system

Abstract

The system according to the embodiment aims to automatically detect and appropriately handle abusive comments in the comment section of a social networking site. [Solution] A system according to an embodiment includes a reception unit, an analysis unit, a detection unit, and a processing unit. The reception unit receives comments. The analysis unit analyzes the comments received by the reception unit. The detection unit detects abusive comments based on the comments analyzed by the analysis unit. The processing unit deletes or converts the abusive comments detected by the detection unit.

Description

[Technical Field]

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

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

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

[0004] Conventional technologies do not adequately detect and appropriately process abusive comments in social networking site comment sections, and there is room for improvement.

[0005] The system according to the embodiment aims to automatically detect and appropriately handle abusive comments in the comment section of a social networking site. [Means for solving the problem]

[0006] The system according to the embodiment includes a reception unit, an analysis unit, a detection unit, and a processing unit. The reception unit receives comments. The analysis unit analyzes the comments received by the reception unit. The detection unit detects abusive comments based on the comments analyzed by the analysis unit. The processing unit deletes or converts the abusive comments detected by the detection unit. [Effects of the Invention]

[0007] The system according to the embodiment can automatically detect abusive comments in the comment section of a social networking site and deal with them appropriately. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE INVENTION

[0009] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), and Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) An SNS comment management system according to an embodiment of the present invention automatically detects abusive comments in SNS comment sections and either deletes them or converts them to a more ladylike tone. When a comment is posted, the SNS comment management system uses AI to analyze the content and determine whether it is a defamatory comment. For example, if a comment contains offensive words such as "idiot" or "die," the AI ​​determines it to be a defamatory comment. If a comment is determined to be a defamatory comment, the AI ​​automatically deletes the comment or converts it to a more ladylike tone. For example, if a comment using the word "idiot" is posted, the AI ​​converts it to a more ladylike tone, such as "you idiot." This conversion is performed based on conversion rules learned by the AI ​​in advance. This makes SNS comment sections a place for healthy communication. Users can post comments with peace of mind, without being bothered by abusive comments. Furthermore, deleting comments prevents the abusive comments from negatively impacting other users. For example, if a user comments "You're an idiot," the AI ​​will convert it to "You're an idiot." This softens the tone of the comment and reduces its aggression toward other users. This allows the SNS comment management system to automatically detect and delete or convert abusive comments in SNS comment sections, providing a healthy forum for communication.

[0029] An SNS comment management system according to an embodiment includes a reception unit, an analysis unit, a detection unit, and a processing unit. The reception unit receives comments. Examples of comments include, but are not limited to, text comments, image comments, and audio comments. The reception unit receives comments, for example, through a posting interface of the SNS. The reception unit can also receive comments from an external system via an API. For example, the reception unit receives comments posted to a comment section of the SNS in real time. When a user posts a comment by voice, the reception unit can convert the voice comment into text and accept the comment using voice recognition technology. The analysis unit analyzes the comments received by the reception unit using natural language processing technology. For example, the analysis unit divides the words in the comment using morphological analysis and performs grammatical analysis. The analysis unit can also understand the content of the comment using semantic analysis. For example, the analysis unit understands the context of the comment and identifies offensive words and phrases. The analysis unit can also analyze the sentiment of the comment using natural language processing technology. The detection unit detects abusive comments based on the comments analyzed by the analysis unit. The detection unit detects abusive comments using, for example, a machine learning algorithm. The detection unit can also use a rule-based algorithm to detect whether a specific word or phrase is included. For example, the detection unit references a predefined list of abusive comments and detects whether a comment matches the list. The processing unit deletes or converts the abusive comments detected by the detection unit. For example, the processing unit can completely delete the abusive comments. The processing unit can also hide the abusive comments. Furthermore, the processing unit can convert the abusive comments into a ladylike tone. For example, the processing unit converts offensive words and phrases into softer expressions based on pre-trained conversion rules. As a result, the SNS comment management system according to the embodiment can automatically detect and delete or convert abusive comments in the comment section of an SNS, thereby providing a forum for healthy communication.

[0030] The analysis unit can analyze the content of the comment using natural language processing technology. Natural language processing technology includes, but is not limited to, morphological analysis, grammatical analysis, and semantic analysis. For example, the analysis unit can use morphological analysis to divide the words in the comment and identify the part of speech of each word. For example, morphological analysis divides each word in the comment and identifies which part of speech each word belongs to, such as a noun, verb, or adjective. The analysis unit can also analyze the sentence structure of the comment using grammatical analysis. For example, grammatical analysis analyzes the order and relationships of words in the comment to understand the sentence structure. Furthermore, the analysis unit can understand the content of the comment using semantic analysis. For example, semantic analysis analyzes the meaning of words and phrases in the comment to understand the meaning of the comment as a whole. This allows the analysis unit to accurately analyze the content of the comment using natural language processing technology. Some or all of the above-mentioned processing in the analysis unit can be performed using, for example, a generation AI, or can be performed without using a generation AI. For example, the analysis unit can input the comment into a generation AI, which then analyzes the content of the comment. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. This allows the analysis unit to accurately analyze the content of the comments by using natural language processing technology.

[0031] The detection unit may use an algorithm for detecting abusive comments. Examples of the algorithm include, but are not limited to, machine learning algorithms and rule-based algorithms. The detection unit may use, for example, a machine learning algorithm to detect abusive comments. For example, the machine learning algorithm learns from a large amount of comment data and extracts characteristics of abusive comments. This allows the detection unit to detect with high accuracy whether a newly posted comment is abusive. The detection unit may also use a rule-based algorithm to detect abusive comments. For example, the rule-based algorithm references a predefined list of abusive comments and detects whether a comment matches the list. This allows the detection unit to quickly detect whether a specific word or phrase is included. Furthermore, the detection unit may also detect abusive comments by combining a machine learning algorithm and a rule-based algorithm. For example, the machine learning algorithm extracts characteristics of abusive comments, and the rule-based algorithm detects specific words or phrases. This allows the detection unit to accurately detect abusive comments. Some or all of the above-described processing in the detection unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the detection unit inputs comments into a generation AI, which then detects abusive comments. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. This allows the detection unit to accurately detect abusive comments.

[0032] The processing unit can delete the abusive comment. Deletion includes, but is not limited to, methods such as complete deletion and hiding. For example, the processing unit completely deletes the abusive comment. For example, the processing unit completely deletes the abusive comment from the database, preventing other users from viewing the comment. The processing unit can also hide the abusive comment. For example, the processing unit sets the abusive comment to be hidden so that only specific users can view the comment. This allows the processing unit to prevent the deletion of the abusive comment from having a negative impact on other users. Some or all of the above-described processing in the processing unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the processing unit can input the abusive comment into a generation AI, which then deletes the comment. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. This allows the processing unit to prevent the deletion of defamatory comments from having a negative impact on other users.

[0033] The processing unit can use conversion rules to convert abusive comments into a ladylike tone. Examples of conversion rules include, but are not limited to, rules for converting offensive words or phrases into softer expressions. For example, the processing unit converts abusive comments into a ladylike tone. For example, the processing unit converts a comment such as "idiot" into "idiot." The processing unit can also convert a comment such as "die" into "please die." By converting abusive comments into a ladylike tone, the processing unit can soften the tone of the comment. Some or all of the above-described processing in the processing unit may be performed using, or without, a generation AI. For example, the processing unit can input abusive comments into a generation AI, which then converts the comments into a ladylike tone. Examples of generation AI include, but are not limited to, a text generation AI (e.g., LLM) and a multimodal generation AI. This allows the processing unit to soften the tone of the comment by converting the slanderous comment into a ladylike tone.

[0034] The reception unit can analyze a user's past comment history and select the optimal reception method. For example, if a user has posted offensive comments in the past, the reception unit displays a warning message before accepting the comment. For example, if a user has a history of posting offensive comments in the past, the reception unit displays a warning message saying, "Please refrain from posting offensive comments" before accepting the comment. The reception unit can also quickly accept comments if a user has posted constructive comments in the past. For example, the reception unit prioritizes accepting comments if the user has a history of posting constructive comments in the past. Furthermore, the reception unit can analyze a user's interest in a specific topic from the user's past comment history and prioritize accepting comments on related topics. For example, if a user has frequently posted comments on a specific topic in the past, the reception unit prioritizes accepting comments related to that topic. This allows the reception unit to select the optimal reception method based on the user's past comment history, enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's past comment history data into the generation AI, which can then select the optimal reception method. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0035] When accepting comments, the acceptance unit can filter the comments based on the user's current activity status and areas of interest. For example, the acceptance unit accepts only relevant comments based on the content of the page the user is currently viewing. For example, if the user is viewing a specific news article, the acceptance unit accepts only comments related to the news article. The acceptance unit can also analyze the user's areas of interest from the user's past browsing history and preferentially accept related comments. For example, if the user has frequently viewed articles on a specific topic in the past, the acceptance unit preferentially accepts comments related to the topic. Furthermore, if the user is participating in a specific event, the acceptance unit can preferentially accept comments related to the event. For example, if the user is participating in a specific event, the acceptance unit preferentially accepts comments related to the event. In this way, the acceptance unit can preferentially accept highly relevant comments by filtering comments based on the user's current activity status and areas of interest. Some or all of the above-described processing in the acceptance unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's activity status and interest field data to the generation AI, which can then filter the comments. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0036] When accepting comments, the reception unit can prioritize accepting highly relevant comments by taking into account the user's geographical location information. For example, if the user is in a specific region, the reception unit prioritizes accepting comments related to that region. For example, if the user is in a specific city, the reception unit prioritizes accepting comments related to that city. Furthermore, if the user is traveling, the reception unit can prioritize accepting comments related to the travel destination. For example, if the reception unit estimates that the user is traveling, the reception unit prioritizes accepting comments related to the travel destination. Furthermore, if the user is at an event venue, the reception unit can prioritize accepting comments related to the event. For example, if the user is at a specific event venue, the reception unit prioritizes accepting comments related to the event. This allows the reception unit to prioritize accepting highly relevant comments based on the user's geographical location information, thereby enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, or without, a generation AI. For example, the reception unit can input the user's geographical location information data into the generation AI, which can then filter the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0037] The reception unit can analyze the user's social media activity when receiving comments and receive relevant comments. For example, if a user frequently uses a specific hashtag, the reception unit can prioritize receiving comments related to that hashtag. For example, if a user frequently uses a specific hashtag, the reception unit can prioritize receiving comments related to that hashtag. Furthermore, if a user belongs to a specific group, the reception unit can prioritize receiving comments related to that group. For example, if a user belongs to a specific group, the reception unit can prioritize receiving comments related to that group. Furthermore, if a user frequently posts about a specific topic, the reception unit can prioritize receiving comments related to that topic. For example, if a user frequently posts about a specific topic, the reception unit can prioritize receiving comments related to that topic. This allows the reception unit to receive relevant comments based on the user's social media activity, thereby enabling appropriate comment reception. Some or all of the above-described processing in the reception unit can be performed using, or without, a generation AI. For example, the reception unit can input the user's social media activity data into the generation AI, which can then filter the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0038] During analysis, the analysis unit can adjust the level of detail of the analysis based on the importance of the comment. For example, the analysis unit performs a detailed analysis on comments with high importance. For example, the analysis unit performs a detailed analysis on comments with high importance to gain a deeper understanding of the content of the comment. The analysis unit can also perform a simplified analysis on comments with low importance. For example, the analysis unit can perform a simplified analysis on comments with low importance to quickly provide results. Furthermore, the analysis unit can adjust the display order of the analysis results based on the importance of the comment. For example, the analysis unit can prioritize displaying the analysis results of comments with high importance and postpone displaying the analysis results of comments with low importance. In this way, the analysis unit can provide appropriate analysis results by adjusting the level of detail of the analysis based on the importance of the comment. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment importance data to the generation AI, which can then adjust the level of detail of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0039] During analysis, the analysis unit can apply different analysis algorithms depending on the category of the comment. For example, the analysis unit applies an analysis algorithm that emphasizes specific political keywords to comments related to politics. For example, the analysis unit applies an analysis algorithm that emphasizes specific political keywords to comments related to politics, thereby gaining a deeper understanding of the content of the comment. The analysis unit can also apply an analysis algorithm that emphasizes sports terms to comments related to sports. For example, the analysis unit applies an analysis algorithm that emphasizes sports terms to comments related to sports, thereby gaining a deeper understanding of the content of the comment. The analysis unit can also apply an analysis algorithm that emphasizes entertainment-related keywords to comments related to entertainment. For example, the analysis unit applies an analysis algorithm that emphasizes entertainment-related keywords to comments related to entertainment, thereby gaining a deeper understanding of the content of the comment. This allows the analysis unit to provide accurate analysis results by applying an appropriate analysis algorithm depending on the category of the comment. Some or all of the above-mentioned processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment category data into the generation AI, which then applies an appropriate analysis algorithm. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0040] During analysis, the analysis unit can determine the analysis priority based on the time when the comments were posted. For example, the analysis unit can prioritize the analysis of the most recent comments and display the results in real time. For example, the analysis unit can prioritize the analysis of the most recent comments and display the results in real time, thereby providing quick feedback to the user. The analysis unit can also determine the analysis priority for past comments based on their importance. For example, the analysis unit can prioritize the analysis of past comments based on their importance and prioritize analyzing important comments. Furthermore, the analysis unit can prioritize the analysis of comments posted during a specific event period. For example, the analysis unit can prioritize the analysis of comments posted during a specific event period and quickly provide information related to the event. In this way, the analysis unit can provide appropriate analysis results by determining the analysis priority based on the time when the comments were posted. Some or all of the above-described processing in the analysis unit can be performed using, or without, a generation AI. For example, the analysis unit can input comment posting time data to the generation AI, which can then determine the analysis priority. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0041] During analysis, the analysis unit can adjust the order of analysis based on the relevance of the comments. For example, the analysis unit prioritizes analysis of highly relevant comments and displays the results. For example, the analysis unit prioritizes analysis of highly relevant comments and displays the results, thereby quickly providing important information to the user. The analysis unit can also perform a simplified analysis of less relevant comments. For example, the analysis unit can perform a simplified analysis of less relevant comments and quickly provide the results. Furthermore, the analysis unit can adjust the display order of the analysis results based on the relevance of the comments. For example, the analysis unit prioritizes displaying the analysis results of highly relevant comments and postpones the analysis results of less relevant comments. In this way, the analysis unit can provide appropriate analysis results by adjusting the analysis order based on the relevance of the comments. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment relevance data into the generation AI, which can then adjust the analysis order. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0042] The detection unit can improve the accuracy of detection by taking into account the interrelationships between comments during detection. The detection unit, for example, analyzes the context of comments to detect abusive comments. For example, the detection unit analyzes the context of comments to detect abusive comments, thereby performing highly accurate detection that takes context into account. The detection unit can also analyze consecutive comments by the same user to detect abusive comments. For example, the detection unit analyzes consecutive comments by the same user to detect abusive comments, thereby accurately understanding the user's intention. Furthermore, the detection unit can also analyze interactions between multiple users to detect abusive comments. For example, the detection unit analyzes interactions between multiple users to detect abusive comments, thereby performing highly accurate detection that takes into account the flow of the dialogue. As a result, the detection unit improves the accuracy of detecting abusive comments by taking into account the interrelationships between comments. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input comment interrelationship data into the generation AI, which may then improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0043] The detection unit can perform detection by taking into account attribute information of the poster of the comment. The detection unit adjusts the detection criteria for defamatory comments based on, for example, the poster's age group. For example, the detection unit adjusts the detection criteria for defamatory comments based on the poster's age group and takes appropriate measures according to the poster's age. The detection unit can also analyze the poster's past comment history to detect defamatory comments. For example, the detection unit analyzes the poster's past comment history to detect defamatory comments, thereby performing highly accurate detection that takes into account the poster's behavioral patterns. Furthermore, the detection unit can also detect defamatory comments by taking into account the poster's regional information. For example, the detection unit takes into account the poster's regional information and performs detection that takes into account words and expressions unique to the region. As a result, the detection unit improves the accuracy of detecting defamatory comments by taking into account the attribute information of the comment poster. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input attribute information data of the poster into the generation AI, which may improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0044] The detection unit can perform detection by taking into account the geographical distribution of comments. For example, the detection unit prioritizes detecting abusive comments that occur frequently in a specific region. For example, the detection unit prioritizes detecting abusive comments that occur frequently in a specific region and responds quickly to issues specific to that region. The detection unit can also detect abusive comments by taking into account cultural and linguistic differences between regions. For example, the detection unit takes into account cultural and linguistic differences between regions and appropriately detects words and expressions specific to the region. Furthermore, the detection unit can also detect abusive comments by analyzing interactions between geographically close users. For example, the detection unit analyzes interactions between geographically close users and detects abusive comments, thereby responding quickly to issues within the local community. In this way, the detection unit can improve the accuracy of detecting abusive comments specific to a region by taking into account the geographical distribution of comments. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input geographic distribution data of comments into the generation AI, which may improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0045] The detection unit can improve the accuracy of detection by referring to literature related to the comment during detection. The detection unit, for example, refers to related research papers to improve the accuracy of detecting defamatory comments. For example, the detection unit refers to related research papers to improve the accuracy of detecting defamatory comments based on the latest research results. The detection unit can also improve the accuracy of detecting defamatory comments by referring to related news articles. For example, the detection unit can detect defamatory comments that reflect the latest social conditions by referring to related news articles. Furthermore, the detection unit can also improve the accuracy of detecting defamatory comments by referring to related books. For example, the detection unit can detect defamatory comments taking into account historical background and cultural context by referring to related books. In this way, the detection unit can improve the accuracy of detecting defamatory comments by referring to literature related to the comment. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input related literature data of the comment into the generation AI, which may improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0046] During processing, the processing unit can improve the accuracy of processing by taking into account the interrelationships between comments. For example, the processing unit analyzes the context of comments and deletes abusive comments. For example, the processing unit analyzes the context of comments and deletes abusive comments, thereby performing highly accurate processing that takes context into account. The processing unit can also analyze consecutive comments from the same user and delete abusive comments. For example, the processing unit analyzes consecutive comments from the same user and deletes abusive comments, thereby accurately understanding the user's intention. Furthermore, the processing unit can analyze interactions between multiple users and delete abusive comments. For example, the processing unit analyzes interactions between multiple users and deletes abusive comments, thereby performing highly accurate processing that takes into account the flow of the dialogue. As a result, the processing unit takes into account the interrelationships between comments, thereby improving the accuracy of processing abusive comments. Some or all of the above-mentioned processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit can input comment interrelationship data to a generation AI, which can improve the accuracy of processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0047] The processing unit can perform processing while taking into account attribute information of the comment poster. The processing unit, for example, adjusts the processing method for abusive comments based on the poster's age group. For example, the processing unit adjusts the processing method for abusive comments based on the poster's age group and takes appropriate action according to the poster's age. The processing unit can also analyze the poster's past comment history and delete or convert abusive comments. For example, the processing unit analyzes the poster's past comment history and deletes or converts abusive comments, thereby performing highly accurate processing that takes into account the poster's behavioral patterns. Furthermore, the processing unit can also delete or convert abusive comments while taking into account the poster's regional information. For example, the processing unit takes into account the poster's regional information and performs processing that takes into account words and expressions specific to the region. As a result, the processing unit takes into account the attribute information of the comment poster, thereby improving the accuracy of processing abusive comments. Some or all of the above-described processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit inputs attribute information data of the poster into the generation AI, which can improve the accuracy of the processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0048] The processing unit may take into account the geographical distribution of comments during processing. For example, the processing unit may prioritize deleting abusive comments that occur frequently in a specific region. For example, the processing unit may prioritize deleting abusive comments that occur frequently in a specific region, thereby quickly addressing issues specific to that region. The processing unit may also delete or convert abusive comments taking into account cultural and linguistic differences between regions. For example, the processing unit may appropriately process words and expressions specific to the region by taking into account cultural and linguistic differences between regions. Furthermore, the processing unit may analyze interactions between geographically close users and delete or convert abusive comments. For example, the processing unit may analyze interactions between geographically close users and delete or convert abusive comments, thereby quickly addressing issues within the local community. In this way, by taking into account the geographical distribution of comments, the processing unit improves the accuracy of processing abusive comments specific to the region. Some or all of the above-described processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit can input geographical distribution data of comments to the generation AI, which can improve the accuracy of processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0049] During processing, the processing unit can improve the accuracy of processing by referring to literature related to the comment. The processing unit, for example, refers to related research papers to improve the accuracy of processing defamatory comments. For example, the processing unit refers to related research papers to improve the accuracy of processing defamatory comments based on the latest research results. The processing unit can also refer to related news articles to improve the accuracy of processing defamatory comments. For example, the processing unit refers to related news articles to process defamatory comments that reflect the latest social conditions. Furthermore, the processing unit can also refer to related books to improve the accuracy of processing defamatory comments. For example, the processing unit refers to related books to process defamatory comments taking into account historical background and cultural context. In this way, by referring to literature related to the comment, the processing unit improves the accuracy of processing defamatory comments. Some or all of the above-mentioned processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit can input the related literature data of the comment into the generation AI, which can improve the accuracy of the processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0050] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0051] The reception unit can analyze a user's past comment history and select the optimal reception method. For example, if a user has posted offensive comments in the past, the reception unit displays a warning message before accepting the comment. For example, if a user has a history of posting offensive comments in the past, the reception unit displays a warning message saying, "Please refrain from posting offensive comments" before accepting the comment. The reception unit can also quickly accept comments if a user has posted constructive comments in the past. For example, the reception unit prioritizes accepting comments if the user has a history of posting constructive comments in the past. Furthermore, the reception unit can analyze a user's interest in a specific topic from the user's past comment history and prioritize accepting comments on related topics. For example, if a user has frequently posted comments on a specific topic in the past, the reception unit prioritizes accepting comments related to that topic. This allows the reception unit to select the optimal reception method based on the user's past comment history, enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's past comment history data into the generation AI, which can then select the optimal reception method. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0052] When accepting comments, the acceptance unit can filter the comments based on the user's current activity status and areas of interest. For example, the acceptance unit accepts only relevant comments based on the content of the page the user is currently viewing. For example, if the user is viewing a specific news article, the acceptance unit accepts only comments related to that news article. The acceptance unit can also analyze the user's areas of interest from the user's past browsing history and preferentially accept related comments. For example, if the user has frequently viewed articles on a specific topic in the past, the acceptance unit preferentially accepts comments related to that topic. Furthermore, if the user is participating in a specific event, the acceptance unit can preferentially accept comments related to that event. For example, if the user is participating in a specific event, the acceptance unit preferentially accepts comments related to that event. In this way, the acceptance unit can preferentially accept highly relevant comments by filtering comments based on the user's current activity status and areas of interest. Some or all of the above-described processing by the acceptance unit may be performed using, or without, a generation AI. For example, the acceptance unit can input the user's activity status and area of ​​interest data into the generation AI, which then filters the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0053] During analysis, the analysis unit can adjust the level of detail of the analysis based on the importance of the comment. For example, the analysis unit performs a detailed analysis on comments with high importance. For example, the analysis unit performs a detailed analysis on comments with high importance to gain a deeper understanding of the content of the comment. The analysis unit can also perform a simplified analysis on comments with low importance. For example, the analysis unit can perform a simplified analysis on comments with low importance to quickly provide results. Furthermore, the analysis unit can adjust the display order of the analysis results based on the importance of the comment. For example, the analysis unit can prioritize displaying the analysis results of comments with high importance and postpone displaying the analysis results of comments with low importance. In this way, the analysis unit can provide appropriate analysis results by adjusting the level of detail of the analysis based on the importance of the comment. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment importance data to the generation AI, which can then adjust the level of detail of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0054] The detection unit can improve the accuracy of detection by taking into account the interrelationships between comments. For example, the detection unit analyzes the context of comments to detect abusive comments. For example, the detection unit analyzes the context of comments to detect abusive comments, thereby performing highly accurate detection that takes context into account. The detection unit can also analyze consecutive comments from the same user to detect abusive comments. For example, the detection unit analyzes consecutive comments from the same user to detect abusive comments, thereby accurately understanding the user's intention. Furthermore, the detection unit can also analyze interactions between multiple users to detect abusive comments. For example, the detection unit analyzes interactions between multiple users to detect abusive comments, thereby performing highly accurate detection that takes into account the flow of the dialogue. As a result, the detection unit improves the accuracy of detecting abusive comments by taking into account the interrelationships between comments. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input comment interrelationship data into the generation AI, which may then improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0055] During processing, the processing unit can improve the accuracy of processing by taking into account the interrelationships between comments. For example, the processing unit analyzes the context of comments and deletes abusive comments. For example, the processing unit analyzes the context of comments and deletes abusive comments, thereby performing highly accurate processing that takes context into account. The processing unit can also analyze consecutive comments from the same user and delete abusive comments. For example, the processing unit analyzes consecutive comments from the same user and deletes abusive comments, thereby accurately understanding the user's intention. Furthermore, the processing unit can analyze interactions between multiple users and delete abusive comments. For example, the processing unit analyzes interactions between multiple users and deletes abusive comments, thereby performing highly accurate processing that takes into account the flow of the dialogue. As a result, the processing unit improves the accuracy of processing abusive comments by taking into account the interrelationships between comments. Some or all of the above-mentioned processing in the processing unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the processing unit can input comment interrelationship data to a generation AI, which can improve the accuracy of processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0056] When accepting comments, the reception unit can prioritize accepting highly relevant comments by taking into account the user's geographical location information. For example, if the user is in a specific region, it can prioritize accepting comments related to that region. For example, if the user is in a specific city, it can prioritize accepting comments related to that city. Furthermore, if the user is traveling, the reception unit can prioritize accepting comments related to the travel destination. For example, if the reception unit estimates that the user is traveling, it can prioritize accepting comments related to the travel destination. Furthermore, if the user is at an event venue, the reception unit can prioritize accepting comments related to the event. For example, if the reception unit detects that the user is at a specific event venue, it can prioritize accepting comments related to the event. This allows the reception unit to prioritize accepting highly relevant comments based on the user's geographical location information, thereby enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, or without, a generation AI. For example, the reception unit can input the user's geographical location information data into the generation AI, which can then filter the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0057] The processing flow of the first embodiment will be briefly explained below.

[0058] Step 1: The reception unit receives comments. Comments include text comments, image comments, and audio comments. The reception unit can receive comments from external systems via the SNS posting interface or API. For example, it can receive comments posted in the SNS comment section in real time. It can also convert audio comments into text using voice recognition technology and receive them. Step 2: The analysis unit uses natural language processing technology to analyze the comments received by the reception unit. For example, it uses morphological analysis to split the words in the comments and perform grammatical analysis. It also uses semantic analysis to understand the content of the comments and identify offensive words and phrases. It can also analyze the sentiment of the comments. Step 3: The detection unit detects abusive comments based on the comments analyzed by the analysis unit. For example, it uses machine learning algorithms or rule-based algorithms to detect whether specific words or phrases are included. It can also refer to a predefined list of abusive comments and detect whether the comment matches that list. Step 4: The processing unit deletes or converts the abusive comments detected by the detection unit. For example, the processing unit can completely delete the abusive comments, hide them, or convert them to a more ladylike tone. The processing unit converts offensive words and phrases into softer expressions based on pre-trained conversion rules.

[0059] (Example 2) An SNS comment management system according to an embodiment of the present invention automatically detects abusive comments in SNS comment sections and either deletes them or converts them to a more ladylike tone. When a comment is posted, the SNS comment management system uses AI to analyze the content and determine whether it is a defamatory comment. For example, if a comment contains offensive words such as "idiot" or "die," the AI ​​determines it to be a defamatory comment. If a comment is determined to be a defamatory comment, the AI ​​automatically deletes the comment or converts it to a more ladylike tone. For example, if a comment using the word "idiot" is posted, the AI ​​converts it to a more ladylike tone, such as "you idiot." This conversion is performed based on conversion rules learned by the AI ​​in advance. This makes SNS comment sections a place for healthy communication. Users can post comments with peace of mind, without being bothered by abusive comments. Furthermore, deleting comments prevents the abusive comments from negatively impacting other users. For example, if a user comments "You're an idiot," the AI ​​will convert it to "You're an idiot." This softens the tone of the comment and reduces its aggression toward other users. This allows the SNS comment management system to automatically detect and delete or convert abusive comments in SNS comment sections, providing a healthy forum for communication.

[0060] An SNS comment management system according to an embodiment includes a reception unit, an analysis unit, a detection unit, and a processing unit. The reception unit receives comments. Examples of comments include, but are not limited to, text comments, image comments, and audio comments. The reception unit receives comments, for example, through a posting interface of the SNS. The reception unit can also receive comments from an external system via an API. For example, the reception unit receives comments posted to a comment section of the SNS in real time. When a user posts a comment by voice, the reception unit can convert the voice comment into text and accept the comment using voice recognition technology. The analysis unit analyzes the comments received by the reception unit using natural language processing technology. For example, the analysis unit divides the words in the comment using morphological analysis and performs grammatical analysis. The analysis unit can also understand the content of the comment using semantic analysis. For example, the analysis unit understands the context of the comment and identifies offensive words and phrases. The analysis unit can also analyze the sentiment of the comment using natural language processing technology. The detection unit detects abusive comments based on the comments analyzed by the analysis unit. The detection unit detects abusive comments using, for example, a machine learning algorithm. The detection unit can also use a rule-based algorithm to detect whether a specific word or phrase is included. For example, the detection unit references a predefined list of abusive comments and detects whether a comment matches the list. The processing unit deletes or converts the abusive comments detected by the detection unit. For example, the processing unit can completely delete the abusive comments. The processing unit can also hide the abusive comments. Furthermore, the processing unit can convert the abusive comments into a ladylike tone. For example, the processing unit converts offensive words and phrases into softer expressions based on pre-trained conversion rules. As a result, the SNS comment management system according to the embodiment can automatically detect and delete or convert abusive comments in the comment section of an SNS, thereby providing a forum for healthy communication.

[0061] The analysis unit can analyze the content of the comment using natural language processing technology. Natural language processing technology includes, but is not limited to, morphological analysis, grammatical analysis, and semantic analysis. For example, the analysis unit can use morphological analysis to divide the words in the comment and identify the part of speech of each word. For example, morphological analysis divides each word in the comment and identifies which part of speech each word belongs to, such as a noun, verb, or adjective. The analysis unit can also analyze the sentence structure of the comment using grammatical analysis. For example, grammatical analysis analyzes the order and relationships of words in the comment to understand the sentence structure. Furthermore, the analysis unit can understand the content of the comment using semantic analysis. For example, semantic analysis analyzes the meaning of words and phrases in the comment to understand the meaning of the comment as a whole. This allows the analysis unit to accurately analyze the content of the comment using natural language processing technology. Some or all of the above-mentioned processing in the analysis unit can be performed using, for example, a generation AI, or can be performed without using a generation AI. For example, the analysis unit can input the comment into a generation AI, which then analyzes the content of the comment. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. This allows the analysis unit to accurately analyze the content of the comments by using natural language processing technology.

[0062] The detection unit may use an algorithm for detecting abusive comments. Examples of the algorithm include, but are not limited to, machine learning algorithms and rule-based algorithms. The detection unit may use, for example, a machine learning algorithm to detect abusive comments. For example, the machine learning algorithm learns from a large amount of comment data and extracts characteristics of abusive comments. This allows the detection unit to detect with high accuracy whether a newly posted comment is abusive. The detection unit may also use a rule-based algorithm to detect abusive comments. For example, the rule-based algorithm references a predefined list of abusive comments and detects whether a comment matches the list. This allows the detection unit to quickly detect whether a specific word or phrase is included. Furthermore, the detection unit may also detect abusive comments by combining a machine learning algorithm and a rule-based algorithm. For example, the machine learning algorithm extracts characteristics of abusive comments, and the rule-based algorithm detects specific words or phrases. This allows the detection unit to accurately detect abusive comments. Some or all of the above-described processing in the detection unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the detection unit inputs comments into a generation AI, which then detects abusive comments. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. This allows the detection unit to accurately detect abusive comments.

[0063] The processing unit can delete the abusive comment. Deletion includes, but is not limited to, methods such as complete deletion and hiding. For example, the processing unit completely deletes the abusive comment. For example, the processing unit completely deletes the abusive comment from the database, preventing other users from viewing the comment. The processing unit can also hide the abusive comment. For example, the processing unit sets the abusive comment to be hidden so that only specific users can view the comment. This allows the processing unit to prevent the deletion of the abusive comment from having a negative impact on other users. Some or all of the above-described processing in the processing unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the processing unit can input the abusive comment into a generation AI, which then deletes the comment. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. This allows the processing unit to prevent the deletion of defamatory comments from having a negative impact on other users.

[0064] The processing unit can use conversion rules to convert abusive comments into a ladylike tone. Examples of conversion rules include, but are not limited to, rules for converting offensive words or phrases into softer expressions. For example, the processing unit converts abusive comments into a ladylike tone. For example, the processing unit converts a comment such as "idiot" into "idiot." The processing unit can also convert a comment such as "die" into "please die." By converting abusive comments into a ladylike tone, the processing unit can soften the tone of the comment. Some or all of the above-described processing in the processing unit may be performed using, or without, a generation AI. For example, the processing unit can input abusive comments into a generation AI, which then converts the comments into a ladylike tone. Examples of generation AI include, but are not limited to, a text generation AI (e.g., LLM) and a multimodal generation AI. This allows the processing unit to soften the tone of the comment by converting the slanderous comment into a ladylike tone.

[0065] The reception unit can estimate the user's emotions and adjust the timing of comment acceptance based on the estimated user emotions. For example, if the user is angry, the reception unit temporarily delays the acceptance of comments to allow the user time to calm down. For example, if the reception unit estimates that the user is angry, the reception unit delays the acceptance of comments for several minutes to allow the user time to calm down. Furthermore, if the user is sad, the reception unit can display an encouraging message to encourage the acceptance of comments. For example, if the reception unit estimates that the user is sad, the reception unit displays an encouraging message to encourage the acceptance of comments. Furthermore, if the user is excited, the reception unit can quickly accept comments to reflect the user's heightened emotions. For example, if the reception unit estimates that the user is excited, the reception unit immediately accepts comments to reflect the user's heightened emotions. In this way, the reception unit can adjust the timing of comment acceptance according to the user's emotions, thereby accepting comments at an appropriate time. Some or all of the above-described processing in the reception unit may be performed, for example, using a generation AI or without using a generation AI. For example, the reception unit inputs the user's emotional data into the generation AI, which then analyzes the emotions and adjusts the timing of receiving comments. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0066] The reception unit can analyze a user's past comment history and select the optimal reception method. For example, if a user has posted offensive comments in the past, the reception unit displays a warning message before accepting the comment. For example, if a user has a history of posting offensive comments in the past, the reception unit displays a warning message saying, "Please refrain from posting offensive comments" before accepting the comment. The reception unit can also quickly accept comments if a user has posted constructive comments in the past. For example, the reception unit prioritizes accepting comments if the user has a history of posting constructive comments in the past. Furthermore, the reception unit can analyze a user's interest in a specific topic from the user's past comment history and prioritize accepting comments on related topics. For example, if a user has frequently posted comments on a specific topic in the past, the reception unit prioritizes accepting comments related to that topic. This allows the reception unit to select the optimal reception method based on the user's past comment history, enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's past comment history data into the generation AI, which can then select the optimal reception method. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0067] When accepting comments, the acceptance unit can filter the comments based on the user's current activity status and areas of interest. For example, the acceptance unit accepts only relevant comments based on the content of the page the user is currently viewing. For example, if the user is viewing a specific news article, the acceptance unit accepts only comments related to the news article. The acceptance unit can also analyze the user's areas of interest from the user's past browsing history and preferentially accept related comments. For example, if the user has frequently viewed articles on a specific topic in the past, the acceptance unit preferentially accepts comments related to the topic. Furthermore, if the user is participating in a specific event, the acceptance unit can preferentially accept comments related to the event. For example, if the user is participating in a specific event, the acceptance unit preferentially accepts comments related to the event. In this way, the acceptance unit can preferentially accept highly relevant comments by filtering comments based on the user's current activity status and areas of interest. Some or all of the above-described processing in the acceptance unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's activity status and interest field data to the generation AI, which can then filter the comments. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0068] The reception unit can estimate the user's emotions and determine the priority of comments to be received based on the estimated user's emotions. For example, if the user is angry, the reception unit can set aggressive comments to a low priority and delay their reception. For example, if the reception unit estimates that the user is angry, the reception unit can set aggressive comments to a low priority and delay their reception for several minutes. Furthermore, if the user is relaxed, the reception unit can set constructive comments to a high priority and quickly receive them. For example, if the reception unit estimates that the user is relaxed, the reception unit can set constructive comments to a high priority and quickly receive them. Furthermore, if the user is excited, the reception unit can set comments reflecting the user's heightened emotions to a high priority and quickly receive them. For example, if the reception unit estimates that the user is excited, the reception unit can set comments reflecting the user's heightened emotions to a high priority and quickly receive them. This allows the reception unit to determine the priority of comments according to the user's emotions, thereby enabling appropriate comment reception. Some or all of the above-described processing in the reception unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's emotional data into the generation AI, which can then determine the priority of the comments. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0069] When accepting comments, the reception unit can prioritize accepting highly relevant comments by taking into account the user's geographical location information. For example, if the user is in a specific region, the reception unit prioritizes accepting comments related to that region. For example, if the user is in a specific city, the reception unit prioritizes accepting comments related to that city. Furthermore, if the user is traveling, the reception unit can prioritize accepting comments related to the travel destination. For example, if the reception unit estimates that the user is traveling, the reception unit prioritizes accepting comments related to the travel destination. Furthermore, if the user is at an event venue, the reception unit can prioritize accepting comments related to the event. For example, if the user is at a specific event venue, the reception unit prioritizes accepting comments related to the event. This allows the reception unit to prioritize accepting highly relevant comments based on the user's geographical location information, thereby enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, or without, a generation AI. For example, the reception unit can input the user's geographical location information data into the generation AI, which can then filter the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0070] The reception unit can analyze the user's social media activity when receiving comments and receive relevant comments. For example, if a user frequently uses a specific hashtag, the reception unit can prioritize receiving comments related to that hashtag. For example, if a user frequently uses a specific hashtag, the reception unit can prioritize receiving comments related to that hashtag. Furthermore, if a user belongs to a specific group, the reception unit can prioritize receiving comments related to that group. For example, if a user belongs to a specific group, the reception unit can prioritize receiving comments related to that group. Furthermore, if a user frequently posts about a specific topic, the reception unit can prioritize receiving comments related to that topic. For example, if a user frequently posts about a specific topic, the reception unit can prioritize receiving comments related to that topic. This allows the reception unit to receive relevant comments based on the user's social media activity, thereby enabling appropriate comment reception. Some or all of the above-described processing in the reception unit can be performed using, or without, a generation AI. For example, the reception unit can input the user's social media activity data into the generation AI, which can then filter the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0071] The analysis unit can estimate the user's emotions and adjust the presentation method of the analysis based on the estimated user's emotions. For example, if the user is angry, the analysis unit displays the analysis results in a calm tone. For example, if the analysis unit estimates that the user is angry, the analysis unit displays the analysis results in a calm tone to calm the user's emotions. The analysis unit can also display detailed analysis results if the user is relaxed. For example, if the analysis unit estimates that the user is relaxed, the analysis unit displays detailed analysis results to provide the user with sufficient information. Furthermore, if the user is excited, the analysis unit can display analysis results with visually stimulating effects. For example, if the analysis unit estimates that the user is excited, the analysis unit displays analysis results with visually stimulating effects to maintain the user's excitement. In this way, the analysis unit can provide appropriate analysis results by adjusting the presentation method of the analysis according to the user's emotions. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input the user's emotion data into a generation AI, which can then adjust the presentation method of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0072] During analysis, the analysis unit can adjust the level of detail of the analysis based on the importance of the comment. For example, the analysis unit performs a detailed analysis on comments with high importance. For example, the analysis unit performs a detailed analysis on comments with high importance to gain a deeper understanding of the content of the comment. The analysis unit can also perform a simplified analysis on comments with low importance. For example, the analysis unit can perform a simplified analysis on comments with low importance to quickly provide results. Furthermore, the analysis unit can adjust the display order of the analysis results based on the importance of the comment. For example, the analysis unit can prioritize displaying the analysis results of comments with high importance and postpone displaying the analysis results of comments with low importance. In this way, the analysis unit can provide appropriate analysis results by adjusting the level of detail of the analysis based on the importance of the comment. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment importance data to the generation AI, which can then adjust the level of detail of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0073] During analysis, the analysis unit can apply different analysis algorithms depending on the category of the comment. For example, the analysis unit applies an analysis algorithm that emphasizes specific political keywords to comments related to politics. For example, the analysis unit applies an analysis algorithm that emphasizes specific political keywords to comments related to politics, thereby gaining a deeper understanding of the content of the comment. The analysis unit can also apply an analysis algorithm that emphasizes sports terms to comments related to sports. For example, the analysis unit applies an analysis algorithm that emphasizes sports terms to comments related to sports, thereby gaining a deeper understanding of the content of the comment. The analysis unit can also apply an analysis algorithm that emphasizes entertainment-related keywords to comments related to entertainment. For example, the analysis unit applies an analysis algorithm that emphasizes entertainment-related keywords to comments related to entertainment, thereby gaining a deeper understanding of the content of the comment. This allows the analysis unit to provide accurate analysis results by applying an appropriate analysis algorithm depending on the category of the comment. Some or all of the above-mentioned processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment category data into the generation AI, which then applies an appropriate analysis algorithm. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0074] The analysis unit can estimate the user's emotions and adjust the length of the analysis based on the estimated user's emotions. For example, if the user is in a hurry, the analysis unit displays a short and concise analysis result. For example, if the analysis unit estimates that the user is in a hurry, the analysis unit displays a short and concise analysis result to quickly provide information. The analysis unit can also display a detailed analysis result if the user is relaxed. For example, if the analysis unit estimates that the user is relaxed, the analysis unit displays a detailed analysis result to provide the user with sufficient information. Furthermore, if the user is excited, the analysis unit can display an analysis result with visually stimulating effects. For example, if the analysis unit estimates that the user is excited, the analysis unit displays an analysis result with visually stimulating effects to maintain the user's excitement. This allows the analysis unit to provide appropriate analysis results by adjusting the length of the analysis according to the user's emotions. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input the user's emotion data into the generation AI, which can then adjust the length of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0075] During analysis, the analysis unit can determine the analysis priority based on the time when the comments were posted. For example, the analysis unit can prioritize the analysis of the most recent comments and display the results in real time. For example, the analysis unit can prioritize the analysis of the most recent comments and display the results in real time, thereby providing quick feedback to the user. The analysis unit can also determine the analysis priority for past comments based on their importance. For example, the analysis unit can prioritize the analysis of past comments based on their importance and prioritize analyzing important comments. Furthermore, the analysis unit can prioritize the analysis of comments posted during a specific event period. For example, the analysis unit can prioritize the analysis of comments posted during a specific event period and quickly provide information related to the event. In this way, the analysis unit can provide appropriate analysis results by determining the analysis priority based on the time when the comments were posted. Some or all of the above-described processing in the analysis unit can be performed using, or without, a generation AI. For example, the analysis unit can input comment posting time data to the generation AI, which can then determine the analysis priority. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0076] During analysis, the analysis unit can adjust the order of analysis based on the relevance of the comments. For example, the analysis unit prioritizes analysis of highly relevant comments and displays the results. For example, the analysis unit prioritizes analysis of highly relevant comments and displays the results, thereby quickly providing important information to the user. The analysis unit can also perform a simplified analysis of less relevant comments. For example, the analysis unit can perform a simplified analysis of less relevant comments and quickly provide the results. Furthermore, the analysis unit can adjust the display order of the analysis results based on the relevance of the comments. For example, the analysis unit prioritizes displaying the analysis results of highly relevant comments and postpones the analysis results of less relevant comments. In this way, the analysis unit can provide appropriate analysis results by adjusting the analysis order based on the relevance of the comments. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment relevance data into the generation AI, which can then adjust the analysis order. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0077] The detection unit can estimate the user's emotions and adjust the detection criteria based on the estimated user's emotions. For example, if the user is angry, the detection unit detects aggressive comments more strictly. For example, if the detection unit estimates that the user is angry, the detection unit detects aggressive comments more strictly and responds promptly. The detection unit can also detect mildly abusive comments more gently if the user is relaxed. For example, if the detection unit estimates that the user is relaxed, the detection unit detects mildly abusive comments more gently and avoids excessive responses. Furthermore, the detection unit can also detect comments that reflect heightened emotions more strictly if the user is excited. For example, if the detection unit estimates that the user is excited, the detection unit detects comments that reflect heightened emotions more strictly and responds promptly. This allows the detection unit to adjust the detection criteria according to the user's emotions, thereby enabling appropriate detection of abusive comments. Some or all of the above-mentioned processing in the detection unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the detection unit can input user emotion data to the generation AI, which can then adjust its detection criteria. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0078] The detection unit can improve the accuracy of detection by taking into account the interrelationships between comments during detection. The detection unit, for example, analyzes the context of comments to detect abusive comments. For example, the detection unit analyzes the context of comments to detect abusive comments, thereby performing highly accurate detection that takes context into account. The detection unit can also analyze consecutive comments by the same user to detect abusive comments. For example, the detection unit analyzes consecutive comments by the same user to detect abusive comments, thereby accurately understanding the user's intention. Furthermore, the detection unit can also analyze interactions between multiple users to detect abusive comments. For example, the detection unit analyzes interactions between multiple users to detect abusive comments, thereby performing highly accurate detection that takes into account the flow of the dialogue. As a result, the detection unit improves the accuracy of detecting abusive comments by taking into account the interrelationships between comments. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input comment interrelationship data into the generation AI, which may then improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0079] The detection unit can perform detection by taking into account attribute information of the poster of the comment. The detection unit adjusts the detection criteria for defamatory comments based on, for example, the poster's age group. For example, the detection unit adjusts the detection criteria for defamatory comments based on the poster's age group and takes appropriate measures according to the poster's age. The detection unit can also analyze the poster's past comment history to detect defamatory comments. For example, the detection unit analyzes the poster's past comment history to detect defamatory comments, thereby performing highly accurate detection that takes into account the poster's behavioral patterns. Furthermore, the detection unit can also detect defamatory comments by taking into account the poster's regional information. For example, the detection unit takes into account the poster's regional information and performs detection that takes into account words and expressions unique to the region. As a result, the detection unit improves the accuracy of detecting defamatory comments by taking into account the attribute information of the comment poster. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input attribute information data of the poster into the generation AI, which may improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0080] The detection unit can estimate the user's emotions and adjust the order in which the detection results are displayed based on the estimated user's emotions. For example, if the user is angry, the detection unit can display aggressive comments first to alert the user. For example, if the detection unit estimates that the user is angry, the detection unit can display aggressive comments first to alert the user. Furthermore, if the user is relaxed, the detection unit can display minor defamatory comments later. For example, if the detection unit estimates that the user is relaxed, the detection unit can display minor defamatory comments later to reduce the user's stress. Furthermore, if the user is excited, the detection unit can display comments that reflect the user's heightened emotions first. For example, if the detection unit estimates that the user is excited, the detection unit can display comments that reflect the user's heightened emotions first to maintain the user's excitement. In this way, the detection unit can adjust the order in which the detection results are displayed based on the user's emotions, thereby enabling appropriate alerts. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit can input user emotion data to the generation AI and adjust the order in which the generation AI displays the detection results. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0081] The detection unit can perform detection by taking into account the geographical distribution of comments. For example, the detection unit prioritizes detecting abusive comments that occur frequently in a specific region. For example, the detection unit prioritizes detecting abusive comments that occur frequently in a specific region and responds quickly to issues specific to that region. The detection unit can also detect abusive comments by taking into account cultural and linguistic differences between regions. For example, the detection unit takes into account cultural and linguistic differences between regions and appropriately detects words and expressions specific to the region. Furthermore, the detection unit can also detect abusive comments by analyzing interactions between geographically close users. For example, the detection unit analyzes interactions between geographically close users and detects abusive comments, thereby responding quickly to issues within the local community. In this way, the detection unit can improve the accuracy of detecting abusive comments specific to a region by taking into account the geographical distribution of comments. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input geographic distribution data of comments into the generation AI, which may improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0082] The detection unit can improve the accuracy of detection by referring to literature related to the comment during detection. The detection unit, for example, refers to related research papers to improve the accuracy of detecting defamatory comments. For example, the detection unit refers to related research papers to improve the accuracy of detecting defamatory comments based on the latest research results. The detection unit can also improve the accuracy of detecting defamatory comments by referring to related news articles. For example, the detection unit can detect defamatory comments that reflect the latest social conditions by referring to related news articles. Furthermore, the detection unit can also improve the accuracy of detecting defamatory comments by referring to related books. For example, the detection unit can detect defamatory comments taking into account historical background and cultural context by referring to related books. In this way, the detection unit can improve the accuracy of detecting defamatory comments by referring to literature related to the comment. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input related literature data of the comment into the generation AI, which may improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0083] The processing unit can estimate the user's emotions and adjust the processing method based on the estimated user's emotions. For example, if the user is angry, the processing unit deletes offensive comments. For example, if the processing unit estimates that the user is angry, the processing unit immediately deletes offensive comments to prevent them from negatively impacting other users. The processing unit can also convert aggressive comments into a ladylike tone if the user is relaxed. For example, if the processing unit estimates that the user is relaxed, the processing unit converts aggressive comments into a ladylike tone and softens the tone of the comments. Furthermore, if the user is excited, the processing unit can convert comments reflecting heightened emotions into a ladylike tone. For example, if the processing unit estimates that the user is excited, the processing unit converts comments reflecting heightened emotions into a ladylike tone and softens the tone of the comments. This allows the processing unit to adjust the processing method according to the user's emotions, thereby enabling appropriate comment processing. Some or all of the above-mentioned processing in the processing unit may be performed using, or without, a generation AI. For example, the processing unit can input the user's emotion data into the generation AI, which then adjusts the processing method. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0084] During processing, the processing unit can improve the accuracy of processing by taking into account the interrelationships between comments. For example, the processing unit analyzes the context of comments and deletes abusive comments. For example, the processing unit analyzes the context of comments and deletes abusive comments, thereby performing highly accurate processing that takes context into account. The processing unit can also analyze consecutive comments from the same user and delete abusive comments. For example, the processing unit analyzes consecutive comments from the same user and deletes abusive comments, thereby accurately understanding the user's intention. Furthermore, the processing unit can analyze interactions between multiple users and delete abusive comments. For example, the processing unit analyzes interactions between multiple users and deletes abusive comments, thereby performing highly accurate processing that takes into account the flow of the dialogue. As a result, the processing unit takes into account the interrelationships between comments, thereby improving the accuracy of processing abusive comments. Some or all of the above-mentioned processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit can input comment interrelationship data to a generation AI, which can improve the accuracy of processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0085] The processing unit can perform processing while taking into account attribute information of the comment poster. The processing unit, for example, adjusts the processing method for abusive comments based on the poster's age group. For example, the processing unit adjusts the processing method for abusive comments based on the poster's age group and takes appropriate action according to the poster's age. The processing unit can also analyze the poster's past comment history and delete or convert abusive comments. For example, the processing unit analyzes the poster's past comment history and deletes or converts abusive comments, thereby performing highly accurate processing that takes into account the poster's behavioral patterns. Furthermore, the processing unit can also delete or convert abusive comments while taking into account the poster's regional information. For example, the processing unit takes into account the poster's regional information and performs processing that takes into account words and expressions specific to the region. As a result, the processing unit takes into account the attribute information of the comment poster, thereby improving the accuracy of processing abusive comments. Some or all of the above-described processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit inputs attribute information data of the poster into the generation AI, which can improve the accuracy of the processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0086] The processing unit can estimate the user's emotions and determine processing priorities based on the estimated user's emotions. For example, if the user is angry, the processing unit prioritizes deleting offensive comments. For example, if the processing unit estimates that the user is angry, the processing unit prioritizes deleting offensive comments to prevent them from negatively impacting other users. The processing unit can also convert aggressive comments into a ladylike tone if the user is relaxed. For example, if the processing unit estimates that the user is relaxed, the processing unit converts aggressive comments into a ladylike tone and softens the tone of the comments. Furthermore, if the user is excited, the processing unit can convert comments reflecting heightened emotions into a ladylike tone. For example, if the processing unit estimates that the user is excited, the processing unit converts comments reflecting heightened emotions into a ladylike tone and softens the tone of the comments. This allows the processing unit to determine processing priorities according to the user's emotions, thereby enabling appropriate comment processing. Some or all of the above-described processing in the processing unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the processing unit can input user emotion data to a generation AI, which can then determine the processing priority. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0087] The processing unit may take into account the geographical distribution of comments during processing. For example, the processing unit may prioritize deleting abusive comments that occur frequently in a specific region. For example, the processing unit may prioritize deleting abusive comments that occur frequently in a specific region, thereby quickly addressing issues specific to that region. The processing unit may also delete or convert abusive comments taking into account cultural and linguistic differences between regions. For example, the processing unit may appropriately process words and expressions specific to the region by taking into account cultural and linguistic differences between regions. Furthermore, the processing unit may analyze interactions between geographically close users and delete or convert abusive comments. For example, the processing unit may analyze interactions between geographically close users and delete or convert abusive comments, thereby quickly addressing issues within the local community. In this way, by taking into account the geographical distribution of comments, the processing unit improves the accuracy of processing abusive comments specific to the region. Some or all of the above-described processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit can input geographical distribution data of comments to the generation AI, which can improve the accuracy of processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0088] During processing, the processing unit can improve the accuracy of processing by referring to literature related to the comment. The processing unit, for example, refers to related research papers to improve the accuracy of processing defamatory comments. For example, the processing unit refers to related research papers to improve the accuracy of processing defamatory comments based on the latest research results. The processing unit can also refer to related news articles to improve the accuracy of processing defamatory comments. For example, the processing unit refers to related news articles to process defamatory comments that reflect the latest social conditions. Furthermore, the processing unit can also refer to related books to improve the accuracy of processing defamatory comments. For example, the processing unit refers to related books to process defamatory comments taking into account historical background and cultural context. In this way, by referring to literature related to the comment, the processing unit improves the accuracy of processing defamatory comments. Some or all of the above-mentioned processing in the processing unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the processing unit can input the related literature data of the comment into the generation AI, which can improve the accuracy of the processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. === Hard Collateral 1-1 === Each of the multiple elements including the above-described reception unit, analysis unit, detection unit, and processing unit is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the reception unit is realized by the reception device 38 of the smart device 14 and receives comments posted to a comment section of an SNS in real time. The analysis unit is realized by the specific processing unit 290 of the data processing device 12 and analyzes the comments using natural language processing technology. The detection unit is realized by the specific processing unit 290 of the data processing device 12 and detects abusive comments using a machine learning algorithm. The processing unit is realized by the specific processing unit 290 of the data processing device 12 and deletes abusive comments or converts them into a ladylike tone. === Hard Collateral 1-2 === Each of the multiple elements including the above-described reception unit, analysis unit, detection unit, and processing unit is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the reception unit is realized by the microphone 238 of the smart glasses 214 and receives comments posted to a comment section of an SNS in real time. The analysis unit is realized by the specific processing unit 290 of the data processing device 12 and analyzes the comments using natural language processing technology. The detection unit is realized by the specific processing unit 290 of the data processing device 12 and detects abusive comments using a machine learning algorithm. The processing unit is realized by the specific processing unit 290 of the data processing device 12 and deletes abusive comments or converts them into a ladylike tone. === Hard Collateral 1-3 === Each of the multiple elements including the above-described reception unit, analysis unit, detection unit, and processing unit is realized, for example, by at least one of the headset-type terminal 314 and the data processing device 12. For example, the reception unit is realized by the microphone 238 of the headset-type terminal 314 and receives comments posted to the comment section of the SNS in real time. The analysis unit is realized by the specific processing unit 290 of the data processing device 12 and analyzes the comments using natural language processing technology. The detection unit is realized by the specific processing unit 290 of the data processing device 12 and detects abusive comments using a machine learning algorithm. The processing unit is realized by the specific processing unit 290 of the data processing device 12 and deletes abusive comments or converts them into a ladylike tone. === Hard Collateral 1-4 === Each of the multiple elements including the above-mentioned reception unit, analysis unit, detection unit, and processing unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the reception unit is realized by the microphone 238 of the robot 414 and receives comments posted to a comment section of an SNS in real time. The analysis unit is realized by the specific processing unit 290 of the data processing device 12 and analyzes the comments using natural language processing technology. The detection unit is realized by the specific processing unit 290 of the data processing device 12 and detects abusive comments using a machine learning algorithm. The processing unit is realized by the specific processing unit 290 of the data processing device 12 and deletes abusive comments or converts them into a ladylike tone.

[0089] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0090] The reception unit can analyze a user's past comment history and select the optimal reception method. For example, if a user has posted offensive comments in the past, the reception unit displays a warning message before accepting the comment. For example, if a user has a history of posting offensive comments in the past, the reception unit displays a warning message saying, "Please refrain from posting offensive comments" before accepting the comment. The reception unit can also quickly accept comments if a user has posted constructive comments in the past. For example, the reception unit prioritizes accepting comments if the user has a history of posting constructive comments in the past. Furthermore, the reception unit can analyze a user's interest in a specific topic from the user's past comment history and prioritize accepting comments on related topics. For example, if a user has frequently posted comments on a specific topic in the past, the reception unit prioritizes accepting comments related to that topic. This allows the reception unit to select the optimal reception method based on the user's past comment history, enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit can input the user's past comment history data into the generation AI, which can then select the optimal reception method. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0091] The analysis unit can estimate the user's emotions and adjust the presentation method of the analysis based on the estimated user's emotions. For example, if the user is angry, the analysis result can be displayed in a calm tone. For example, if the analysis unit estimates that the user is angry, the analysis result can be displayed in a calm tone to calm the user's emotions. The analysis unit can also display detailed analysis results if the user is relaxed. For example, if the analysis unit estimates that the user is relaxed, the analysis unit displays detailed analysis results to provide the user with sufficient information. Furthermore, if the user is excited, the analysis unit can display analysis results with visually stimulating effects. For example, if the analysis unit estimates that the user is excited, the analysis unit displays analysis results with visually stimulating effects to maintain the user's excitement. In this way, the analysis unit can provide appropriate analysis results by adjusting the presentation method of the analysis according to the user's emotions. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input the user's emotion data into a generation AI, which can then adjust the presentation method of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0092] The detection unit can estimate the user's emotions and adjust the detection criteria based on the estimated user's emotions. For example, if the user is angry, the detection unit can strictly detect offensive comments. For example, if the detection unit estimates that the user is angry, the detection unit can strictly detect offensive comments and respond quickly. The detection unit can also gently detect mildly abusive comments if the user is relaxed. For example, if the detection unit estimates that the user is relaxed, the detection unit can gently detect mildly abusive comments and avoid excessive responses. Furthermore, the detection unit can also strictly detect comments that reflect heightened emotions if the user is excited. For example, if the detection unit estimates that the user is excited, the detection unit can strictly detect comments that reflect heightened emotions and respond quickly. This allows the detection unit to adjust the detection criteria according to the user's emotions, thereby enabling appropriate detection of abusive comments. Some or all of the above-mentioned processing in the detection unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the detection unit can input user emotion data to the generation AI, which can then adjust its detection criteria. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0093] The processing unit can estimate the user's emotions and adjust the processing method based on the estimated user's emotions. For example, if the user is angry, the processing unit deletes offensive comments. For example, if the processing unit estimates that the user is angry, the processing unit immediately deletes the offensive comments to prevent them from negatively impacting other users. The processing unit can also convert aggressive comments into a ladylike tone if the user is relaxed. For example, if the processing unit estimates that the user is relaxed, the processing unit converts aggressive comments into a ladylike tone and softens the tone of the comments. Furthermore, if the user is excited, the processing unit can convert comments reflecting heightened emotions into a ladylike tone. For example, if the processing unit estimates that the user is excited, the processing unit converts comments reflecting heightened emotions into a ladylike tone and softens the tone of the comments. This allows the processing unit to adjust the processing method according to the user's emotions, thereby enabling appropriate comment processing. Some or all of the above-described processing in the processing unit may be performed using, or without, a generation AI. For example, the processing unit can input the user's emotion data into the generation AI, which then adjusts the processing method. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0094] The reception unit can estimate the user's emotions and adjust the timing of comment acceptance based on the estimated user emotions. For example, if the user is angry, the reception unit temporarily delays the acceptance of comments to allow the user time to calm down. For example, if the reception unit estimates that the user is angry, the reception unit delays the acceptance of comments for several minutes to allow the user time to calm down. Furthermore, if the user is sad, the reception unit can display an encouraging message to encourage the acceptance of comments. For example, if the reception unit estimates that the user is sad, the reception unit displays an encouraging message to encourage the acceptance of comments. Furthermore, if the user is excited, the reception unit can quickly accept comments to reflect the user's heightened emotions. For example, if the reception unit estimates that the user is excited, the reception unit immediately accepts comments to reflect the user's heightened emotions. In this way, the reception unit can adjust the timing of comment acceptance according to the user's emotions, thereby accepting comments at an appropriate time. Some or all of the above-described processing in the reception unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the reception unit inputs the user's emotional data into the generation AI, which then analyzes the emotions and adjusts the timing of receiving comments. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0095] When accepting comments, the acceptance unit can filter the comments based on the user's current activity status and areas of interest. For example, the acceptance unit accepts only relevant comments based on the content of the page the user is currently viewing. For example, if the user is viewing a specific news article, the acceptance unit accepts only comments related to that news article. The acceptance unit can also analyze the user's areas of interest from the user's past browsing history and preferentially accept related comments. For example, if the user has frequently viewed articles on a specific topic in the past, the acceptance unit preferentially accepts comments related to that topic. Furthermore, if the user is participating in a specific event, the acceptance unit can preferentially accept comments related to that event. For example, if the user is participating in a specific event, the acceptance unit preferentially accepts comments related to that event. In this way, the acceptance unit can preferentially accept highly relevant comments by filtering comments based on the user's current activity status and areas of interest. Some or all of the above-described processing by the acceptance unit may be performed using, or without, a generation AI. For example, the acceptance unit can input the user's activity status and area of ​​interest data into the generation AI, which then filters the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0096] During analysis, the analysis unit can adjust the level of detail of the analysis based on the importance of the comment. For example, the analysis unit performs a detailed analysis on comments with high importance. For example, the analysis unit performs a detailed analysis on comments with high importance to gain a deeper understanding of the content of the comment. The analysis unit can also perform a simplified analysis on comments with low importance. For example, the analysis unit can perform a simplified analysis on comments with low importance to quickly provide results. Furthermore, the analysis unit can adjust the display order of the analysis results based on the importance of the comment. For example, the analysis unit can prioritize displaying the analysis results of comments with high importance and postpone displaying the analysis results of comments with low importance. In this way, the analysis unit can provide appropriate analysis results by adjusting the level of detail of the analysis based on the importance of the comment. Some or all of the above-described processing in the analysis unit may be performed using, or without, a generation AI. For example, the analysis unit can input comment importance data to the generation AI, which can then adjust the level of detail of the analysis. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0097] The detection unit can improve the accuracy of detection by taking into account the interrelationships between comments. For example, the detection unit analyzes the context of comments to detect abusive comments. For example, the detection unit analyzes the context of comments to detect abusive comments, thereby performing highly accurate detection that takes context into account. The detection unit can also analyze consecutive comments from the same user to detect abusive comments. For example, the detection unit analyzes consecutive comments from the same user to detect abusive comments, thereby accurately understanding the user's intention. Furthermore, the detection unit can also analyze interactions between multiple users to detect abusive comments. For example, the detection unit analyzes interactions between multiple users to detect abusive comments, thereby performing highly accurate detection that takes into account the flow of the dialogue. As a result, the detection unit improves the accuracy of detecting abusive comments by taking into account the interrelationships between comments. Some or all of the above-described processing in the detection unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the detection unit may input comment interrelationship data into the generation AI, which may then improve the accuracy of detection. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0098] During processing, the processing unit can improve the accuracy of processing by taking into account the interrelationships between comments. For example, the processing unit analyzes the context of comments and deletes abusive comments. For example, the processing unit analyzes the context of comments and deletes abusive comments, thereby performing highly accurate processing that takes context into account. The processing unit can also analyze consecutive comments from the same user and delete abusive comments. For example, the processing unit analyzes consecutive comments from the same user and deletes abusive comments, thereby accurately understanding the user's intention. Furthermore, the processing unit can analyze interactions between multiple users and delete abusive comments. For example, the processing unit analyzes interactions between multiple users and deletes abusive comments, thereby performing highly accurate processing that takes into account the flow of the dialogue. As a result, the processing unit improves the accuracy of processing abusive comments by taking into account the interrelationships between comments. Some or all of the above-mentioned processing in the processing unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the processing unit can input comment interrelationship data to a generation AI, which can improve the accuracy of processing. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI.

[0099] When accepting comments, the reception unit can prioritize accepting highly relevant comments by taking into account the user's geographical location information. For example, if the user is in a specific region, it can prioritize accepting comments related to that region. For example, if the user is in a specific city, it can prioritize accepting comments related to that city. Furthermore, if the user is traveling, the reception unit can prioritize accepting comments related to the travel destination. For example, if the reception unit estimates that the user is traveling, it can prioritize accepting comments related to the travel destination. Furthermore, if the user is at an event venue, the reception unit can prioritize accepting comments related to the event. For example, if the reception unit detects that the user is at a specific event venue, it can prioritize accepting comments related to the event. This allows the reception unit to prioritize accepting highly relevant comments based on the user's geographical location information, thereby enabling appropriate comment acceptance. Some or all of the above-described processing in the reception unit may be performed using, or without, a generation AI. For example, the reception unit can input the user's geographical location information data into the generation AI, which can then filter the comments. The generation AI may be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples.

[0100] The processing flow of the second embodiment will be briefly explained below.

[0101] Step 1: The reception unit receives comments. Comments include text comments, image comments, and audio comments. The reception unit can receive comments from external systems via the SNS posting interface or API. For example, it can receive comments posted in the SNS comment section in real time. It can also convert audio comments into text using voice recognition technology and receive them. Step 2: The analysis unit uses natural language processing technology to analyze the comments received by the reception unit. For example, it uses morphological analysis to split the words in the comments and perform grammatical analysis. It also uses semantic analysis to understand the content of the comments and identify offensive words and phrases. It can also analyze the sentiment of the comments. Step 3: The detection unit detects abusive comments based on the comments analyzed by the analysis unit. For example, it uses machine learning algorithms or rule-based algorithms to detect whether specific words or phrases are included. It can also refer to a predefined list of abusive comments and detect whether the comment matches that list. Step 4: The processing unit deletes or converts the abusive comments detected by the detection unit. For example, the processing unit can completely delete the abusive comments, hide them, or convert them to a more ladylike tone. The processing unit converts offensive words and phrases into softer expressions based on pre-trained conversion rules.

[0102] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0103] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of the generative AI include a neural network (NN) and a neural network (NN). The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, image data, etc. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and may perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-mentioned parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. The processing performed by an AI including the generative AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI including the generative AI.

[0104] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0105] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0106] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0107] 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.

[0108] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0109] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0110] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0111] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0112] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0113] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0114] 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.

[0115] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0116] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0117] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0118] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0119] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0120] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0121] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0122] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0123] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0124] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0125] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0126] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0127] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0128] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0129] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0130] 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.

[0131] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0132] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0133] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0134] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0135] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0136] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0137] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0138] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0139] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0140] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0141] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0142] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0143] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0144] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0145] The control 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0146] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0147] 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.

[0148] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0149] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0150] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0151] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0152] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0153] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0154] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0155] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0156] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0157] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0158] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0159] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0160] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[0161] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[0162] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0163] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0164] 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.

[0165] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0166] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. A processor also includes a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0167] The hardware resource that executes the specific process may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0168] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0169] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0170] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0171] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0172] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0173] [Explanation of symbols]

[0174] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. a reception unit for receiving comments; an analysis unit that analyzes the comments received by the reception unit; a detection unit that detects defamatory comments based on the comments analyzed by the analysis unit; a processing unit that deletes or converts the defamatory comments detected by the detection unit. A system characterized by:

2. The analysis unit Analyzing the content of comments using natural language processing technology 2. The system of claim 1.

3. The detection unit Use an algorithm to detect abusive comments 2. The system of claim 1.

4. The processing unit Delete abusive comments 2. The system of claim 1.

5. The processing unit Use conversion rules to convert abusive comments into ladylike speech 2. The system of claim 1.

6. The reception unit Analyze user sentiment and adjust the timing of comment acceptance based on the analyzed user sentiment.

2. The system of claim 1.

7. The reception unit Analyze the user's past comment history and select the optimal reception method 2. The system of claim 1.

8. The reception unit Filter comments based on your current activity and interests 2. The system of claim 1.

Images