allows users to prevent inappropriate posts on social media. Through this process, the system provides an effective means of maintaining the credibility of companies and individuals and reducing risks.

[0248] The following describes the processing flow.

[0249] Step 1:

[0250] Users prepare text, images, audio, and video data they want to post to social media using their devices and upload them to the system. The devices then send this data to the server.

[0251] Step 2:

[0252] The server receives data sent from the terminal. The received data is classified into text data, image data, audio data, and video data.

[0253] Step 3:

[0254] The server applies natural language processing algorithms to the text data to analyze its content. This process identifies whether the data contains offensive language or discriminatory words.

[0255] Step 4:

[0256] The server uses image recognition technology to analyze the image data. It analyzes the poses of objects and people in the image and evaluates whether any inappropriate elements are included.

[0257] Step 5:

[0258] The server converts the audio data into text using speech recognition technology. The converted text is then checked again using natural language processing to ensure it does not contain any inappropriate content.

[0259] Step 6:

[0260] The server extracts frames from the video data and analyzes them using image recognition technology. Furthermore, audio within the video is analyzed separately.

[0261] Step 7:

[0262] The server integrates all analysis results and evaluates whether the information matches past online firestorm incidents. It calculates a firestorm risk score and estimates the expected amount of loss.

[0263] Step 8:

[0264] The server notifies the user of the evaluation results. This includes feedback detailing the risks and possible mitigation strategies. The user then uses this information to consider whether to revise their post.

[0265] (Example 1)

[0266] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0267] There is a growing need to prevent inappropriate posts on social networking services and other platforms. However, it is currently difficult for users to accurately judge the risk of their posts causing controversy. This invention aims to reduce inappropriate posts by automatically analyzing the content of posts, evaluating the risks, and providing feedback before users post them.

[0268] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0269] In this invention, the server includes means for receiving user information, means for classifying the received information into data formats, and means for processing based on the classified data formats. This makes it possible to analyze the risk of user posts becoming controversial in advance and provide appropriate feedback.

[0270] "Means for receiving user information" refers to the function by which the server receives text data, visual data, audio data, and video data sent from the terminal by the user.

[0271] "Means of classifying data into data formats" refers to the function of classifying received information into text data, visual data, audio data, and video data according to its content.

[0272] "Processing means for performing text analysis" refers to a function that analyzes received text data based on natural language processing technology to understand its content and meaning.

[0273] "Image analysis processing means" refers to a function that uses received visual data to detect specific patterns or inappropriate content.

[0274] "A processing means for performing audio analysis" refers to a function that converts received audio data into text and analyzes its meaning and emotion based on that text data.

[0275] "Video analysis processing means" refers to a function that analyzes received video data frame by frame and performs image and audio analysis necessary to understand the content.

[0276] A "storage means for retaining past data" refers to a function that stores past cases and data in a database for future analysis and evaluation.

[0277] A "means for assessing risk" is a function that quantifies the potential risks posed by a post by comparing the analysis results from the processing method with past data.

[0278] "Means for providing predictions and suggestions" refers to a function that generates and provides recommendations when conveying the results of a risk assessment to the user as feedback.

[0279] This system analyzes the content of posts made by users on social networking services before they are submitted, in order to predict the risk of controversy and inappropriateness. The specific configuration and operation of the system are described below.

[0280] Users first send the text, images, audio, or video data they wish to post to the system via their device. The device then transmits this data to the server over the internet. This communication utilizes the HTTPS protocol, ensuring secure data transfer.

[0281] The server has the ability to immediately check received data and classify it by type based on its format. For example, text data is analyzed using natural language processing techniques. Specific software used includes Python's NLTK library and spaCy. By utilizing these, aggressive language and negative sentiments in text can be identified.

[0282] Next, the image data is used for image recognition using TensorFlow, a machine learning framework. For example, a pre-trained model is used to detect images containing inappropriate symbols or content. ;

[0283] ; ; For audio data, it is converted to text using the Google Speech-to-Text API, and then sentiment analysis is performed on the text again. This is used to identify aggressive or emotional content contained in the audio. ;

[0284] ; ; Video data is analyzed frame by frame using OpenCV to detect inappropriate content from both image and audio elements. ;

[0285] ; ; Based on these analysis results, the server compares with the past database to evaluate the potential risk of an online feud caused by the post. In this process, a risk score is calculated, and users can judge whether to review or postpone the content based on this score. ;

[0286] ; ; As a specific example, when a user tries to post a message saying "This product is really bad. It's better not to use it!", the server can warn of this aggressiveness through text analysis. Also, as an example of a prompt sentence for the generative AI model, something like "Regarding the following content planned to be posted on SNS, please analyze whether there is a risk of aggression or an online feud: 'This product is really bad. It's better not to use it!'" can be considered. ;

[0287] ; ; This enables the system to provide appropriate feedback to users and prevent online feuds and misunderstandings on SNS. It plays an important role in maintaining the reliability of users for the entire system and protecting the brand value of companies and individuals. ;

[0288] ; ; The flow of the specific process in Example 1 will be described using FIG. 11. ;

[0289] ; ; Step 1: ;

[0290] ; ;Users input text, images, audio, and video data through their devices and send that data. The input data is sent to the server as posted content. This process involves data transfer over the internet, and the HTTPS protocol is used to ensure security.

[0291] Step 2:

[0292] The server immediately analyzes received data by classifying it by type. Specifically, it identifies each data type as text data, visual data, audio data, or video data, and stores them in the appropriate directory or database. This classification allows subsequent processing steps to proceed more efficiently.

[0293] Step 3:

[0294] When the server receives text data, it performs text analysis using natural language processing techniques. The input is the received text data, and the output is the analysis results regarding the sentiment and aggression contained in the text. A Python library is used to segment the text and calculate the frequency of occurrence of specific keywords.

[0295] Step 4:

[0296] When the server receives visual data, it starts image analysis using an image recognition algorithm. The input is classified visual data, and the output is the analysis result identifying inappropriate content and symbols. Using TensorFlow or similar tools, a pre-trained model is applied to identify and evaluate objects within the image.

[0297] Step 5:

[0298] When the server receives audio data, it performs speech recognition processing to convert the audio data into text. The input is audio data, and the output is the converted text result. The Google Speech-to-Text API is used to convert the audio input into text data, and the result is then subjected to further text analysis.

[0299] Step 6:

[0300] When the server receives video data, it extracts all frames and applies image recognition technology. The input is video data, and the output is a report of inappropriateness for each frame obtained from the video. Using OpenCV, the video is decomposed into image frames, and the content of each frame is evaluated through image analysis.

[0301] Step 7:

[0302] Based on the results of the analysis, the server evaluates the risk by comparing it with past data stored in the database. Using the analyzed data as input information, the server calculates the similarity with past fire incidents and outputs the result as a risk score.

[0303] Step 8:

[0304] Based on the evaluation results, the server notifies the user of appropriate feedback. The input includes the risk score and recommended actions, and the output is the warning message received by the user. For example, when the risk is high, a specific proposal such as "This post may be inappropriate. We recommend making changes" is sent.

[0305] Through this series of steps, a system is constructed that helps users reduce risks before posting and enables safe and appropriate communication.

[0306] (Application Example 1)

[0307] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0308] In recent years, with the rapid dissemination of information in the real world, the spread of inappropriate content from individuals and businesses has increased. As a result, there have been numerous cases where the reputation of companies and individuals has been negatively affected. In particular, brick-and-mortar stores face the challenge of managing the risks associated with social media posts by staff and customers, as these posts spread instantly. It is necessary to prevent this from happening and make the flow of information in society safer and more reliable.

[0309] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0310] In this invention, the server includes a receiving means for receiving input from an information processing device, a classification means for classifying the received input into multiple formats, an analysis means for performing analysis based on the classified formats, a storage means for storing records of past risk cases, an evaluation means for comparing the results of the analysis means with the records in the storage means to evaluate the risk, a notification means for notifying the user in real time of loss predictions and recommendations based on the evaluation results, and a management means for managing information dissemination at real-world interaction locations and protecting reputation. This makes it possible to manage the risks of information dissemination to SNS even in physical stores, and to prevent problems caused by posted content.

[0311] An "information processing device" refers to a computer system that receives, transmits, and processes data, and is a device that receives user input in various formats.

[0312] A "receiving means" is a means that provides the function of receiving data from an external source to an information processing device, and plays the role of receiving various input data provided by the user.

[0313] A "classification method" is a means that provides a function to organize received data based on its attributes and format, and to divide it into appropriate categories.

[0314] "Analysis methods" refer to methods that use algorithms and techniques to apply advanced processing to classified data, understand its content, and analyze it.

[0315] A "memory device" is a means of storing past cases and data and providing a function to manage them so that they can be quickly searched and referenced.

[0316] An "evaluation tool" is a means that provides a function to predict and evaluate the likelihood of specific risks or problems by comparing analyzed data with stored data.

[0317] A "notification method" is a means that, based on the evaluation results, provides users with necessary information and communicates recommended actions and precautions.

[0318] "Management measures" refer to means for comprehensively monitoring and controlling the dissemination, distribution, and protection of information, and provide methods for appropriately managing information in the real world.

[0319] The server implements a system for managing information dissemination risks in physical stores. This system receives data transmitted from users' smart devices, classifies and analyzes it appropriately, compares it with past risk cases, and provides real-time feedback of the evaluation results.

[0320] On the smart device side, for example, a smartphone or tablet is used, and an application is installed on it. This application uses a program based on Python and TensorFlow as its execution environment and sends input such as text, images, and audio from the user to the server.

[0321] On the server, received data is automatically analyzed using natural language processing (NLP), image recognition, and speech recognition technologies. Specifically, for text, NLP is used to detect offensive or inappropriate expressions, and for images, a Convolutional Neural Network (CNN) is used to recognize inappropriate symbols and trademarks. Audio data is analyzed in a similar manner.

[0322] Next, the evaluation system compares the analysis results with a database of past risk cases and calculates a risk score. Based on this score, a notification is presented to the user's device. This allows the user to confirm the safety of immediate information dissemination within the store and to review or cancel the posted content as needed.

[0323] For example, if a customer who visited a cafe posts a review with a photo on social media, the app will detect if the photo contains a trademark and issue a warning. An example of a prompt to the generative AI model might be, "Please assess the risks of images containing specific brand names or trademarks."

[0324] In this way, this system can strengthen risk management for information dissemination in physical stores and provide a safe and secure communication environment for businesses and customers.

[0325] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0326] Step 1:

[0327] The device receives user input. During this process, the user inputs data such as text, images, and audio into the application via a smartphone or tablet. The input data is then prepared to be sent to the server.

[0328] Step 2:

[0329] The terminal sends the received data to the server. The data is transferred to the server via the network, where it is classified into an appropriate format for analysis. Input can include text, images, and audio.

[0330] Step 3:

[0331] The server receives the transmitted data and first organizes it into different formats using a classification mechanism. For example, text data is converted into string format, image data into pixel information, and audio data into a spectrum.

[0332] Step 4:

[0333] The server performs different analyses depending on the classified format. It uses NLP for text, CNN for images, and speech recognition technology for audio. For example, in text analysis, it performs grammatical analysis and keyword extraction to detect offensive expressions.

[0334] Step 5:

[0335] Based on the analysis results, the server compares them with a database of past risk cases. The evaluation system calculates a risk score, quantifying the level of risk. The risk score is determined based on the analysis results compared with the database.

[0336] Step 6:

[0337] The server sends the assessed risk score to the user's device via a notification system. The user receives the notification and decides whether to re-evaluate the posted content. An example of a prompt message generated by the AI ​​model is, "Please assess the risk of images containing specific brand names or trademarks."

[0338] Step 7:

[0339] Users will receive feedback from the server and, if necessary, revise their posts appropriately. If the re-evaluation reveals that a post contains problems, the user will edit the post to ensure safe information dissemination.

[0340] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0341] The present invention's system is designed to assess in advance the potential risk of online backlash caused by posts on social media, and further features analysis of the user's emotional state. When a user inputs the content they wish to post using a terminal, the terminal sends that data to a server. The server classifies the received data into text, audio, images, and videos, and applies an appropriate analysis process to each.

[0342] The server analyzes text data using natural language processing algorithms. Particular attention is paid to evaluating whether aggressive language or negative emotions are present. Image data is examined using image recognition technology to check for symbolic or inappropriate elements. Audio data is converted to text using speech recognition, and an emotion engine analyzes voice tone and intensity, allowing for a detailed examination of the emotional nuances of the posted content.

[0343] After these analyses are complete, the server integrates the results and calculates a crisis risk score by comparing them to past crisis cases. Sentiment analysis plays a significant role in this process, as the emotions the user is trying to convey have a major influence. The server then makes a loss prediction based on this and notifies the user as appropriate feedback. At this point, the user can obtain specific information to consider reviewing or deleting their post.

[0344] For example, if a user attempts to post a long text message containing many emotionally charged words, the server will perform a sentiment analysis of the text. If it determines that negative emotions are dominant, it will warn the user that there is a high risk of the message causing a social media firestorm. Similarly, in the analysis of audio data, if strong emotions such as anger or sadness are expressed, the server will alert the user to help them make appropriate decisions. In this way, the present invention provides a means to realize safer and healthier use of social media.

[0345] The following describes the processing flow.

[0346] Step 1:

[0347] Users input text, images, audio, and video data they want to post to social media into the system via their device and upload it. The device then sends this data to the server.

[0348] Step 2:

[0349] The server receives data from the terminal and classifies it into text, image, audio, and video formats. Based on this classification, the server initiates an analysis process appropriate to each format.

[0350] Step 3:

[0351] The server applies natural language processing algorithms to the text data. It extracts keywords and performs sentiment analysis to identify inappropriate expressions and negative emotions.

[0352] Step 4:

[0353] The server uses image recognition technology to detect inappropriate symbols and recognizable objects in the image data, paying particular attention to elements that may cause offense.

[0354] Step 5:

[0355] The server converts the audio data into text using speech recognition technology, and then analyzes that text using an emotion engine. It analyzes the tone and intensity of the voice and evaluates the underlying emotions.

[0356] Step 6:

[0357] For video data, the server extracts frames and performs image recognition on them. It also performs audio analysis, including the video's sound. This allows for the evaluation of the emotional impact of the combined visual and auditory elements.

[0358] Step 7:

[0359] After all analyses are complete, the server integrates the results. This allows for comparison and analysis with past online controversies, and calculates a controversy risk score for the posted content.

[0360] Step 8:

[0361] Based on the analysis results and risk assessment obtained by the server, the server sends specific feedback to the user. The user reviews the posted content based on the information provided and considers modifying or canceling the post if necessary.

[0362] (Example 2)

[0363] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0364] The challenge lies in proactively evaluating the potential risks of online backlash and negative emotional impacts of content posted by users on social media, thereby promoting safe and healthy communication. In today's information society, new methods are needed to mitigate the risks caused by inappropriate or emotionally charged posts.

[0365] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0366] In this invention, the server includes means for receiving user input from an information processing device, means for classifying the received input into multiple data formats, and means for performing text analysis using a natural language processing algorithm based on the classified text data. This enables appropriate analysis according to each data format posted by the user, allows for appropriate feedback after understanding the risks in advance, and supports safe communication.

[0367] An "information processing device" is an electronic device that has the functions of receiving, processing, and outputting data.

[0368] A "user" refers to a person or entity that operates the system and enters data.

[0369] "Receiving means" refers to the protocol or interface for receiving data transmitted from an information processing device.

[0370] "Data format" refers to the structure or type of information that is represented, and includes text, audio, images, and video.

[0371] A "classification method" refers to the process or mechanism of sorting received data according to its format.

[0372] "Means of performing text analysis" refers to the process of analyzing the content of text data using natural language processing techniques.

[0373] "Image recognition technology" refers to algorithms and systems used to extract and analyze specific information from image data.

[0374] "Speech recognition technology" is a technology that converts speech data into text data and analyzes its characteristics and emotions.

[0375] "Emotion inference" is the process of evaluating the type and intensity of emotions from analyzed data.

[0376] A "storage medium" is a device or medium that stores data and keeps it accessible at a later date.

[0377] "Means of risk assessment" refers to the process of quantifying or qualitatively evaluating the degree of potential risk based on the analysis results.

[0378] "Means of generating notifications" refers to a system that creates and presents messages and warnings to convey information to users.

[0379] This invention is a system that assesses the risks of content posted by users to social networking services (SNS) via an information processing device, thereby promoting safe communication. First, the user inputs text, audio, image, or video data they wish to post using a terminal. The terminal then transmits this input data to a server.

[0380] The server classifies the received data according to its format and applies the appropriate analysis process to each. Text data is analyzed using natural language processing libraries. Specifically, techniques such as extracting emotions and aggressive expressions from text are used, employing Python's NLTK and spaCy.

[0381] The audio data is converted to text on the server using speech recognition technology. The Google Speech-to-Text API is used, and then IBM Watson is utilized to analyze the tone and emotion of the voice. This allows for the evaluation of the emotional nuances of the audio content.

[0382] Image data is analyzed using TensorFlow and common image recognition techniques to check for symbolic or inappropriate elements. For example, provocative gestures or inappropriate objects can be automatically detected.

[0383] The analysis results are integrated, and the server generates a crisis risk score by comparing it to a database of past cases. Finally, based on this, a loss prediction is made and communicated to the user as specific feedback. This allows users to understand the risks of their posts in advance and take appropriate action.

[0384] For example, if a user attempts to post emotional text such as "This product is completely useless!", the server will analyze the text and issue a warning if negative sentiment is detected. Another example of a prompt message is, "Evaluate the degree of negative sentiment based on the content the user is attempting to post and generate an appropriate risk warning."

[0385] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0386] Step 1:

[0387] The user enters the content they want to post to social media using their device. The entered data can be in text, audio, image, or video format. Specifically, the user enters the content to be posted into an input form on their smartphone or computer and presses the submit button. The entered data is then ready to be sent to the server.

[0388] Step 2:

[0389] The terminal sends input data to the server. The data is in the form of an HTTP request sent to the server over the internet. Once the input data is transferred to the server, a series of analysis preparation processes begin.

[0390] Step 3:

[0391] The server classifies the received data. Specifically, it analyzes metadata to identify the data format and categorizes it into text, audio, image, and video. The classified data is then passed to the appropriate analysis process according to its format.

[0392] Step 4:

[0393] The server analyzes text data using natural language processing algorithms. The input is text data, and the output is the analysis result showing its emotional characteristics. Specifically, the server uses Python's natural language processing library to analyze and evaluate aggressive expressions and negative emotions from the text data.

[0394] Step 5:

[0395] The server converts audio data into text and then performs sentiment analysis. The input is audio data, and the output is the result of evaluating the emotions contained in the audio. Specifically, it uses speech recognition technology to convert the audio into text, and then uses an emotion engine to analyze the tone and emotions of the voice.

[0396] Step 6:

[0397] The server analyzes image data using image recognition technology. The input is image data, and the output is an evaluation based on the image content. Specifically, it uses an image recognition algorithm to detect symbolic or inappropriate elements within the image and generates results.

[0398] Step 7:

[0399] The server integrates the results of each analysis and calculates a crisis risk score. The input is the results of each analysis, and the output is an overall risk assessment. Specifically, the server compares the analysis results with past case data and objectively evaluates the risks involved.

[0400] Step 8:

[0401] The server generates notifications based on the evaluation results and provides feedback to the user. The input is the risk score, and the output is the feedback notification. Specifically, the server generates a warning message according to the risk level and sends it to the user. This gives the user an opportunity to review their posts.

[0402] (Application Example 2)

[0403] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0404] In online advertising, deploying ads without predicting consumer emotions and reactions can lead to backlash and negative responses, potentially harming a company's brand image and sales. Therefore, it is essential to evaluate how the content will be received by consumers and the level of backlash risk before the ad is delivered, and to take preventative measures.

[0405] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0406] In this invention, the server includes receiving means for receiving user input from an information processing device, classification means for classifying the received input into multiple formats, and advertising analysis means for evaluating in advance how advertising content will affect the user's emotions and providing feedback. This makes it possible to predict consumer reactions before advertising is delivered and reduce the risk of backlash.

[0407] An "information processing device" is a computer or server used to receive and analyze data.

[0408] A "receiving means" is an element that performs the function of collecting data provided by the user and incorporating it into the system.

[0409] A "classification means" is an element that has the function of identifying and sorting received data into specific formats such as text, images, audio, and video.

[0410] An "analysis tool" is an element that has the function of analyzing data of each classified format and evaluating its content.

[0411] A "memory device" is a medium or device for accumulating and storing past online firestorm incidents that are compared with the analyzed data.

[0412] "Evaluation methods" refer to a mechanism that assesses the likelihood of a social media firestorm by comparing the analysis results with past data.

[0413] A "notification method" is an element that has a method or technique for providing feedback on evaluation results to the user.

[0414] "Advertising analysis tools" are elements that have the function of evaluating the emotional impact that advertising content has on consumers and predicting the risk of backlash.

[0415] To realize this invention, a server, an information processing device, a user terminal, and a communication network connecting them are necessary. The server uses software such as Python, TensorFlow, NLTK, and spaCy to analyze the received data. Specifically, the server acquires advertising content received from the user terminal via a receiving means and classifies it into different formats. Using a classification means, it separates the content into text, image, audio, and video formats, and an analysis means performs an analysis appropriate to each format.

[0416] For text data, natural language processing algorithms are applied to extract offensive words and negative expressions. Image data is analyzed using image recognition technology to attempt to detect inappropriate elements and symbolic content. Audio data is converted to text using speech recognition technology, and then an emotion engine evaluates the emotional nuances. Similar processing is performed on video data.

[0417] The analysis results are compared with past online controversies stored in memory, and an evaluation tool calculates the risk of a controversy. This score is then provided to the user visually or audibly through a notification tool. Additionally, an advertising analysis tool evaluates the emotional impact of the advertisement on consumers and provides feedback. In this process, a generative AI model assists by generating or suggesting revisions to the advertisement based on prompt text.

[0418] As a concrete example, a user uploads an ad video to a server before launching a new advertising campaign. The prompt might be something like, "Analyze the new campaign ad video and predict its social impact." The server analyzes this and returns feedback such as, "There is a high probability of a negative reaction," allowing the user to gain specific guidance for improving the campaign.

[0419] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0420] Step 1:

[0421] The user inputs advertising content using their device and sends it to the server. The input data can take the form of text, images, audio, or video in digital format. The server receives this input and begins processing it within the system.

[0422] Step 2:

[0423] The server classifies received content into four categories—text, images, audio, and video—using classification mechanisms. This classification is necessary to apply an analysis process appropriate to the data format. For example, the data format is identified by determining the file extension.

[0424] Step 3:

[0425] The server uses natural language processing algorithms to extract aggressive expressions and negative emotions from text data. The input is raw text data, and the output is a set of analyzed emotion information. This process is performed using a natural language processing library.

[0426] Step 4:

[0427] Image data is analyzed using image recognition technology to detect symbolic or inappropriate elements. The input is image data, and the output is the analysis result as metadata. An image processing library is used for this process.

[0428] Step 5:

[0429] The audio data is converted to text by a speech recognition engine, and then sentiment analysis is performed, similar to that for text. The input is an audio file, and the output is the corresponding text along with its sentiment information. Speech recognition is performed using speech analysis software.

[0430] Step 6:

[0431] The video data is analyzed using a video analysis tool to extract still images from each frame, and then the image recognition process from step 4 is applied to these still images. The input is a video file, and the output is the analysis results for each frame.

[0432] Step 7:

[0433] The server compares the analysis results with past online firestorm incidents stored in a memory device and calculates a firestorm risk score using an evaluation device. The input is the analysis results for each data format, and the output is the firestorm risk score. A historical database is used for comparison.

[0434] Step 8:

[0435] Through a notification system, users are provided with a risk score for online controversy and feedback based on that score. The input is the score, and the output is text and warning messages displayed on the user's device. This allows users to review and modify their advertising content.

[0436] This series of processes provides users with guidelines for generating or modifying ad content based on prompts from the generation AI model.

[0437] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0438] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0439] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0440] [Third Embodiment]

[0441] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0442] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0443] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0444] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0445] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0446] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0447] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0448] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0449] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0450] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0451] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0452] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0453] The system of this invention is designed to analyze the content of a post and predict its risks before the user posts it on social media. The user transmits text, images, audio, and video data they wish to post to the system via a terminal. This data is received by a server and classified according to its type. Based on the received data, the server applies natural language processing technology to the text, image recognition technology to the images, and speech recognition technology to the audio, performing analysis accordingly.

[0454] For example, if a user attempts to post offensive content, the server analyzes the post's text and compares it to a database of past online controversies. If similar problematic expressions are detected, the server determines that the risk of online backlash is high. Furthermore, if a posted image contains inappropriate symbols, the server analyzes them and issues a warning to the user.

[0455] Furthermore, the server integrates these analysis results and calculates the expected losses caused by the posted content. These losses include potential economic losses, including legal compensation. Based on the evaluation results, the server provides feedback to the user and suggests reviewing or canceling the post. This allows users to prevent inappropriate posts on social media. Through this process, the system provides an effective means of maintaining the credibility of companies and individuals and reducing risks.

[0456] The following describes the processing flow.

[0457] Step 1:

[0458] Users prepare text, images, audio, and video data they want to post to social media using their devices and upload them to the system. The devices then send this data to the server.

[0459] Step 2:

[0460] The server receives data sent from the terminal. The received data is classified into text data, image data, audio data, and video data.

[0461] Step 3:

[0462] The server applies natural language processing algorithms to the text data to analyze its content. This process identifies whether the data contains offensive language or discriminatory words.

[0463] Step 4:

[0464] The server uses image recognition technology to analyze the image data. It analyzes the poses of objects and people in the image and evaluates whether any inappropriate elements are included.

[0465] Step 5:

[0466] The server converts the audio data into text using speech recognition technology. The converted text is then checked again using natural language processing to ensure it does not contain any inappropriate content.

[0467] Step 6:

[0468] The server extracts frames from the video data and analyzes them using image recognition technology. Furthermore, audio within the video is analyzed separately.

[0469] Step 7:

[0470] The server integrates all analysis results and evaluates whether the information matches past online firestorm incidents. It calculates a firestorm risk score and estimates the expected amount of loss.

[0471] Step 8:

[0472] The server notifies the user of the evaluation results. This includes feedback detailing the risks and possible mitigation strategies. The user then uses this information to consider whether to revise their post.

[0473] (Example 1)

[0474] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0475] There is a growing need to prevent inappropriate posts on social networking services and other platforms. However, it is currently difficult for users to accurately judge the risk of their posts causing controversy. This invention aims to reduce inappropriate posts by automatically analyzing the content of posts, evaluating the risks, and providing feedback before users post them.

[0476] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0477] In this invention, the server includes means for receiving user information, means for classifying the received information into data formats, and means for processing based on the classified data formats. This makes it possible to analyze the risk of user posts becoming controversial in advance and provide appropriate feedback.

[0478] "Means for receiving user information" refers to the function by which the server receives text data, visual data, audio data, and video data sent from the terminal by the user.

[0479] "Means of classifying data into data formats" refers to the function of classifying received information into text data, visual data, audio data, and video data according to its content.

[0480] "Processing means for performing text analysis" refers to a function that analyzes received text data based on natural language processing technology to understand its content and meaning.

[0481] "Image analysis processing means" refers to a function that uses received visual data to detect specific patterns or inappropriate content.

[0482] "A processing means for performing audio analysis" refers to a function that converts received audio data into text and analyzes its meaning and emotion based on that text data.

[0483] "Video analysis processing means" refers to a function that analyzes received video data frame by frame and performs image and audio analysis necessary to understand the content.

[0484] A "storage means for retaining past data" refers to a function that stores past cases and data in a database for future analysis and evaluation.

[0485] A "means for assessing risk" is a function that quantifies the potential risks posed by a post by comparing the analysis results from the processing method with past data.

[0486] "Means for providing predictions and suggestions" refers to a function that generates and provides recommendations when conveying the results of a risk assessment to the user as feedback.

[0487] This system analyzes the content of posts made by users on social networking services before they are submitted, in order to predict the risk of controversy and inappropriateness. The specific configuration and operation of the system are described below.

[0488] Users first send the text, images, audio, or video data they wish to post to the system via their device. The device then transmits this data to the server over the internet. This communication utilizes the HTTPS protocol, ensuring secure data transfer.

[0489] The server has the ability to immediately check received data and classify it by type based on its format. For example, text data is analyzed using natural language processing techniques. Specific software used includes Python's NLTK library and spaCy. By utilizing these, aggressive language and negative sentiments in text can be identified.

[0490] Next, the image data is subjected to image recognition using the machine learning framework TensorFlow. For example, a pre-trained model is used to detect images containing inappropriate symbols or content.

[0491] The audio data is converted to text using the Google Speech-to-Text API, and then sentiment analysis is performed on the text. This identifies aggressive remarks and emotional content contained in the audio.

[0492] The video data is analyzed frame by frame using OpenCV to detect inappropriate content from both image and audio elements.

[0493] The server uses these analysis results to compare with past database data and assess the potential risk of a post causing a backlash. This process calculates a risk score, which users can use to decide whether to revise or postpone their post.

[0494] For example, if a user tries to post a message saying, "This product is really terrible. You shouldn't use it!", the server can warn about this aggression through text analysis. Another example of a prompt for a generative AI model might be, "Please analyze whether the following content I plan to post on social media is offensive or at risk of causing controversy: 'This product is really terrible. You shouldn't use it!'"

[0495] This allows the system to provide users with appropriate feedback and prevent online controversies and misunderstandings. The system as a whole plays a crucial role in maintaining user trust and protecting the brand value of companies and individuals.

[0496] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0497] Step 1:

[0498] Users input text, images, audio, and video data through their devices and send that data. The input data is sent to the server as posted content. This process involves data transfer over the internet, and the HTTPS protocol is used to ensure security.

[0499] Step 2:

[0500] The server immediately analyzes received data by classifying it by type. Specifically, it identifies each data type as text data, visual data, audio data, or video data, and stores them in the appropriate directory or database. This classification allows subsequent processing steps to proceed more efficiently.

[0501] Step 3:

[0502] When the server receives text data, it performs text analysis using natural language processing techniques. The input is the received text data, and the output is the analysis results regarding the sentiment and aggression contained in the text. A Python library is used to segment the text and calculate the frequency of occurrence of specific keywords.

[0503] Step 4:

[0504] When the server receives visual data, it starts image analysis using an image recognition algorithm. The input is classified visual data, and the output is the analysis result identifying inappropriate content and symbols. Using TensorFlow or similar tools, a pre-trained model is applied to identify and evaluate objects within the image.

[0505] Step 5:

[0506] When the server receives audio data, it performs speech recognition processing to convert the audio data into text. The input is audio data, and the output is the converted text result. The Google Speech-to-Text API is used to convert the audio input into text data, and the result is then subjected to further text analysis.

[0507] Step 6:

[0508] When the server receives video data, it extracts all frames and applies image recognition technology. The input is video data, and the output is a report of inappropriateness for each frame obtained from the video. Using OpenCV, the video is decomposed into image frames, and the content of each frame is evaluated through image analysis.

[0509] Step 7:

[0510] Based on the analysis results, the server assesses the risk by comparing it with historical data stored in the database. Using the analyzed data as input, the server calculates the similarity to past online firestorm incidents and outputs the result as a risk score.

[0511] Step 8:

[0512] Based on the evaluation results, the server notifies the user of appropriate feedback. The input includes a risk score and recommended actions, and the output is a warning message the user receives. For example, if the risk is high, it will send a specific suggestion such as, "This post may be inappropriate. We recommend you make changes."

[0513] This series of steps creates a system that helps users mitigate risks before posting and supports safe and appropriate communication.

[0514] (Application Example 1)

[0515] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0516] In recent years, with the rapid dissemination of information in the real world, the spread of inappropriate content from individuals and businesses has increased. As a result, there have been numerous cases where the reputation of companies and individuals has been negatively affected. In particular, brick-and-mortar stores face the challenge of managing the risks associated with social media posts by staff and customers, as these posts spread instantly. It is necessary to prevent this from happening and make the flow of information in society safer and more reliable.

[0517] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0518] In this invention, the server includes a receiving means for receiving input from an information processing device, a classification means for classifying the received input into multiple formats, an analysis means for performing analysis based on the classified formats, a storage means for storing records of past risk cases, an evaluation means for comparing the results of the analysis means with the records in the storage means to evaluate the risk, a notification means for notifying the user in real time of loss predictions and recommendations based on the evaluation results, and a management means for managing information dissemination at real-world interaction locations and protecting reputation. This makes it possible to manage the risks of information dissemination to SNS even in physical stores, and to prevent problems caused by posted content.

[0519] An "information processing device" refers to a computer system that receives, transmits, and processes data, and is a device that receives user input in various formats.

[0520] A "receiving means" is a means that provides the function of receiving data from an external source to an information processing device, and plays the role of receiving various input data provided by the user.

[0521] A "classification method" is a means that provides a function to organize received data based on its attributes and format, and to divide it into appropriate categories.

[0522] "Analysis methods" refer to methods that use algorithms and techniques to apply advanced processing to classified data, understand its content, and analyze it.

[0523] A "memory device" is a means of storing past cases and data and providing a function to manage them so that they can be quickly searched and referenced.

[0524] An "evaluation tool" is a means that provides a function to predict and evaluate the likelihood of specific risks or problems by comparing analyzed data with stored data.

[0525] A "notification method" is a means that, based on the evaluation results, provides users with necessary information and communicates recommended actions and precautions.

[0526] "Management measures" refer to means for comprehensively monitoring and controlling the dissemination, distribution, and protection of information, and provide methods for appropriately managing information in the real world.

[0527] The server implements a system for managing information dissemination risks in physical stores. This system receives data transmitted from users' smart devices, classifies and analyzes it appropriately, compares it with past risk cases, and provides real-time feedback of the evaluation results.

[0528] On the smart device side, for example, a smartphone or tablet is used, and an application is installed on it. This application uses a program based on Python and TensorFlow as its execution environment and sends input such as text, images, and audio from the user to the server.

[0529] On the server, received data is automatically analyzed using natural language processing (NLP), image recognition, and speech recognition technologies. Specifically, for text, NLP is used to detect offensive or inappropriate expressions, and for images, a Convolutional Neural Network (CNN) is used to recognize inappropriate symbols and trademarks. Audio data is analyzed in a similar manner.

[0530] Next, the evaluation system compares the analysis results with a database of past risk cases and calculates a risk score. Based on this score, a notification is presented to the user's device. This allows the user to confirm the safety of immediate information dissemination within the store and to review or cancel the posted content as needed.

[0531] For example, if a customer who visited a cafe posts a review with a photo on social media, the app will detect if the photo contains a trademark and issue a warning. An example of a prompt to the generative AI model might be, "Please assess the risks of images containing specific brand names or trademarks."

[0532] In this way, this system can strengthen risk management for information dissemination in physical stores and provide a safe and secure communication environment for businesses and customers.

[0533] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0534] Step 1:

[0535] The device receives user input. During this process, the user inputs data such as text, images, and audio into the application via a smartphone or tablet. The input data is then prepared to be sent to the server.

[0536] Step 2:

[0537] The terminal sends the received data to the server. The data is transferred to the server via the network, where it is classified into an appropriate format for analysis. Input can include text, images, and audio.

[0538] Step 3:

[0539] The server receives the transmitted data and first organizes it into different formats using a classification mechanism. For example, text data is converted into string format, image data into pixel information, and audio data into a spectrum.

[0540] Step 4:

[0541] The server performs different analyses depending on the classified format. It uses NLP for text, CNN for images, and speech recognition technology for audio. For example, in text analysis, it performs grammatical analysis and keyword extraction to detect offensive expressions.

[0542] Step 5:

[0543] Based on the analysis results, the server compares them with a database of past risk cases. The evaluation system calculates a risk score, quantifying the level of risk. The risk score is determined based on the analysis results compared with the database.

[0544] Step 6:

[0545] The server sends the assessed risk score to the user's device via a notification system. The user receives the notification and decides whether to re-evaluate the posted content. An example of a prompt message generated by the AI ​​model is, "Please assess the risk of images containing specific brand names or trademarks."

[0546] Step 7:

[0547] Users will receive feedback from the server and, if necessary, revise their posts appropriately. If the re-evaluation reveals that a post contains problems, the user will edit the post to ensure safe information dissemination.

[0548] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0549] The present invention's system is designed to assess in advance the potential risk of online backlash caused by posts on social media, and further features analysis of the user's emotional state. When a user inputs the content they wish to post using a terminal, the terminal sends that data to a server. The server classifies the received data into text, audio, images, and videos, and applies an appropriate analysis process to each.

[0550] The server analyzes text data using natural language processing algorithms. Particular attention is paid to evaluating whether aggressive language or negative emotions are present. Image data is examined using image recognition technology to check for symbolic or inappropriate elements. Audio data is converted to text using speech recognition, and an emotion engine analyzes voice tone and intensity, allowing for a detailed examination of the emotional nuances of the posted content.

[0551] After these analyses are complete, the server integrates the results and calculates a crisis risk score by comparing them to past crisis cases. Sentiment analysis plays a significant role in this process, as the emotions the user is trying to convey have a major influence. The server then makes a loss prediction based on this and notifies the user as appropriate feedback. At this point, the user can obtain specific information to consider reviewing or deleting their post.

[0552] For example, if a user attempts to post a long text message containing many emotionally charged words, the server will perform a sentiment analysis of the text. If it determines that negative emotions are dominant, it will warn the user that there is a high risk of the message causing a social media firestorm. Similarly, in the analysis of audio data, if strong emotions such as anger or sadness are expressed, the server will alert the user to help them make appropriate decisions. In this way, the present invention provides a means to realize safer and healthier use of social media.

[0553] The following describes the processing flow.

[0554] Step 1:

[0555] Users input text, images, audio, and video data they want to post to social media into the system via their device and upload it. The device then sends this data to the server.

[0556] Step 2:

[0557] The server receives data from the terminal and classifies it into text, image, audio, and video formats. Based on this classification, the server initiates an analysis process appropriate to each format.

[0558] Step 3:

[0559] The server applies natural language processing algorithms to the text data. It extracts keywords and performs sentiment analysis to identify inappropriate expressions and negative emotions.

[0560] Step 4:

[0561] The server uses image recognition technology to detect inappropriate symbols and recognizable objects in the image data, paying particular attention to elements that may cause offense.

[0562] Step 5:

[0563] The server converts the audio data into text using speech recognition technology, and then analyzes that text using an emotion engine. It analyzes the tone and intensity of the voice and evaluates the underlying emotions.

[0564] Step 6:

[0565] For video data, the server extracts frames and performs image recognition on them. It also performs audio analysis, including the video's sound. This allows for the evaluation of the emotional impact of the combined visual and auditory elements.

[0566] Step 7:

[0567] After all analyses are complete, the server integrates the results. This allows for comparison and analysis with past online controversies, and calculates a controversy risk score for the posted content.

[0568] Step 8:

[0569] Based on the analysis results and risk assessment obtained by the server, the server sends specific feedback to the user. The user reviews the posted content based on the information provided and considers modifying or canceling the post if necessary.

[0570] (Example 2)

[0571] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0572] The challenge lies in proactively evaluating the potential risks of online backlash and negative emotional impacts of content posted by users on social media, thereby promoting safe and healthy communication. In today's information society, new methods are needed to mitigate the risks caused by inappropriate or emotionally charged posts.

[0573] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0574] In this invention, the server includes means for receiving user input from an information processing device, means for classifying the received input into multiple data formats, and means for performing text analysis using a natural language processing algorithm based on the classified text data. This enables appropriate analysis according to each data format posted by the user, allows for appropriate feedback after understanding the risks in advance, and supports safe communication.

[0575] An "information processing device" is an electronic device that has the functions of receiving, processing, and outputting data.

[0576] A "user" refers to a person or entity that operates the system and enters data.

[0577] "Receiving means" refers to the protocol or interface for receiving data transmitted from an information processing device.

[0578] "Data format" refers to the structure or type of information that is represented, and includes text, audio, images, and video.

[0579] A "classification method" refers to the process or mechanism of sorting received data according to its format.

[0580] "Means of performing text analysis" refers to the process of analyzing the content of text data using natural language processing techniques.

[0581] "Image recognition technology" refers to algorithms and systems used to extract and analyze specific information from image data.

[0582] "Speech recognition technology" is a technology that converts speech data into text data and analyzes its characteristics and emotions.

[0583] "Emotion inference" is the process of evaluating the type and intensity of emotions from analyzed data.

[0584] A "storage medium" is a device or medium that stores data and keeps it accessible at a later date.

[0585] "Means of risk assessment" refers to the process of quantifying or qualitatively evaluating the degree of potential risk based on the analysis results.

[0586] "Means of generating notifications" refers to a system that creates and presents messages and warnings to convey information to users.

[0587] This invention is a system that assesses the risks of content posted by users to social networking services (SNS) via an information processing device, thereby promoting safe communication. First, the user inputs text, audio, image, or video data they wish to post using a terminal. The terminal then transmits this input data to a server.

[0588] The server classifies the received data according to its format and applies the appropriate analysis process to each. Text data is analyzed using natural language processing libraries. Specifically, techniques such as extracting emotions and aggressive expressions from text are used, employing Python's NLTK and spaCy.

[0589] The audio data is converted to text on the server using speech recognition technology. The Google Speech-to-Text API is used, and then IBM Watson is utilized to analyze the tone and emotion of the voice. This allows for the evaluation of the emotional nuances of the audio content.

[0590] Image data is analyzed using TensorFlow and common image recognition techniques to check for symbolic or inappropriate elements. For example, provocative gestures or inappropriate objects can be automatically detected.

[0591] The analysis results are integrated, and the server generates a crisis risk score by comparing it to a database of past cases. Finally, based on this, a loss prediction is made and communicated to the user as specific feedback. This allows users to understand the risks of their posts in advance and take appropriate action.

[0592] For example, if a user attempts to post emotional text such as "This product is completely useless!", the server will analyze the text and issue a warning if negative sentiment is detected. Another example of a prompt message is, "Evaluate the degree of negative sentiment based on the content the user is attempting to post and generate an appropriate risk warning."

[0593] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0594] Step 1:

[0595] The user enters the content they want to post to social media using their device. The entered data can be in text, audio, image, or video format. Specifically, the user enters the content to be posted into an input form on their smartphone or computer and presses the submit button. The entered data is then ready to be sent to the server.

[0596] Step 2:

[0597] The terminal sends input data to the server. The data is in the form of an HTTP request sent to the server over the internet. Once the input data is transferred to the server, a series of analysis preparation processes begin.

[0598] Step 3:

[0599] The server classifies the received data. Specifically, it analyzes metadata to identify the data format and categorizes it into text, audio, image, and video. The classified data is then passed to the appropriate analysis process according to its format.

[0600] Step 4:

[0601] The server analyzes text data using natural language processing algorithms. The input is text data, and the output is the analysis result showing its emotional characteristics. Specifically, the server uses Python's natural language processing library to analyze and evaluate aggressive expressions and negative emotions from the text data.

[0602] Step 5:

[0603] The server converts audio data into text and then performs sentiment analysis. The input is audio data, and the output is the result of evaluating the emotions contained in the audio. Specifically, it uses speech recognition technology to convert the audio into text, and then uses an emotion engine to analyze the tone and emotions of the voice.

[0604] Step 6:

[0605] The server analyzes image data using image recognition technology. The input is image data, and the output is an evaluation based on the image content. Specifically, it uses an image recognition algorithm to detect symbolic or inappropriate elements within the image and generates results.

[0606] Step 7:

[0607] The server integrates the results of each analysis and calculates a crisis risk score. The input is the results of each analysis, and the output is an overall risk assessment. Specifically, the server compares the analysis results with past case data and objectively evaluates the risks involved.

[0608] Step 8:

[0609] The server generates notifications based on the evaluation results and provides feedback to the user. The input is the risk score, and the output is the feedback notification. Specifically, the server generates a warning message according to the risk level and sends it to the user. This gives the user an opportunity to review their posts.

[0610] (Application Example 2)

[0611] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0612] In online advertising, deploying ads without predicting consumer emotions and reactions can lead to backlash and negative responses, potentially harming a company's brand image and sales. Therefore, it is essential to evaluate how the content will be received by consumers and the level of backlash risk before the ad is delivered, and to take preventative measures.

[0613] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0614] In this invention, the server includes receiving means for receiving user input from an information processing device, classification means for classifying the received input into multiple formats, and advertising analysis means for evaluating in advance how advertising content will affect the user's emotions and providing feedback. This makes it possible to predict consumer reactions before advertising is delivered and reduce the risk of backlash.

[0615] An "information processing device" is a computer or server used to receive and analyze data.

[0616] A "receiving means" is an element that performs the function of collecting data provided by the user and incorporating it into the system.

[0617] A "classification means" is an element that has the function of identifying and sorting received data into specific formats such as text, images, audio, and video.

[0618] An "analysis tool" is an element that has the function of analyzing data of each classified format and evaluating its content.

[0619] A "memory device" is a medium or device for accumulating and storing past online firestorm incidents that are compared with the analyzed data.

[0620] "Evaluation methods" refer to a mechanism that assesses the likelihood of a social media firestorm by comparing the analysis results with past data.

[0621] A "notification method" is an element that has a method or technique for providing feedback on evaluation results to the user.

[0622] "Advertising analysis tools" are elements that have the function of evaluating the emotional impact that advertising content has on consumers and predicting the risk of backlash.

[0623] To realize this invention, a server, an information processing device, a user terminal, and a communication network connecting them are necessary. The server uses software such as Python, TensorFlow, NLTK, and spaCy to analyze the received data. Specifically, the server acquires advertising content received from the user terminal via a receiving means and classifies it into different formats. Using a classification means, it separates the content into text, image, audio, and video formats, and an analysis means performs an analysis appropriate to each format.

[0624] For text data, natural language processing algorithms are applied to extract offensive words and negative expressions. Image data is analyzed using image recognition technology to attempt to detect inappropriate elements and symbolic content. Audio data is converted to text using speech recognition technology, and then an emotion engine evaluates the emotional nuances. Similar processing is performed on video data.

[0625] The analysis results are compared with past online controversies stored in memory, and an evaluation tool calculates the risk of a controversy. This score is then provided to the user visually or audibly through a notification tool. Additionally, an advertising analysis tool evaluates the emotional impact of the advertisement on consumers and provides feedback. In this process, a generative AI model assists by generating or suggesting revisions to the advertisement based on prompt text.

[0626] As a concrete example, a user uploads an ad video to a server before launching a new advertising campaign. The prompt might be something like, "Analyze the new campaign ad video and predict its social impact." The server analyzes this and returns feedback such as, "There is a high probability of a negative reaction," allowing the user to gain specific guidance for improving the campaign.

[0627] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0628] Step 1:

[0629] The user inputs advertising content using their device and sends it to the server. The input data can take the form of text, images, audio, or video in digital format. The server receives this input and begins processing it within the system.

[0630] Step 2:

[0631] The server classifies received content into four categories—text, images, audio, and video—using classification mechanisms. This classification is necessary to apply an analysis process appropriate to the data format. For example, the data format is identified by determining the file extension.

[0632] Step 3:

[0633] The server uses natural language processing algorithms to extract aggressive expressions and negative emotions from text data. The input is raw text data, and the output is a set of analyzed emotion information. This process is performed using a natural language processing library.

[0634] Step 4:

[0635] Image data is analyzed using image recognition technology to detect symbolic or inappropriate elements. The input is image data, and the output is the analysis result as metadata. An image processing library is used for this process.

[0636] Step 5:

[0637] The audio data is converted to text by a speech recognition engine, and then sentiment analysis is performed, similar to that for text. The input is an audio file, and the output is the corresponding text along with its sentiment information. Speech recognition is performed using speech analysis software.

[0638] Step 6:

[0639] The video data is analyzed using a video analysis tool to extract still images from each frame, and then the image recognition process from step 4 is applied to these still images. The input is a video file, and the output is the analysis results for each frame.

[0640] Step 7:

[0641] The server compares the analysis results with past online firestorm incidents stored in a memory device and calculates a firestorm risk score using an evaluation device. The input is the analysis results for each data format, and the output is the firestorm risk score. A historical database is used for comparison.

[0642] Step 8:

[0643] Through a notification system, users are provided with a risk score for online controversy and feedback based on that score. The input is the score, and the output is text and warning messages displayed on the user's device. This allows users to review and modify their advertising content.

[0644] This series of processes provides users with guidelines for generating or modifying ad content based on prompts from the generation AI model.

[0645] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0646] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0647] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0648] [Fourth Embodiment]

[0649] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0650] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0651] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0652] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0653] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0654] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0655] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0656] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0657] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0658] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0659] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0660] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0661] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0662] The system of this invention is designed to analyze the content of a post and predict its risks before the user posts it on social media. The user transmits text, images, audio, and video data they wish to post to the system via a terminal. This data is received by a server and classified according to its type. Based on the received data, the server applies natural language processing technology to the text, image recognition technology to the images, and speech recognition technology to the audio, performing analysis accordingly.

[0663] For example, if a user attempts to post offensive content, the server analyzes the post's text and compares it to a database of past online controversies. If similar problematic expressions are detected, the server determines that the risk of online backlash is high. Furthermore, if a posted image contains inappropriate symbols, the server analyzes them and issues a warning to the user.

[0664] Furthermore, the server integrates these analysis results and calculates the expected losses caused by the posted content. These losses include potential economic losses, including legal compensation. Based on the evaluation results, the server provides feedback to the user and suggests reviewing or canceling the post. This allows users to prevent inappropriate posts on social media. Through this process, the system provides an effective means of maintaining the credibility of companies and individuals and reducing risks.

[0665] The following describes the processing flow.

[0666] Step 1:

[0667] Users prepare text, images, audio, and video data they want to post to social media using their devices and upload them to the system. The devices then send this data to the server.

[0668] Step 2:

[0669] The server receives data sent from the terminal. The received data is classified into text data, image data, audio data, and video data.

[0670] Step 3:

[0671] The server applies natural language processing algorithms to the text data to analyze its content. This process identifies whether the data contains offensive language or discriminatory words.

[0672] Step 4:

[0673] The server uses image recognition technology to analyze the image data. It analyzes the poses of objects and people in the image and evaluates whether any inappropriate elements are included.

[0674] Step 5:

[0675] The server converts the audio data into text using speech recognition technology. The converted text is then checked again using natural language processing to ensure it does not contain any inappropriate content.

[0676] Step 6:

[0677] The server extracts frames from the video data and analyzes them using image recognition technology. Furthermore, audio within the video is analyzed separately.

[0678] Step 7:

[0679] The server integrates all analysis results and evaluates whether the information matches past online firestorm incidents. It calculates a firestorm risk score and estimates the expected amount of loss.

[0680] Step 8:

[0681] The server notifies the user of the evaluation results. This includes feedback detailing the risks and possible mitigation strategies. The user then uses this information to consider whether to revise their post.

[0682] (Example 1)

[0683] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0684] There is a growing need to prevent inappropriate posts on social networking services and other platforms. However, it is currently difficult for users to accurately judge the risk of their posts causing controversy. This invention aims to reduce inappropriate posts by automatically analyzing the content of posts, evaluating the risks, and providing feedback before users post them.

[0685] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0686] In this invention, the server includes means for receiving user information, means for classifying the received information into data formats, and means for processing based on the classified data formats. This makes it possible to analyze the risk of user posts becoming controversial in advance and provide appropriate feedback.

[0687] "Means for receiving user information" refers to the function by which the server receives text data, visual data, audio data, and video data sent from the terminal by the user.

[0688] "Means of classifying data into data formats" refers to the function of classifying received information into text data, visual data, audio data, and video data according to its content.

[0689] "Processing means for performing text analysis" refers to a function that analyzes received text data based on natural language processing technology to understand its content and meaning.

[0690] "Image analysis processing means" refers to a function that uses received visual data to detect specific patterns or inappropriate content.

[0691] "A processing means for performing audio analysis" refers to a function that converts received audio data into text and analyzes its meaning and emotion based on that text data.

[0692] "Video analysis processing means" refers to a function that analyzes received video data frame by frame and performs image and audio analysis necessary to understand the content.

[0693] A "storage means for retaining past data" refers to a function that stores past cases and data in a database for future analysis and evaluation.

[0694] A "means for assessing risk" is a function that quantifies the potential risks posed by a post by comparing the analysis results from the processing method with past data.

[0695] "Means for providing predictions and suggestions" refers to a function that generates and provides recommendations when conveying the results of a risk assessment to the user as feedback.

[0696] This system analyzes the content of posts made by users on social networking services before they are submitted, in order to predict the risk of controversy and inappropriateness. The specific configuration and operation of the system are described below.

[0697] Users first send the text, images, audio, or video data they wish to post to the system via their device. The device then transmits this data to the server over the internet. This communication utilizes the HTTPS protocol, ensuring secure data transfer.

[0698] The server has the ability to immediately check received data and classify it by type based on its format. For example, text data is analyzed using natural language processing techniques. Specific software used includes Python's NLTK library and spaCy. By utilizing these, aggressive language and negative sentiments in text can be identified.

[0699] Next, the image data is subjected to image recognition using the machine learning framework TensorFlow. For example, a pre-trained model is used to detect images containing inappropriate symbols or content.

[0700] The audio data is converted to text using the Google Speech-to-Text API, and then sentiment analysis is performed on the text. This identifies aggressive remarks and emotional content contained in the audio.

[0701] The video data is analyzed frame by frame using OpenCV to detect inappropriate content from both image and audio elements.

[0702] The server uses these analysis results to compare with past database data and assess the potential risk of a post causing a backlash. This process calculates a risk score, which users can use to decide whether to revise or postpone their post.

[0703] For example, if a user tries to post a message saying, "This product is really terrible. You shouldn't use it!", the server can warn about this aggression through text analysis. Another example of a prompt for a generative AI model might be, "Please analyze whether the following content I plan to post on social media is offensive or at risk of causing controversy: 'This product is really terrible. You shouldn't use it!'"

[0704] This allows the system to provide users with appropriate feedback and prevent online controversies and misunderstandings. The system as a whole plays a crucial role in maintaining user trust and protecting the brand value of companies and individuals.

[0705] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0706] Step 1:

[0707] Users input text, images, audio, and video data through their devices and send that data. The input data is sent to the server as posted content. This process involves data transfer over the internet, and the HTTPS protocol is used to ensure security.

[0708] Step 2:

[0709] The server immediately analyzes received data by classifying it by type. Specifically, it identifies each data type as text data, visual data, audio data, or video data, and stores them in the appropriate directory or database. This classification allows subsequent processing steps to proceed more efficiently.

[0710] Step 3:

[0711] When the server receives text data, it performs text analysis using natural language processing techniques. The input is the received text data, and the output is the analysis results regarding the sentiment and aggression contained in the text. A Python library is used to segment the text and calculate the frequency of occurrence of specific keywords.

[0712] Step 4:

[0713] When the server receives visual data, it starts image analysis using an image recognition algorithm. The input is classified visual data, and the output is the analysis result identifying inappropriate content and symbols. Using TensorFlow or similar tools, a pre-trained model is applied to identify and evaluate objects within the image.

[0714] Step 5:

[0715] When the server receives audio data, it performs speech recognition processing to convert the audio data into text. The input is audio data, and the output is the converted text result. The Google Speech-to-Text API is used to convert the audio input into text data, and the result is then subjected to further text analysis.

[0716] Step 6:

[0717] When the server receives video data, it extracts all frames and applies image recognition technology. The input is video data, and the output is a report of inappropriateness for each frame obtained from the video. Using OpenCV, the video is decomposed into image frames, and the content of each frame is evaluated through image analysis.

[0718] Step 7:

[0719] Based on the analysis results, the server assesses the risk by comparing it with historical data stored in the database. Using the analyzed data as input, the server calculates the similarity to past online firestorm incidents and outputs the result as a risk score.

[0720] Step 8:

[0721] Based on the evaluation results, the server notifies the user of appropriate feedback. The input includes a risk score and recommended actions, and the output is a warning message the user receives. For example, if the risk is high, it will send a specific suggestion such as, "This post may be inappropriate. We recommend you make changes."

[0722] This series of steps creates a system that helps users mitigate risks before posting and supports safe and appropriate communication.

[0723] (Application Example 1)

[0724] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0725] In recent years, with the rapid dissemination of information in the real world, the spread of inappropriate content from individuals and businesses has increased. As a result, there have been numerous cases where the reputation of companies and individuals has been negatively affected. In particular, brick-and-mortar stores face the challenge of managing the risks associated with social media posts by staff and customers, as these posts spread instantly. It is necessary to prevent this from happening and make the flow of information in society safer and more reliable.

[0726] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0727] In this invention, the server includes a receiving means for receiving input from an information processing device, a classification means for classifying the received input into multiple formats, an analysis means for performing analysis based on the classified formats, a storage means for storing records of past risk cases, an evaluation means for comparing the results of the analysis means with the records in the storage means to evaluate the risk, a notification means for notifying the user in real time of loss predictions and recommendations based on the evaluation results, and a management means for managing information dissemination at real-world interaction locations and protecting reputation. This makes it possible to manage the risks of information dissemination to SNS even in physical stores, and to prevent problems caused by posted content.

[0728] An "information processing device" refers to a computer system that receives, transmits, and processes data, and is a device that receives user input in various formats.

[0729] A "receiving means" is a means that provides the function of receiving data from an external source to an information processing device, and plays the role of receiving various input data provided by the user.

[0730] A "classification method" is a means that provides a function to organize received data based on its attributes and format, and to divide it into appropriate categories.

[0731] "Analysis methods" refer to methods that use algorithms and techniques to apply advanced processing to classified data, understand its content, and analyze it.

[0732] A "memory device" is a means of storing past cases and data and providing a function to manage them so that they can be quickly searched and referenced.

[0733] An "evaluation tool" is a means that provides a function to predict and evaluate the likelihood of specific risks or problems by comparing analyzed data with stored data.

[0734] A "notification method" is a means that, based on the evaluation results, provides users with necessary information and communicates recommended actions and precautions.

[0735] "Management measures" refer to means for comprehensively monitoring and controlling the dissemination, distribution, and protection of information, and provide methods for appropriately managing information in the real world.

[0736] The server implements a system for managing information dissemination risks in physical stores. This system receives data transmitted from users' smart devices, classifies and analyzes it appropriately, compares it with past risk cases, and provides real-time feedback of the evaluation results.

[0737] On the smart device side, for example, a smartphone or tablet is used, and an application is installed on it. This application uses a program based on Python and TensorFlow as its execution environment and sends input such as text, images, and audio from the user to the server.

[0738] On the server, received data is automatically analyzed using natural language processing (NLP), image recognition, and speech recognition technologies. Specifically, for text, NLP is used to detect offensive or inappropriate expressions, and for images, a Convolutional Neural Network (CNN) is used to recognize inappropriate symbols and trademarks. Audio data is analyzed in a similar manner.

[0739] Next, the evaluation system compares the analysis results with a database of past risk cases and calculates a risk score. Based on this score, a notification is presented to the user's device. This allows the user to confirm the safety of immediate information dissemination within the store and to review or cancel the posted content as needed.

[0740] For example, if a customer who visited a cafe posts a review with a photo on social media, the app will detect if the photo contains a trademark and issue a warning. An example of a prompt to the generative AI model might be, "Please assess the risks of images containing specific brand names or trademarks."

[0741] In this way, this system can strengthen risk management for information dissemination in physical stores and provide a safe and secure communication environment for businesses and customers.

[0742] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0743] Step 1:

[0744] The device receives user input. During this process, the user inputs data such as text, images, and audio into the application via a smartphone or tablet. The input data is then prepared to be sent to the server.

[0745] Step 2:

[0746] The terminal sends the received data to the server. The data is transferred to the server via the network, where it is classified into an appropriate format for analysis. Input can include text, images, and audio.

[0747] Step 3:

[0748] The server receives the transmitted data and first organizes it into different formats using a classification mechanism. For example, text data is converted into string format, image data into pixel information, and audio data into a spectrum.

[0749] Step 4:

[0750] The server performs different analyses depending on the classified format. It uses NLP for text, CNN for images, and speech recognition technology for audio. For example, in text analysis, it performs grammatical analysis and keyword extraction to detect offensive expressions.

[0751] Step 5:

[0752] Based on the analysis results, the server compares them with a database of past risk cases. The evaluation system calculates a risk score, quantifying the level of risk. The risk score is determined based on the analysis results compared with the database.

[0753] Step 6:

[0754] The server sends the assessed risk score to the user's device via a notification system. The user receives the notification and decides whether to re-evaluate the posted content. An example of a prompt message generated by the AI ​​model is, "Please assess the risk of images containing specific brand names or trademarks."

[0755] Step 7:

[0756] Users will receive feedback from the server and, if necessary, revise their posts appropriately. If the re-evaluation reveals that a post contains problems, the user will edit the post to ensure safe information dissemination.

[0757] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0758] The present invention's system is designed to assess in advance the potential risk of online backlash caused by posts on social media, and further features analysis of the user's emotional state. When a user inputs the content they wish to post using a terminal, the terminal sends that data to a server. The server classifies the received data into text, audio, images, and videos, and applies an appropriate analysis process to each.

[0759] The server analyzes text data using natural language processing algorithms. Particular attention is paid to evaluating whether aggressive language or negative emotions are present. Image data is examined using image recognition technology to check for symbolic or inappropriate elements. Audio data is converted to text using speech recognition, and an emotion engine analyzes voice tone and intensity, allowing for a detailed examination of the emotional nuances of the posted content.

[0760] After these analyses are complete, the server integrates the results and calculates a crisis risk score by comparing them to past crisis cases. Sentiment analysis plays a significant role in this process, as the emotions the user is trying to convey have a major influence. The server then makes a loss prediction based on this and notifies the user as appropriate feedback. At this point, the user can obtain specific information to consider reviewing or deleting their post.

[0761] For example, if a user attempts to post a long text message containing many emotionally charged words, the server will perform a sentiment analysis of the text. If it determines that negative emotions are dominant, it will warn the user that there is a high risk of the message causing a social media firestorm. Similarly, in the analysis of audio data, if strong emotions such as anger or sadness are expressed, the server will alert the user to help them make appropriate decisions. In this way, the present invention provides a means to realize safer and healthier use of social media.

[0762] The following describes the processing flow.

[0763] Step 1:

[0764] Users input text, images, audio, and video data they want to post to social media into the system via their device and upload it. The device then sends this data to the server.

[0765] Step 2:

[0766] The server receives data from the terminal and classifies it into text, image, audio, and video formats. Based on this classification, the server initiates an analysis process appropriate to each format.

[0767] Step 3:

[0768] The server applies natural language processing algorithms to the text data. It extracts keywords and performs sentiment analysis to identify inappropriate expressions and negative emotions.

[0769] Step 4:

[0770] The server uses image recognition technology to detect inappropriate symbols and recognizable objects in the image data, paying particular attention to elements that may cause offense.

[0771] Step 5:

[0772] The server converts the audio data into text using speech recognition technology, and then analyzes that text using an emotion engine. It analyzes the tone and intensity of the voice and evaluates the underlying emotions.

[0773] Step 6:

[0774] For video data, the server extracts frames and performs image recognition on them. It also performs audio analysis, including the video's sound. This allows for the evaluation of the emotional impact of the combined visual and auditory elements.

[0775] Step 7:

[0776] After all analyses are complete, the server integrates the results. This allows for comparison and analysis with past online controversies, and calculates a controversy risk score for the posted content.

[0777] Step 8:

[0778] Based on the analysis results and risk assessment obtained by the server, the server sends specific feedback to the user. The user reviews the posted content based on the information provided and considers modifying or canceling the post if necessary.

[0779] (Example 2)

[0780] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0781] The challenge lies in proactively evaluating the potential risks of online backlash and negative emotional impacts of content posted by users on social media, thereby promoting safe and healthy communication. In today's information society, new methods are needed to mitigate the risks caused by inappropriate or emotionally charged posts.

[0782] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0783] In this invention, the server includes means for receiving user input from an information processing device, means for classifying the received input into multiple data formats, and means for performing text analysis using a natural language processing algorithm based on the classified text data. This enables appropriate analysis according to each data format posted by the user, allows for appropriate feedback after understanding the risks in advance, and supports safe communication.

[0784] An "information processing device" is an electronic device that has the functions of receiving, processing, and outputting data.

[0785] A "user" refers to a person or entity that operates the system and enters data.

[0786] "Receiving means" refers to the protocol or interface for receiving data transmitted from an information processing device.

[0787] "Data format" refers to the structure or type of information that is represented, and includes text, audio, images, and video.

[0788] A "classification method" refers to the process or mechanism of sorting received data according to its format.

[0789] "Means of performing text analysis" refers to the process of analyzing the content of text data using natural language processing techniques.

[0790] "Image recognition technology" refers to algorithms and systems used to extract and analyze specific information from image data.

[0791] "Speech recognition technology" is a technology that converts speech data into text data and analyzes its characteristics and emotions.

[0792] "Emotion inference" is the process of evaluating the type and intensity of emotions from analyzed data.

[0793] A "storage medium" is a device or medium that stores data and keeps it accessible at a later date.

[0794] "Means of risk assessment" refers to the process of quantifying or qualitatively evaluating the degree of potential risk based on the analysis results.

[0795] "Means of generating notifications" refers to a system that creates and presents messages and warnings to convey information to users.

[0796] This invention is a system that assesses the risks of content posted by users to social networking services (SNS) via an information processing device, thereby promoting safe communication. First, the user inputs text, audio, image, or video data they wish to post using a terminal. The terminal then transmits this input data to a server.

[0797] The server classifies the received data according to its format and applies the appropriate analysis process to each. Text data is analyzed using natural language processing libraries. Specifically, techniques such as extracting emotions and aggressive expressions from text are used, employing Python's NLTK and spaCy.

[0798] The audio data is converted to text on the server using speech recognition technology. The Google Speech-to-Text API is used, and then IBM Watson is utilized to analyze the tone and emotion of the voice. This allows for the evaluation of the emotional nuances of the audio content.

[0799] Image data is analyzed using TensorFlow and common image recognition techniques to check for symbolic or inappropriate elements. For example, provocative gestures or inappropriate objects can be automatically detected.

[0800] The analysis results are integrated, and the server generates a crisis risk score by comparing it to a database of past cases. Finally, based on this, a loss prediction is made and communicated to the user as specific feedback. This allows users to understand the risks of their posts in advance and take appropriate action.

[0801] For example, if a user attempts to post emotional text such as "This product is completely useless!", the server will analyze the text and issue a warning if negative sentiment is detected. Another example of a prompt message is, "Evaluate the degree of negative sentiment based on the content the user is attempting to post and generate an appropriate risk warning."

[0802] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0803] Step 1:

[0804] The user enters the content they want to post to social media using their device. The entered data can be in text, audio, image, or video format. Specifically, the user enters the content to be posted into an input form on their smartphone or computer and presses the submit button. The entered data is then ready to be sent to the server.

[0805] Step 2:

[0806] The terminal sends input data to the server. The data is in the form of an HTTP request sent to the server over the internet. Once the input data is transferred to the server, a series of analysis preparation processes begin.

[0807] Step 3:

[0808] The server classifies the received data. Specifically, it analyzes metadata to identify the data format and categorizes it into text, audio, image, and video. The classified data is then passed to the appropriate analysis process according to its format.

[0809] Step 4:

[0810] The server analyzes text data using natural language processing algorithms. The input is text data, and the output is the analysis result showing its emotional characteristics. Specifically, the server uses Python's natural language processing library to analyze and evaluate aggressive expressions and negative emotions from the text data.

[0811] Step 5:

[0812] The server converts audio data into text and then performs sentiment analysis. The input is audio data, and the output is the result of evaluating the emotions contained in the audio. Specifically, it uses speech recognition technology to convert the audio into text, and then uses an emotion engine to analyze the tone and emotions of the voice.

[0813] Step 6:

[0814] The server analyzes image data using image recognition technology. The input is image data, and the output is an evaluation based on the image content. Specifically, it uses an image recognition algorithm to detect symbolic or inappropriate elements within the image and generates results.

[0815] Step 7:

[0816] The server integrates the results of each analysis and calculates a crisis risk score. The input is the results of each analysis, and the output is an overall risk assessment. Specifically, the server compares the analysis results with past case data and objectively evaluates the risks involved.

[0817] Step 8:

[0818] The server generates notifications based on the evaluation results and provides feedback to the user. The input is the risk score, and the output is the feedback notification. Specifically, the server generates a warning message according to the risk level and sends it to the user. This gives the user an opportunity to review their posts.

[0819] (Application Example 2)

[0820] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0821] In online advertising, deploying ads without predicting consumer emotions and reactions can lead to backlash and negative responses, potentially harming a company's brand image and sales. Therefore, it is essential to evaluate how the content will be received by consumers and the level of backlash risk before the ad is delivered, and to take preventative measures.

[0822] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0823] In this invention, the server includes receiving means for receiving user input from an information processing device, classification means for classifying the received input into multiple formats, and advertising analysis means for evaluating in advance how advertising content will affect the user's emotions and providing feedback. This makes it possible to predict consumer reactions before advertising is delivered and reduce the risk of backlash.

[0824] An "information processing device" is a computer or server used to receive and analyze data.

[0825] A "receiving means" is an element that performs the function of collecting data provided by the user and incorporating it into the system.

[0826] A "classification means" is an element that has the function of identifying and sorting received data into specific formats such as text, images, audio, and video.

[0827] An "analysis tool" is an element that has the function of analyzing data of each classified format and evaluating its content.

[0828] A "memory device" is a medium or device for accumulating and storing past online firestorm incidents that are compared with the analyzed data.

[0829] "Evaluation methods" refer to a mechanism that assesses the likelihood of a social media firestorm by comparing the analysis results with past data.

[0830] A "notification method" is an element that has a method or technique for providing feedback on evaluation results to the user.

[0831] "Advertising analysis tools" are elements that have the function of evaluating the emotional impact that advertising content has on consumers and predicting the risk of backlash.

[0832] To realize this invention, a server, an information processing device, a user terminal, and a communication network connecting them are necessary. The server uses software such as Python, TensorFlow, NLTK, and spaCy to analyze the received data. Specifically, the server acquires advertising content received from the user terminal via a receiving means and classifies it into different formats. Using a classification means, it separates the content into text, image, audio, and video formats, and an analysis means performs an analysis appropriate to each format.

[0833] For text data, natural language processing algorithms are applied to extract offensive words and negative expressions. Image data is analyzed using image recognition technology to attempt to detect inappropriate elements and symbolic content. Audio data is converted to text using speech recognition technology, and then an emotion engine evaluates the emotional nuances. Similar processing is performed on video data.

[0834] The analysis results are compared with past online controversies stored in memory, and an evaluation tool calculates the risk of a controversy. This score is then provided to the user visually or audibly through a notification tool. Additionally, an advertising analysis tool evaluates the emotional impact of the advertisement on consumers and provides feedback. In this process, a generative AI model assists by generating or suggesting revisions to the advertisement based on prompt text.

[0835] As a concrete example, a user uploads an ad video to a server before launching a new advertising campaign. The prompt might be something like, "Analyze the new campaign ad video and predict its social impact." The server analyzes this and returns feedback such as, "There is a high probability of a negative reaction," allowing the user to gain specific guidance for improving the campaign.

[0836] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0837] Step 1:

[0838] The user inputs advertising content using their device and sends it to the server. The input data can take the form of text, images, audio, or video in digital format. The server receives this input and begins processing it within the system.

[0839] Step 2:

[0840] The server classifies received content into four categories—text, images, audio, and video—using classification mechanisms. This classification is necessary to apply an analysis process appropriate to the data format. For example, the data format is identified by determining the file extension.

[0841] Step 3:

[0842] The server uses natural language processing algorithms to extract aggressive expressions and negative emotions from text data. The input is raw text data, and the output is a set of analyzed emotion information. This process is performed using a natural language processing library.

[0843] Step 4:

[0844] Image data is analyzed using image recognition technology to detect symbolic or inappropriate elements. The input is image data, and the output is the analysis result as metadata. An image processing library is used for this process.

[0845] Step 5:

[0846] The audio data is converted to text by a speech recognition engine, and then sentiment analysis is performed, similar to that for text. The input is an audio file, and the output is the corresponding text along with its sentiment information. Speech recognition is performed using speech analysis software.

[0847] Step 6:

[0848] The video data is analyzed using a video analysis tool to extract still images from each frame, and then the image recognition process from step 4 is applied to these still images. The input is a video file, and the output is the analysis results for each frame.

[0849] Step 7:

[0850] The server compares the analysis results with past online firestorm incidents stored in a memory device and calculates a firestorm risk score using an evaluation device. The input is the analysis results for each data format, and the output is the firestorm risk score. A historical database is used for comparison.

[0851] Step 8:

[0852] Through a notification system, users are provided with a risk score for online controversy and feedback based on that score. The input is the score, and the output is text and warning messages displayed on the user's device. This allows users to review and modify their advertising content.

[0853] This series of processes provides users with guidelines for generating or modifying ad content based on prompts from the generation AI model.

[0854] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0855] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0856] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0857] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0858] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0859] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0860] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0861] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0862] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0863] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0864] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0865] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0866] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0867] 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.

[0868] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0869] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0870] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0871] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0872] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0873] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0874] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0875] The following is further disclosed regarding the embodiments described above.

[0876] (Claim 1)

[0877] A receiving means for receiving user input from an information processing device,

[0878] A classification means for classifying the received input into multiple formats,

[0879] An analysis means for performing text analysis based on the aforementioned classified format,

[0880] An analysis means for performing video analysis based on the aforementioned classified format,

[0881] An analysis means for performing speech analysis based on the aforementioned classified format,

[0882] A means of storing records of past online controversies,

[0883] An evaluation means for evaluating the risk of fire by comparing the results obtained by the analysis means with the records of the storage means,

[0884] A notification means that outputs loss predictions and recommendations based on the evaluation results,

[0885] A system that includes this.

[0886] (Claim 2)

[0887] The system according to claim 1, wherein the receiving means receives the input in at least one of the following formats: text data, image data, audio data, and video data.

[0888] (Claim 3)

[0889] The system according to claim 1, wherein the evaluation means calculates a firestorm risk score according to the similarity of past firestorm incidents, and the notification means presents the score to the user visually or audibly.

[0890] "Example 1"

[0891] (Claim 1)

[0892] Means for receiving user information,

[0893] A means for classifying the received information into a data format,

[0894] A processing means for performing text analysis based on the classified data format,

[0895] A processing means for performing image analysis based on the classified data format,

[0896] A processing means for performing speech analysis based on the classified data format,

[0897] A processing means for performing video analysis based on the classified data format,

[0898] A storage means for retaining past data,

[0899] A means for comparing the results obtained by the processing means with the data in the storage means and evaluating the risk,

[0900] A means for providing predictions and proposals based on the aforementioned evaluation results,

[0901] A system that includes this.

[0902] (Claim 2)

[0903] The system according to claim 1, wherein the receiving means receives information in at least one of the following formats: character data, visual data, audio data, and video data.

[0904] (Claim 3)

[0905] The system according to claim 1, wherein the evaluation means calculates a risk score according to the similarity of past data, and the providing means displays or audibly communicates the score to the user.

[0906] "Application Example 1"

[0907] (Claim 1)

[0908] A receiving means for receiving input from an information processing device,

[0909] A classification means for classifying the received input into multiple formats,

[0910] An analysis means for performing text analysis based on the aforementioned classified format,

[0911] An analysis means for performing video analysis based on the aforementioned classified format,

[0912] An analysis means for performing speech analysis based on the aforementioned classified format,

[0913] A storage means for holding records of past risk cases,

[0914] An evaluation means that evaluates the risk by comparing the results obtained by the analysis means with the records of the storage means,

[0915] Based on the aforementioned evaluation results, a notification means outputs loss predictions and recommendations and notifies the user in real time.

[0916] A management system to control information dissemination in real-world interaction spaces and protect reputation,

[0917] A system that includes this.

[0918] (Claim 2)

[0919] The system according to claim 1, wherein the receiving means receives the input as a plurality of data formats, and the notification means includes instructions for re-evaluation.

[0920] (Claim 3)

[0921] The system according to claim 1, wherein the evaluation means calculates a risk score according to the similarity of past risk cases, and the notification means presents the score to the user and proposes reviewing or canceling the information dissemination.

[0922] "Example 2 of combining an emotion engine"

[0923] (Claim 1)

[0924] A means for receiving user input from an information processing device,

[0925] The means for classifying the received input into multiple data formats,

[0926] A means for performing text analysis using a natural language processing algorithm based on the classified text data,

[0927] A means for performing video analysis using image recognition technology based on the classified image data,

[0928] Based on the classified audio data, an analysis means for performing speech recognition and emotion inference is provided.

[0929] A method using a storage medium that holds records of past cases,

[0930] A means for performing a risk assessment by comparing the results obtained by the analysis means with the records on the storage medium,

[0931] A means for generating a notification that outputs recommendations based on the evaluation results,

[0932] A system that includes this.

[0933] (Claim 2)

[0934] The system according to claim 1, wherein the receiving means receives the input as at least one of the data formats of text, image, audio, and video.

[0935] (Claim 3)

[0936] The system according to claim 1, wherein the evaluation means calculates a risk score according to the similarity of past cases, and the notification means presents the score to the user in multiple formats.

[0937] "Application example 2 of combining emotional engines"

[0938] (Claim 1)

[0939] A receiving means for receiving user input from an information processing device,

[0940] A classification means for classifying the received input into multiple formats,

[0941] An analysis means for performing text analysis based on the aforementioned classified format,

[0942] An analysis means for performing video analysis based on the aforementioned classified format,

[0943] An analysis means for performing speech analysis based on the aforementioned classified format,

[0944] A means of storing records of past online controversies,

[0945] An evaluation means for evaluating the risk of fire by comparing the results obtained by the analysis means with the records of the storage means,

[0946] A notification means that outputs loss predictions and recommendations based on the evaluation results,

[0947] An advertising analytics tool that evaluates in advance how advertising content will affect users' emotions and provides feedback,

[0948] A system that includes this.

[0949] (Claim 2)

[0950] The system according to claim 1, wherein the receiving means receives the input in at least one of the following formats: text data, image data, audio data, and video data.

[0951] (Claim 3)

[0952] The system according to claim 1, wherein the evaluation means calculates a firestorm risk score according to the similarity of past firestorm incidents, and the notification means presents the score to the user visually or audibly. [Explanation of Symbols]

[0953] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A receiving means for receiving user input from an information processing device, A classification means for classifying the received input into multiple formats, An analysis means for performing text analysis based on the aforementioned classified format, An analysis means for performing video analysis based on the aforementioned classified format, An analysis means for performing speech analysis based on the aforementioned classified format, A means of storing records of past online controversies, An evaluation means for evaluating the risk of fire by comparing the results obtained by the analysis means with the records of the storage means, A notification means that outputs loss predictions and recommendations based on the evaluation results, A system that includes this.

2. The system according to claim 1, wherein the receiving means receives the input in at least one format from text data, image data, audio data, and video data.

3. The system according to claim 1, wherein the evaluation means calculates a firestorm risk score according to the similarity of past firestorm incidents, and the notification means presents the score to the user visually or audibly.

Images