System

Abstract

To provide a system capable of effectively managing improper content and spam in an online community.SOLUTION: A system comprises a detection unit, a deletion unit, an identification unit, a block unit, a classification unit, a tagging unit, and a guidance unit. The detection unit detects contents which have been determined to be improper from among contents posted on an online community. The deletion unit deletes contents which have been determined to be improper. The identification unit determines whether the content to be posted is a content containing a spam content or a content including an unnecessary link. The block unit blocks the posting of content which has been determined to a content containing a spam content or an unnecessary link. The classification unit classifies a content into related topics. The tagging unit assigns a tag to the classified contents. The guidance unit guides a community guide and rules to a new member.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including a description and related instruction sentence regarding 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] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional techniques do not adequately manage inappropriate content and spam in online communities, and there is room for improvement.

[0005] The system according to the embodiment aims to effectively manage inappropriate content and spam in online communities. [Means for solving the problem]

[0006] The system according to the embodiment includes a detection unit, a deletion unit, an identification unit, a blocking unit, a classification unit, a tagging unit, and a guidance unit. The detection unit detects content determined to be inappropriate from among content posted to the online community. The deletion unit deletes the content determined to be inappropriate by the detection unit. The identification unit determines whether or not content posted to the online community is spam content or content including unnecessary links. The blocking unit blocks the posting of content determined by the identification unit to be spam content or content including unnecessary links. The classification unit classifies content posted to the online community into related topics. The tagging unit tags the content classified by the classification unit. The guidance unit provides a community guide and rules to new members of the online community. Effect of the Invention

[0007] The system according to the embodiment can effectively manage inappropriate content and spam in online communities. [Brief description 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. [Diagram 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. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 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. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 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 PREFERRED EMBODIMENTS

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

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

[0011] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one 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 signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

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

[0014] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or 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. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "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, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[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 (e.g., 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., voice and / or text). Display 40A displays visible information such as text and images according to instructions from processor 46. Speaker 40B outputs voice according to 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 Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

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

[0024] As shown in Fig. 2, 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. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process 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 process 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[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 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. Also, the data processing device 12 may be a server device, or may be a terminal device (e.g., a mobile phone, a robot, a home appliance, etc.) owned by a user. Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) The online community moderator system according to an embodiment of the present invention is a system that detects and deletes inappropriate content, takes measures against spam, classifies and tags topics, and provides guidance to users. This system detects inappropriate content from content posted to an online community. For example, it detects content that includes discriminatory or violent expressions. This detection is performed by AI, and the content is automatically deleted without waiting for a report by the user. Next, as a measure against spam, spam content and unnecessary links are identified and the posting of these contents is blocked. The identification is performed using a machine learning model to keep the community clean. Furthermore, the system improves navigation by classifying posts into related topics and attaching correct tags. This allows users to quickly find the information they need. In addition, the system provides new members with a community guide and a guide to the rules to support smooth participation. This allows new members to understand the rules of the community and participate appropriately. Furthermore, the system analyzes the behavioral history of members of the online community and predicts inappropriate behavior. It issues warnings for predicted inappropriate behavior, and provides positive feedback when improved behavior is confirmed. In addition, it performs sentiment analysis of the content posted by members and provides counseling based on specific emotions. For example, if a member is feeling stressed or anxious, it provides a real-time chat with an expert. Finally, the evaluations of the posted content are tallied and the display order is adjusted based on the evaluations. This allows high quality content to be preferentially displayed. In this way, it is possible to support the sound operation of the online community and improve user satisfaction. In this way, the online community moderator system can support the sound operation of the online community and improve user satisfaction.

[0029] The online community moderator system according to the embodiment includes a detection unit, a deletion unit, an identification unit, a blocking unit, a classification unit, a tagging unit, and a guide unit. The detection unit detects inappropriate content from among the content posted to the online community. For example, it detects content including discriminatory expressions or violent expressions. The detection unit automatically detects these contents using AI. The deletion unit deletes the inappropriate content detected by the detection unit. The deletion unit can also automatically delete the content using AI. The identification unit identifies whether the content posted to the online community is spam content or content including unnecessary links. The identification unit identifies these contents using a machine learning model. The blocking unit blocks the posting of the content identified by the identification unit as spam content or content including unnecessary links. The blocking unit can also automatically block the content using AI. The classification unit classifies the content posted to the online community into related topics. The classification unit can also classify the content using AI. The tagging unit tags the content classified by the classification unit. The tagging unit can also automatically tag the content using AI. The guidance unit provides the community guide and rules to new members of the online community. The guidance unit can also provide the guidance automatically using AI. In this way, the online community moderator system according to the embodiment can support the sound operation of the online community and improve user satisfaction.

[0030] The detection unit detects inappropriate content from among the content posted to the online community. For example, it detects content that includes discriminatory or violent expressions. The detection unit automatically detects these contents using AI. Specifically, it uses natural language processing (NLP) technology to analyze the content of the posted text. The AI ​​model is trained using a large data set and can identify discriminatory or violent expressions with high accuracy. For example, it determines whether the posted content is inappropriate by taking into account the frequency of occurrence of specific keywords and phrases, context, sentiment analysis, etc. In addition, it also analyzes image and video content using image recognition technology and computer vision to detect inappropriate content. This allows the detection unit to detect inappropriate content with high accuracy not only for text but also for multimedia content. In addition, the detection unit also has a function to accept reports and feedback from users, and when a user reports inappropriate content, it can update the AI ​​model based on that information and improve the detection accuracy. This allows the detection unit to support the sound operation of online communities and ensure the safety of users.

[0031] The deletion unit deletes the inappropriate content detected by the detection unit. The deletion unit can also automatically delete the content using AI. Specifically, the deletion unit receives information about the inappropriate content sent from the detection unit and immediately deletes the content. The deletion unit can determine the priority of deletion taking into account the type of content and the poster's past behavior history. For example, the content of a user who has posted inappropriate content multiple times in the past can be set to be deleted immediately. The deletion unit also has a function to save a log of deleted content so that the administrator can check it later. This ensures transparency of the deletion and allows the decision to delete to be reviewed if necessary. Furthermore, the deletion unit can automatically send a warning message to the poster for the deleted content to encourage them to prevent recurrence. The warning message includes the specific reason for deletion and the relevant part of the community guidelines, which can serve as a reference for users to review their own behavior. This allows the deletion unit to quickly and effectively delete inappropriate content and support the healthy operation of online communities.

[0032] The identification unit identifies whether or not content posted to an online community is spam content or content containing unnecessary links. The identification unit identifies such content using a machine learning model. Specifically, the posted content is analyzed using a model that has learned the characteristics of spam content. For example, whether or not the content is spam is determined based on the frequency of specific keywords, the number and type of links, the frequency of posting, and the like. In addition, the model is trained using data on past spam content so that it can respond to new spam techniques. The identification unit can analyze not only text but also links and metadata contained in images and videos, and identify spam content with high accuracy. Furthermore, the identification unit can identify users who are likely to be spammers by taking into account the user's behavior history and posting patterns, and strengthen monitoring. This allows the identification unit to prevent the spread of spam content and maintain the quality of the online community.

[0033] The blocking unit blocks the posting of content that is identified by the identification unit as spam content or content containing unnecessary links. The blocking unit can also automatically block using AI. Specifically, it receives information about spam content sent from the identification unit and blocks the content before it is posted. The blocking unit sends a warning message to the poster to prevent the recurrence of spam activity. The blocking unit also has a function to temporarily freeze the account of a user who attempts to post spam content, and can also permanently freeze the account if the spam activity continues. Furthermore, the blocking unit can learn the patterns of spam content and predict and prevent future spam activity. In this way, the blocking unit can effectively prevent the spread of spam content and support the healthy operation of online communities.

[0034] The classification unit classifies the content posted to the online community into related topics. The classification unit can also classify these contents using AI. Specifically, the classification unit uses natural language processing technology to analyze the posted content and classify it by topic. For example, the classification unit analyzes keywords, phrases, and context of the posted content and automatically classifies it into an appropriate topic. The classification unit improves the convenience of the online community by allowing users to easily find content related to topics of interest. In addition, the classification unit can also suggest optimal topics to individual users by taking into account the user's past posting history and browsing history. In this way, the classification unit can improve the user experience and promote the revitalization of the online community.

[0035] The tagging unit tags the content classified by the classification unit. The tagging unit can also automatically tag using AI. Specifically, it analyzes keywords and phrases in the posted content and automatically generates appropriate tags. The tagging unit improves the convenience of the online community by allowing users to easily search for related content. The tagging unit can also suggest optimal tags to individual users, taking into account the user's past tagging history and interests. This allows the tagging unit to improve the user experience and promote the revitalization of the online community. Furthermore, the tagging unit also has a function of referring to an existing tag list to maintain tag consistency, and obtaining approval from an administrator when a new tag is added. This prevents tags from becoming too numerous and keeps the online community organized.

[0036] The guidance unit provides the community guide and rules to new members of the online community. The guidance unit can also automatically provide guidance using AI. Specifically, when a new member joins the community, the guidance unit provides the community guidelines and rules along with a welcome message. The guidance unit uses a chatbot to answer the new member's questions in real time to deepen the new member's understanding of how to use the community and the rules. The guidance unit can also monitor the new member's behavior history and provide additional guidance and support as necessary. For example, when a new member posts for the first time, the guidance unit guides the new member on how to select appropriate topics and tags. The guidance unit also periodically notifies the new member of updates to the community guidelines, providing the new member with the latest information at all times. This allows the guidance unit to support the new member in smoothly adapting to the community and actively participating in it. Furthermore, the guidance unit can collect feedback from new members and use it to improve the community guidelines and guidance content. This allows the guidance unit to support the sound operation of the online community and improve user satisfaction.

[0037] The detection unit can detect each of the content including discriminatory expressions and the content including violent expressions as content determined to be inappropriate. The detection unit detects, for example, the content including discriminatory expressions. The discriminatory expressions include, but are not limited to, racism, sexism, and religious discrimination. The detection unit can also detect the content including violent expressions. The violent expressions include, but are not limited to, depictions of violent acts and threatening language. This makes it possible to effectively detect inappropriate content including discriminatory expressions and violent expressions. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the content including discriminatory expressions and violent expressions into an AI model, and the AI ​​model determines that the content is inappropriate.

[0038] The identification unit can use the machine learning model to identify whether the content posted to the online community is spam content and content including unnecessary links. The identification unit, for example, uses the machine learning model to identify spam content. The spam content includes, but is not limited to, advertising posts and unrelated links. The identification unit can also identify content including unnecessary links. The unnecessary links include, but are not limited to, links to phishing sites and links to unrelated external sites. As a result, the accuracy of identifying spam content and content including unnecessary links is improved by using the machine learning model. A part or all of the above-mentioned processing in the identification unit may be performed using AI, or may be performed without using AI. For example, the identification unit inputs content including spam content and unnecessary links into an AI model, and the AI ​​model identifies these contents.

[0039] The prediction unit can analyze the behavioral history of members of the online community and predict behavior that is determined to be inappropriate. The prediction unit, for example, collects behavioral history such as posting history, comment history, and browsing history of members, and analyzes this data. AI can be used for the analysis. For example, the prediction unit inputs the behavioral history using an AI model and predicts inappropriate behavior. In this way, by analyzing the behavioral history of members, inappropriate behavior can be predicted in advance and measures can be taken. A part or all of the above-mentioned processing in the prediction unit may be performed using AI, or may be performed without using AI. For example, the prediction unit inputs the behavioral history into an AI model, and the AI ​​model predicts inappropriate behavior.

[0040] The warning unit can warn of behavior that is judged to be inappropriate and predicted by the prediction unit. For example, the warning unit issues a warning against inappropriate behavior predicted by the prediction unit. AI can be used for the warning. For example, the warning unit predicts inappropriate behavior using an AI model and generates a warning message. As a result, by issuing a warning against the predicted inappropriate behavior, the behavior of the member can be improved. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit predicts inappropriate behavior using an AI model and generates a warning message.

[0041] The sentiment analysis unit can analyze the sentiment of the posted content of the members of the online community. For example, the sentiment analysis unit analyzes the posted content of the members and identifies the sentiment. The sentiment includes, but is not limited to, positive, negative, neutral, etc. AI can be used for the sentiment analysis. For example, the sentiment analysis unit inputs the posted content using an AI model and identifies the sentiment. This makes it possible to respond based on a specific sentiment by performing sentiment analysis of the posted content of the members. A part or all of the above-mentioned processing in the sentiment analysis unit may be performed using AI, or may be performed without using AI. For example, the sentiment analysis unit inputs the posted content to an AI model, and the AI ​​model identifies the sentiment.

[0042] The counseling unit can provide counseling to the member based on the specific emotion detected by the emotion analysis unit. The counseling unit provides counseling to the member based on, for example, the specific emotion detected by the emotion analysis unit. AI can be used for counseling. For example, the counseling unit inputs the specific emotion using an AI model and generates a counseling message. This makes it possible to provide counseling based on the specific emotion and provide mental support to the member. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs the specific emotion using an AI model and generates a counseling message.

[0043] The evaluation tallying unit can tally up evaluations of content posted in an online community. The evaluation tallying unit collects evaluation data such as user reviews, votes, and feedback, and tally up this data. AI can be used for the tallying. For example, the evaluation tallying unit inputs the evaluation data using an AI model and calculates an evaluation score. This allows the evaluations of the posted content to be tallied, so that high-quality content can be preferentially displayed. A part or all of the above-mentioned processing in the evaluation tallying unit may be performed using AI, or may be performed without using AI. For example, the evaluation tallying unit inputs the evaluation data into an AI model, and the AI ​​model calculates an evaluation score.

[0044] The display order adjustment unit can adjust the display order of the posted content based on the evaluations collected by the evaluation collection unit. The display order adjustment unit adjusts the display order of the posted content based on, for example, the evaluation scores collected by the evaluation collection unit. AI can be used for the adjustment. For example, the display order adjustment unit inputs the evaluation scores using an AI model and determines the display order. As a result, high-quality content is preferentially displayed by adjusting the display order based on the evaluations. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the evaluation scores into an AI model, and the AI ​​model determines the display order.

[0045] The detection unit can analyze the context of the content at the time of detection, and detect inappropriate content with higher accuracy. AI can be used for analyzing the context. For example, the detection unit inputs the context of the content using an AI model, and detects inappropriate expressions based on the context. The detection unit can also analyze the frequency of occurrence of keywords in the content, and identify inappropriate content based on the context. Furthermore, the detection unit can analyze the tone and nuance of the content, and detect inappropriate content based on the context. As a result, by analyzing the context of the content, it is possible to detect inappropriate content with higher accuracy. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the context data of the content into an AI model, and the AI ​​model detects inappropriate content.

[0046] The detection unit can optimize the detection algorithm by referring to the past detection history at the time of detection. The detection unit, for example, learns the pattern of inappropriate content detected in the past and optimizes the detection algorithm. AI can be used for optimization. For example, the detection unit inputs the past detection history using an AI model and optimizes the detection algorithm. The detection unit can also analyze the tendency of inappropriate content posted by a specific user based on the past detection history. Furthermore, the detection unit can also build a feedback loop for improving the accuracy of the detection algorithm by referring to the past detection history. As a result, the accuracy of the detection algorithm can be improved by referring to the past detection history. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the past detection history data into an AI model, and the AI ​​model optimizes the detection algorithm.

[0047] The deletion unit can evaluate the influence of the content at the time of deletion, and can delete the content with high influence preferentially. For example, the deletion unit preferentially deletes the content viewed by many users. AI can be used for the evaluation of the influence. For example, the deletion unit inputs the influence of the content using an AI model, and identifies the content with high influence. The deletion unit can also preferentially delete the inappropriate content posted by an influential user. Furthermore, the deletion unit can evaluate the diffusion degree of the content, and can also preferentially delete the content with high influence. In this way, by evaluating the influence of the content, it is possible to preferentially delete the inappropriate content with high influence. A part or all of the above-mentioned processing in the deletion unit may be performed using AI, or may be performed without using AI. For example, the deletion unit inputs the influence data of the content into an AI model, and the AI ​​model identifies the content with high influence.

[0048] The deletion unit can optimize the deletion algorithm by referring to the past deletion history at the time of deletion. For example, the deletion unit learns the pattern of inappropriate content deleted in the past and optimizes the deletion algorithm. AI can be used for optimization. For example, the deletion unit inputs the past deletion history using an AI model and optimizes the deletion algorithm. The deletion unit can also analyze the tendency of inappropriate content posted by a specific user based on the past deletion history. Furthermore, the deletion unit can also build a feedback loop for improving the accuracy of the deletion algorithm by referring to the past deletion history. As a result, the accuracy of the deletion algorithm can be improved by referring to the past deletion history. A part or all of the above-mentioned processing in the deletion unit may be performed using AI, or may be performed without using AI. For example, the deletion unit inputs the past deletion history data into an AI model, and the AI ​​model optimizes the deletion algorithm.

[0049] The identification unit can analyze the context of the content during identification and identify spam content with higher accuracy. The identification unit, for example, analyzes the context of the content and identifies spam content based on the context. AI can be used for analyzing the context. For example, the identification unit inputs the context of the content using an AI model and identifies spam content based on the context. The identification unit can also analyze the frequency of appearance of keywords in the content and identify spam content based on the context. Furthermore, the identification unit can analyze the tone and nuance of the content and identify spam content based on the context. As a result, by analyzing the context of the content, it is possible to identify spam content with higher accuracy. A part or all of the above-mentioned processing in the identification unit may be performed using AI, or may be performed without using AI. For example, the identification unit inputs the context data of the content to an AI model, and the AI ​​model identifies spam content.

[0050] The identification unit can optimize the identification algorithm by referring to the past identification history at the time of identification. For example, the identification unit learns the patterns of spam content identified in the past and optimizes the identification algorithm. AI can be used for the optimization. For example, the identification unit inputs the past identification history using an AI model and optimizes the identification algorithm. The identification unit can also analyze the tendency of spam content posted by a specific user based on the past identification history. Furthermore, the identification unit can also build a feedback loop for improving the accuracy of the identification algorithm by referring to the past identification history. As a result, the accuracy of the identification algorithm can be improved by referring to the past identification history. A part or all of the above-mentioned processing in the identification unit may be performed using AI, or may be performed without using AI. For example, the identification unit inputs the past identification history data into an AI model, and the AI ​​model optimizes the identification algorithm.

[0051] The blocking unit can evaluate the influence of the content at the time of blocking, and block the content with high influence preferentially. For example, the blocking unit blocks spam content viewed by many users preferentially. AI can be used for the evaluation of the influence. For example, the blocking unit inputs the influence of the content using an AI model, and identifies the content with high influence. The blocking unit can also preferentially block spam content posted by an influential user. Furthermore, the blocking unit can evaluate the diffusion degree of the content, and preferentially block spam content with high influence. In this way, by evaluating the influence of the content, it is possible to block spam content with high influence preferentially. A part or all of the above-mentioned processing in the blocking unit may be performed using AI, or may be performed without using AI. For example, the blocking unit inputs the influence data of the content into an AI model, and the AI ​​model identifies the content with high influence.

[0052] The blocking unit can optimize the blocking algorithm by referring to the past blocking history at the time of blocking. For example, the blocking unit learns the patterns of spam content blocked in the past and optimizes the blocking algorithm. AI can be used for optimization. For example, the blocking unit inputs the past blocking history using an AI model and optimizes the blocking algorithm. The blocking unit can also analyze the tendency of spam content posted by a specific user based on the past blocking history. Furthermore, the blocking unit can also build a feedback loop for improving the accuracy of the blocking algorithm by referring to the past blocking history. As a result, the accuracy of the blocking algorithm can be improved by referring to the past blocking history. A part or all of the above-mentioned processing in the blocking unit may be performed using AI, or may be performed without using AI. For example, the blocking unit inputs the past blocking history data into an AI model, and the AI ​​model optimizes the blocking algorithm.

[0053] The classification unit can analyze the context of the content at the time of classification and perform classification with higher accuracy. The classification unit, for example, analyzes the context of the content and classifies the content based on the context. AI can be used for analyzing the context. For example, the classification unit inputs the context of the content using an AI model and classifies the content based on the context. The classification unit can also analyze the frequency of appearance of keywords in the content and identify the content based on the context. Furthermore, the classification unit can analyze the tone and nuance of the content and classify the content based on the context. As a result, more accurate classification is possible by analyzing the context of the content. A part or all of the above-mentioned processing in the classification unit may be performed using AI, or may be performed without using AI. For example, the classification unit inputs the context data of the content to an AI model, and the AI ​​model classifies the content.

[0054] The classification unit can optimize the classification algorithm by referring to the past classification history at the time of classification. For example, the classification unit learns the patterns of the contents classified in the past and optimizes the classification algorithm. AI can be used for optimization. For example, the classification unit inputs the past classification history using an AI model and optimizes the classification algorithm. The classification unit can also analyze the tendency of the contents posted by a specific user based on the past classification history. Furthermore, the classification unit can also build a feedback loop for improving the accuracy of the classification algorithm by referring to the past classification history. As a result, the accuracy of the classification algorithm can be improved by referring to the past classification history. A part or all of the above-mentioned processing in the classification unit may be performed using AI, or may be performed without using AI. For example, the classification unit inputs the past classification history data into an AI model, and the AI ​​model optimizes the classification algorithm.

[0055] The tagging unit can analyze the context of the content at the time of tagging and perform tagging with higher accuracy. The tagging unit, for example, analyzes the context of the content and performs tagging based on the context. AI can be used for analyzing the context. For example, the tagging unit inputs the context of the content using an AI model and performs tagging based on the context. The tagging unit can also analyze the frequency of appearance of keywords in the content and perform tagging based on the context. Furthermore, the tagging unit can analyze the tone and nuance of the content and perform tagging based on the context. This enables tagging with higher accuracy by analyzing the context of the content. A part or all of the above-mentioned processing in the tagging unit may be performed using AI, or may be performed without using AI. For example, the tagging unit inputs the context data of the content to an AI model, and the AI ​​model performs tagging.

[0056] The tagging unit can optimize the tagging algorithm by referring to the past tagging history at the time of tagging. For example, the tagging unit learns the patterns of the content tagged in the past and optimizes the tagging algorithm. AI can be used for optimization. For example, the tagging unit inputs the past tagging history using an AI model and optimizes the tagging algorithm. The tagging unit can also analyze the tendency of the content posted by a specific user based on the past tagging history. Furthermore, the tagging unit can also build a feedback loop for improving the accuracy of the tagging algorithm by referring to the past tagging history. As a result, the accuracy of the tagging algorithm can be improved by referring to the past tagging history. A part or all of the above-mentioned processing in the tagging unit may be performed using AI or may be performed without using AI. For example, the tagging unit inputs the past tagging history data into an AI model, and the AI ​​model optimizes the tagging algorithm.

[0057] The guidance unit can provide optimal guidance by referring to the user's past behavior history when providing guidance. AI can be used to refer to the behavior history. For example, the guidance unit inputs the past behavior history using an AI model to generate optimal guidance. The guidance unit can also predict the required guidance content from the user's past behavior history. Furthermore, the guidance unit can analyze the user's past behavior history and provide the most effective guidance method. As a result, optimal guidance can be provided by referring to the user's past behavior history. A part or all of the above-mentioned processing in the guidance unit may be performed using AI, or may be performed without using AI. For example, the guidance unit inputs past behavior history data into an AI model, and the AI ​​model generates optimal guidance.

[0058] The guidance unit can provide customized guidance based on the user's field of interest at the time of guidance. For example, the guidance unit preferentially guides information related to the user's field of interest. AI can be used to identify the field of interest. For example, the guidance unit inputs the user's field of interest using an AI model and generates customized guidance. The guidance unit can also provide customized guidance content based on the user's field of interest. Furthermore, the guidance unit can analyze the user's field of interest and provide an optimal guidance method. As a result, more effective guidance can be provided by providing customized guidance based on the user's field of interest. A part or all of the above-mentioned processing in the guidance unit may be performed using AI, or may be performed without using AI. For example, the guidance unit inputs the user's field of interest data into an AI model, and the AI ​​model generates customized guidance.

[0059] The prediction unit can optimize the prediction algorithm by referring to the past behavior history at the time of prediction. The prediction unit, for example, learns the pattern of inappropriate behavior predicted in the past and optimizes the prediction algorithm. AI can be used for optimization. For example, the prediction unit inputs the past behavior history using an AI model and optimizes the prediction algorithm. The prediction unit can also analyze the tendency of inappropriate behavior performed by a specific user based on the past behavior history. Furthermore, the prediction unit can also build a feedback loop for improving the accuracy of the prediction algorithm by referring to the past behavior history. As a result, the accuracy of the prediction algorithm can be improved by referring to the past behavior history. A part or all of the above-mentioned processing in the prediction unit may be performed using AI, or may be performed without using AI. For example, the prediction unit inputs the past behavior history data into an AI model, and the AI ​​model optimizes the prediction algorithm.

[0060] The prediction unit can analyze the context of the content at the time of prediction and predict inappropriate behavior with higher accuracy. The prediction unit, for example, analyzes the context of the content and predicts inappropriate behavior based on the context. AI can be used for analyzing the context. For example, the prediction unit inputs the context of the content using an AI model and predicts inappropriate behavior based on the context. The prediction unit can also analyze the frequency of occurrence of keywords in the content and identify inappropriate behavior based on the context. Furthermore, the prediction unit can analyze the tone and nuance of the content and predict inappropriate behavior based on the context. As a result, by analyzing the context of the content, it is possible to predict inappropriate behavior with higher accuracy. A part or all of the above-mentioned processing in the prediction unit may be performed using AI, or may be performed without using AI. For example, the prediction unit inputs context data of the content into an AI model, and the AI ​​model predicts inappropriate behavior.

[0061] The warning unit can optimize the warning algorithm by referring to the past warning history at the time of warning. For example, the warning unit learns the pattern of behavior that has been warned in the past and optimizes the warning algorithm. AI can be used for optimization. For example, the warning unit inputs the past warning history using an AI model and optimizes the warning algorithm. The warning unit can also analyze the tendency of inappropriate behavior performed by a specific user based on the past warning history. Furthermore, the warning unit can also build a feedback loop for improving the accuracy of the warning algorithm by referring to the past warning history. As a result, the accuracy of the warning algorithm can be improved by referring to the past warning history. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit inputs past warning history data into an AI model, and the AI ​​model optimizes the warning algorithm.

[0062] The warning unit can analyze the user's behavior history and select the optimal warning timing at the time of warning. For example, the warning unit analyzes the timing of inappropriate behavior performed by the user in the past and selects the optimal warning timing. AI can be used to analyze the behavior history. For example, the warning unit inputs past behavior history using an AI model and selects the optimal warning timing. The warning unit can also adjust the warning timing based on the user's behavior history. Furthermore, the warning unit can analyze the user's behavior pattern and select the most effective warning timing. As a result, a warning can be issued at the optimal timing by analyzing the user's behavior history. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit inputs past behavior history data into an AI model, and the AI ​​model selects the optimal warning timing.

[0063] The sentiment analysis unit can analyze the context of the content during sentiment analysis and perform sentiment analysis with higher accuracy. The sentiment analysis unit, for example, analyzes the context of the content and performs sentiment analysis based on the context. AI can be used for analyzing the context. For example, the sentiment analysis unit inputs the context of the content using an AI model and performs sentiment analysis based on the context. The sentiment analysis unit can also analyze the frequency of appearance of keywords in the content and identify emotions based on the context. Furthermore, the sentiment analysis unit can analyze the tone and nuance of the content and perform sentiment analysis based on the context. This enables more accurate sentiment analysis by analyzing the context of the content. A part or all of the above-mentioned processing in the sentiment analysis unit may be performed using AI, or may be performed without using AI. For example, the sentiment analysis unit inputs context data of the content into an AI model, and the AI ​​model identifies emotions.

[0064] The sentiment analysis unit can optimize the sentiment analysis algorithm by referring to the past sentiment analysis history during sentiment analysis. The sentiment analysis unit, for example, learns the patterns of emotions analyzed in the past and optimizes the sentiment analysis algorithm. AI can be used for optimization. For example, the sentiment analysis unit inputs the past sentiment analysis history using an AI model and optimizes the sentiment analysis algorithm. The sentiment analysis unit can also analyze the tendency of emotions expressed by a specific user based on the past sentiment analysis history. Furthermore, the sentiment analysis unit can also refer to the past sentiment analysis history and build a feedback loop for improving the accuracy of the sentiment analysis algorithm. As a result, the accuracy of the sentiment analysis algorithm can be improved by referring to the past sentiment analysis history. A part or all of the above-mentioned processing in the sentiment analysis unit may be performed using AI, or may be performed without using AI. For example, the sentiment analysis unit inputs the past sentiment analysis history data into an AI model, and the AI ​​model optimizes the sentiment analysis algorithm.

[0065] The counseling unit can optimize the counseling algorithm by referring to the past counseling history during counseling. The counseling unit, for example, learns the patterns of counseling provided in the past and optimizes the counseling algorithm. AI can be used for optimization. For example, the counseling unit inputs the past counseling history using an AI model and optimizes the counseling algorithm. The counseling unit can also analyze counseling trends for a specific user based on the past counseling history. Furthermore, the counseling unit can also build a feedback loop for improving the accuracy of the counseling algorithm by referring to the past counseling history. As a result, the accuracy of the counseling algorithm can be improved by referring to the past counseling history. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs past counseling history data into an AI model, and the AI ​​model optimizes the counseling algorithm.

[0066] The counseling unit can analyze the user's behavior history during counseling and select the optimal counseling timing. The counseling unit, for example, analyzes the timing of actions taken by the user in the past and selects the optimal counseling timing. AI can be used to analyze the behavior history. For example, the counseling unit inputs the past behavior history using an AI model and selects the optimal counseling timing. The counseling unit can also adjust the timing of counseling based on the user's behavior history. Furthermore, the counseling unit can analyze the user's behavior pattern and select the most effective counseling timing. As a result, counseling can be provided at the optimal timing by analyzing the user's behavior history. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs past behavior history data into an AI model, and the AI ​​model selects the optimal counseling timing.

[0067] The evaluation aggregation unit can analyze the context of the content when aggregating the evaluations, and perform more accurate evaluation aggregation. The evaluation aggregation unit, for example, analyzes the context of the content and performs evaluation aggregation based on the context. AI can be used for analyzing the context. For example, the evaluation aggregation unit inputs the context of the content using an AI model, and performs evaluation aggregation based on the context. The evaluation aggregation unit can also analyze the frequency of appearance of keywords in the content, and identify the evaluation based on the context. Furthermore, the evaluation aggregation unit can analyze the tone and nuance of the content, and perform evaluation aggregation based on the context. As a result, more accurate evaluation aggregation is possible by analyzing the context of the content. A part or all of the above-mentioned processing in the evaluation aggregation unit may be performed using AI, or may be performed without using AI. For example, the evaluation aggregation unit inputs the context data of the content into an AI model, and the AI ​​model identifies the evaluation.

[0068] The evaluation aggregation unit can optimize the evaluation aggregation algorithm by referring to the past evaluation aggregation history when collating the evaluations. The evaluation aggregation unit, for example, learns the patterns of evaluations collected in the past and optimizes the evaluation aggregation algorithm. AI can be used for the optimization. For example, the evaluation aggregation unit inputs the past evaluation aggregation history using an AI model and optimizes the evaluation aggregation algorithm. The evaluation aggregation unit can also analyze the tendency of evaluations made by a specific user based on the past evaluation aggregation history. Furthermore, the evaluation aggregation unit can also refer to the past evaluation aggregation history and build a feedback loop for improving the accuracy of the evaluation aggregation algorithm. As a result, the accuracy of the evaluation aggregation algorithm can be improved by referring to the past evaluation aggregation history. A part or all of the above-mentioned processing in the evaluation aggregation unit may be performed using AI, or may be performed without using AI. For example, the evaluation aggregation unit inputs the past evaluation aggregation history data into an AI model, and the AI ​​model optimizes the evaluation aggregation algorithm.

[0069] The display order adjustment unit can analyze the context of the content when adjusting the display order, and adjust the display order more accurately. The display order adjustment unit, for example, analyzes the context of the content, and adjusts the display order based on the context. AI can be used for analyzing the context. For example, the display order adjustment unit inputs the context of the content using an AI model, and adjusts the display order based on the context. The display order adjustment unit can also analyze the frequency of appearance of keywords in the content, and adjust the display order based on the context. Furthermore, the display order adjustment unit can analyze the tone and nuance of the content, and adjust the display order based on the context. As a result, by analyzing the context of the content, it is possible to adjust the display order more accurately. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the context data of the content into an AI model, and the AI ​​model adjusts the display order.

[0070] The display order adjustment unit can optimize the display order adjustment algorithm by referring to the past display order adjustment history when adjusting the display order. The display order adjustment unit, for example, learns the pattern of the display order adjusted in the past and optimizes the display order adjustment algorithm. AI can be used for optimization. For example, the display order adjustment unit inputs the past display order adjustment history using an AI model and optimizes the display order adjustment algorithm. The display order adjustment unit can also analyze the tendency of the content preferred by a specific user based on the past display order adjustment history. Furthermore, the display order adjustment unit can also build a feedback loop for improving the accuracy of the display order adjustment algorithm by referring to the past display order adjustment history. As a result, the accuracy of the display order adjustment algorithm can be improved by referring to the past display order adjustment history. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the past display order adjustment history data into an AI model, and the AI ​​model optimizes the display order adjustment algorithm.

[0071] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0072] The online community moderator system may further include a personalized feedback unit based on the user's behavioral history. The personalized feedback unit analyzes the user's past behavioral history and provides individual feedback. For example, if the user has made an inappropriate post in the past, specific improvements to that behavior may be presented. Also, if the user has made a positive contribution, a message praising the contribution may be sent. Furthermore, the user's behavioral patterns may be analyzed and advice for improving future behavior may be provided. This allows the user to reflect on their own behavior and aim for better community participation.

[0073] The identification unit can have a content evaluation function for the contents posted by users. For example, it can not only evaluate whether the posted contents are spam, but also evaluate the quality of the contents. AI can be used for the evaluation. For example, the identification unit inputs the posted contents using an AI model and evaluates the quality of the contents. It can also provide feedback to the user based on the evaluation results. Furthermore, it can adjust the display order of the posted contents based on the evaluation results. This allows high-quality content to be displayed preferentially.

[0074] The prediction unit can not only predict future behavior based on the user's behavioral history, but also analyze the motivation behind the behavior. For example, if a user frequently posts inappropriate content on a specific topic, the motivation behind the posting can be identified. AI can be used to analyze the motivation. For example, the prediction unit inputs the behavioral history using an AI model to identify the motivation. In addition, the prediction unit can provide appropriate feedback and advice to the user based on the motivation. This can fundamentally improve the user's behavior.

[0075] The warning unit can customize the contents of the warning based on the user's behavior history. For example, a stricter warning can be issued to a user who has engaged in similar inappropriate behavior in the past. Also, a gentler warning can be issued to a user who has engaged in inappropriate behavior for the first time. Furthermore, the warning unit can analyze the user's behavioral patterns and select the most effective warning method. This can effectively improve the user's behavior.

[0076] The evaluation aggregation unit can evaluate the reliability of the evaluations based on the user's evaluation history. For example, it can give more weight to evaluations from users who have consistently given high ratings in the past. It can also disregard evaluations from users who have given biased ratings in the past. It can also analyze the evaluation history to identify the tendency of evaluations made by a particular user. This improves the reliability of the evaluations and enables more accurate evaluation aggregation.

[0077] The display order adjustment unit can customize the display order based on the user's browsing history. For example, it can preferentially display content related to topics that the user has frequently browsed in the past. It can also preferentially display content related to specific keywords in which the user is interested. It can also analyze the user's browsing history and display content that is most likely to attract the user's attention. This allows the display order to be adjusted based on the user's interests, making it possible to provide content more effectively.

[0078] The process flow of the first embodiment will be briefly described below.

[0079] Step 1: The detection unit detects inappropriate content from among the content posted to the online community. For example, it detects content that contains discriminatory or violent language. The detection unit automatically detects this content using AI. Step 2: The deletion unit deletes the inappropriate content detected by the detection unit. The deletion unit can also perform the deletion automatically using AI. Step 3: The identification unit identifies whether the content posted to the online community is spam content or content containing unwanted links. The identification unit identifies such content using a machine learning model. Step 4: The blocking unit blocks the posting of content that is identified by the identification unit as spam content or content containing unnecessary links. The blocking unit can also perform blocking automatically using AI. Step 5: The classifier categorizes the content posted to the online community into relevant topics. The classifier can also use AI to categorize these contents. Step 6: The tagger tagges the content classified by the classifier. The tagger can also tag automatically using AI. Step 7: The information section provides the new members of the online community with the community guide and rules. The information section can also provide automatic guidance using AI.

[0080] (Example 2) The online community moderator system according to an embodiment of the present invention is a system that detects and deletes inappropriate content, takes measures against spam, classifies and tags topics, and provides guidance to users. This system detects inappropriate content from content posted to an online community. For example, it detects content that includes discriminatory or violent expressions. This detection is performed by AI, and the content is automatically deleted without waiting for a report by the user. Next, as a measure against spam, spam content and unnecessary links are identified and the posting of these contents is blocked. The identification is performed using a machine learning model to keep the community clean. Furthermore, the system improves navigation by classifying posts into related topics and attaching correct tags. This allows users to quickly find the information they need. In addition, the system provides new members with a community guide and a guide to the rules to support smooth participation. This allows new members to understand the rules of the community and participate appropriately. Furthermore, the system analyzes the behavioral history of members of the online community and predicts inappropriate behavior. It issues warnings for predicted inappropriate behavior, and provides positive feedback when improved behavior is confirmed. In addition, it performs sentiment analysis of the content posted by members and provides counseling based on specific emotions. For example, if a member is feeling stressed or anxious, it provides a real-time chat with an expert. Finally, the evaluations of the posted content are tallied and the display order is adjusted based on the evaluations. This allows high quality content to be preferentially displayed. In this way, it is possible to support the sound operation of the online community and improve user satisfaction. In this way, the online community moderator system can support the sound operation of the online community and improve user satisfaction.

[0081] The online community moderator system according to the embodiment includes a detection unit, a deletion unit, an identification unit, a blocking unit, a classification unit, a tagging unit, and a guide unit. The detection unit detects inappropriate content from among the content posted to the online community. For example, it detects content including discriminatory expressions or violent expressions. The detection unit automatically detects these contents using AI. The deletion unit deletes the inappropriate content detected by the detection unit. The deletion unit can also automatically delete the content using AI. The identification unit identifies whether the content posted to the online community is spam content or content including unnecessary links. The identification unit identifies these contents using a machine learning model. The blocking unit blocks the posting of the content identified by the identification unit as spam content or content including unnecessary links. The blocking unit can also automatically block the content using AI. The classification unit classifies the content posted to the online community into related topics. The classification unit can also classify the content using AI. The tagging unit tags the content classified by the classification unit. The tagging unit can also automatically tag the content using AI. The guidance unit provides the community guide and rules to new members of the online community. The guidance unit can also provide the guidance automatically using AI. In this way, the online community moderator system according to the embodiment can support the sound operation of the online community and improve user satisfaction.

[0082] The detection unit detects inappropriate content from among the content posted to the online community. For example, it detects content that includes discriminatory or violent expressions. The detection unit automatically detects these contents using AI. Specifically, it uses natural language processing (NLP) technology to analyze the content of the posted text. The AI ​​model is trained using a large data set and can identify discriminatory or violent expressions with high accuracy. For example, it determines whether the posted content is inappropriate by taking into account the frequency of occurrence of specific keywords and phrases, context, sentiment analysis, etc. In addition, it also analyzes image and video content using image recognition technology and computer vision to detect inappropriate content. This allows the detection unit to detect inappropriate content with high accuracy not only for text but also for multimedia content. In addition, the detection unit also has a function to accept reports and feedback from users, and when a user reports inappropriate content, it can update the AI ​​model based on that information and improve the detection accuracy. This allows the detection unit to support the sound operation of online communities and ensure the safety of users.

[0083] The deletion unit deletes the inappropriate content detected by the detection unit. The deletion unit can also automatically delete the content using AI. Specifically, the deletion unit receives information about the inappropriate content sent from the detection unit and immediately deletes the content. The deletion unit can determine the priority of deletion taking into account the type of content and the poster's past behavior history. For example, the content of a user who has posted inappropriate content multiple times in the past can be set to be deleted immediately. The deletion unit also has a function to save a log of deleted content so that the administrator can check it later. This ensures transparency of the deletion and allows the decision to delete to be reviewed if necessary. Furthermore, the deletion unit can automatically send a warning message to the poster for the deleted content to encourage them to prevent recurrence. The warning message includes the specific reason for deletion and the relevant part of the community guidelines, which can serve as a reference for users to review their own behavior. This allows the deletion unit to quickly and effectively delete inappropriate content and support the healthy operation of online communities.

[0084] The identification unit identifies whether or not content posted to an online community is spam content or content containing unnecessary links. The identification unit identifies such content using a machine learning model. Specifically, the posted content is analyzed using a model that has learned the characteristics of spam content. For example, whether or not the content is spam is determined based on the frequency of specific keywords, the number and type of links, the frequency of posting, and the like. In addition, the model is trained using data on past spam content so that it can respond to new spam techniques. The identification unit can analyze not only text but also links and metadata contained in images and videos, and identify spam content with high accuracy. Furthermore, the identification unit can identify users who are likely to be spammers by taking into account the user's behavior history and posting patterns, and strengthen monitoring. This allows the identification unit to prevent the spread of spam content and maintain the quality of the online community.

[0085] The blocking unit blocks the posting of content that is identified by the identification unit as spam content or content containing unnecessary links. The blocking unit can also automatically block using AI. Specifically, it receives information about spam content sent from the identification unit and blocks the content before it is posted. The blocking unit sends a warning message to the poster to prevent the recurrence of spam activity. The blocking unit also has a function to temporarily freeze the account of a user who attempts to post spam content, and can also permanently freeze the account if the spam activity continues. Furthermore, the blocking unit can learn the patterns of spam content and predict and prevent future spam activity. In this way, the blocking unit can effectively prevent the spread of spam content and support the healthy operation of online communities.

[0086] The classification unit classifies the content posted to the online community into related topics. The classification unit can also classify these contents using AI. Specifically, the classification unit uses natural language processing technology to analyze the posted content and classify it by topic. For example, the classification unit analyzes keywords, phrases, and context of the posted content and automatically classifies it into an appropriate topic. The classification unit improves the convenience of the online community by allowing users to easily find content related to topics of interest. In addition, the classification unit can also suggest optimal topics to individual users by taking into account the user's past posting history and browsing history. In this way, the classification unit can improve the user experience and promote the revitalization of the online community.

[0087] The tagging unit tags the content classified by the classification unit. The tagging unit can also automatically tag using AI. Specifically, it analyzes keywords and phrases in the posted content and automatically generates appropriate tags. The tagging unit improves the convenience of the online community by allowing users to easily search for related content. The tagging unit can also suggest optimal tags to individual users, taking into account the user's past tagging history and interests. This allows the tagging unit to improve the user experience and promote the revitalization of the online community. Furthermore, the tagging unit also has a function of referring to an existing tag list to maintain tag consistency, and obtaining approval from an administrator when a new tag is added. This prevents tags from becoming too numerous and keeps the online community organized.

[0088] The guidance unit provides the community guide and rules to new members of the online community. The guidance unit can also automatically provide guidance using AI. Specifically, when a new member joins the community, the guidance unit provides the community guidelines and rules along with a welcome message. The guidance unit uses a chatbot to answer the new member's questions in real time to deepen the new member's understanding of how to use the community and the rules. The guidance unit can also monitor the new member's behavior history and provide additional guidance and support as necessary. For example, when a new member posts for the first time, the guidance unit guides the new member on how to select appropriate topics and tags. The guidance unit also periodically notifies the new member of updates to the community guidelines, providing the new member with the latest information at all times. This allows the guidance unit to support the new member in smoothly adapting to the community and actively participating in it. Furthermore, the guidance unit can collect feedback from new members and use it to improve the community guidelines and guidance content. This allows the guidance unit to support the sound operation of the online community and improve user satisfaction.

[0089] The detection unit can detect each of the content including discriminatory expressions and the content including violent expressions as content determined to be inappropriate. The detection unit detects, for example, the content including discriminatory expressions. The discriminatory expressions include, but are not limited to, racism, sexism, and religious discrimination. The detection unit can also detect the content including violent expressions. The violent expressions include, but are not limited to, depictions of violent acts and threatening language. This makes it possible to effectively detect inappropriate content including discriminatory expressions and violent expressions. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the content including discriminatory expressions and violent expressions into an AI model, and the AI ​​model determines that the content is inappropriate.

[0090] The identification unit can use the machine learning model to identify whether the content posted to the online community is spam content and content including unnecessary links. The identification unit, for example, uses the machine learning model to identify spam content. The spam content includes, but is not limited to, advertising posts and unrelated links. The identification unit can also identify content including unnecessary links. The unnecessary links include, but are not limited to, links to phishing sites and links to unrelated external sites. As a result, the accuracy of identifying spam content and content including unnecessary links is improved by using the machine learning model. A part or all of the above-mentioned processing in the identification unit may be performed using AI, or may be performed without using AI. For example, the identification unit inputs content including spam content and unnecessary links into an AI model, and the AI ​​model identifies these contents.

[0091] The prediction unit can analyze the behavioral history of members of the online community and predict behavior that is determined to be inappropriate. The prediction unit, for example, collects behavioral history such as posting history, comment history, and browsing history of members, and analyzes this data. AI can be used for the analysis. For example, the prediction unit inputs the behavioral history using an AI model and predicts inappropriate behavior. In this way, by analyzing the behavioral history of members, inappropriate behavior can be predicted in advance and measures can be taken. A part or all of the above-mentioned processing in the prediction unit may be performed using AI, or may be performed without using AI. For example, the prediction unit inputs the behavioral history into an AI model, and the AI ​​model predicts inappropriate behavior.

[0092] The warning unit can warn of behavior that is judged to be inappropriate and predicted by the prediction unit. For example, the warning unit issues a warning against inappropriate behavior predicted by the prediction unit. AI can be used for the warning. For example, the warning unit predicts inappropriate behavior using an AI model and generates a warning message. As a result, by issuing a warning against the predicted inappropriate behavior, the behavior of the member can be improved. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit predicts inappropriate behavior using an AI model and generates a warning message.

[0093] The sentiment analysis unit can analyze the sentiment of the posted content of the members of the online community. For example, the sentiment analysis unit analyzes the posted content of the members and identifies the sentiment. The sentiment includes, but is not limited to, positive, negative, neutral, etc. AI can be used for the sentiment analysis. For example, the sentiment analysis unit inputs the posted content using an AI model and identifies the sentiment. This makes it possible to respond based on a specific sentiment by performing sentiment analysis of the posted content of the members. A part or all of the above-mentioned processing in the sentiment analysis unit may be performed using AI, or may be performed without using AI. For example, the sentiment analysis unit inputs the posted content to an AI model, and the AI ​​model identifies the sentiment.

[0094] The counseling unit can provide counseling to the member based on the specific emotion detected by the emotion analysis unit. The counseling unit provides counseling to the member based on, for example, the specific emotion detected by the emotion analysis unit. AI can be used for counseling. For example, the counseling unit inputs the specific emotion using an AI model and generates a counseling message. This makes it possible to provide counseling based on the specific emotion and provide mental support to the member. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs the specific emotion using an AI model and generates a counseling message.

[0095] The evaluation tallying unit can tally up evaluations of content posted in an online community. The evaluation tallying unit collects evaluation data such as user reviews, votes, and feedback, and tally up this data. AI can be used for the tallying. For example, the evaluation tallying unit inputs the evaluation data using an AI model and calculates an evaluation score. This allows the evaluations of the posted content to be tallied, so that high-quality content can be preferentially displayed. A part or all of the above-mentioned processing in the evaluation tallying unit may be performed using AI, or may be performed without using AI. For example, the evaluation tallying unit inputs the evaluation data into an AI model, and the AI ​​model calculates an evaluation score.

[0096] The display order adjustment unit can adjust the display order of the posted content based on the evaluations collected by the evaluation collection unit. The display order adjustment unit adjusts the display order of the posted content based on, for example, the evaluation scores collected by the evaluation collection unit. AI can be used for the adjustment. For example, the display order adjustment unit inputs the evaluation scores using an AI model and determines the display order. As a result, high-quality content is preferentially displayed by adjusting the display order based on the evaluations. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the evaluation scores into an AI model, and the AI ​​model determines the display order.

[0097] The detection unit can estimate the user's emotion and adjust the detection criteria for inappropriate content based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the detection unit detects inappropriate content with stricter criteria when the user is stressed. Also, when the user is relaxed, the detection unit can detect inappropriate content with normal criteria. Furthermore, when the user is excited, the detection unit can preferentially detect content including emotional expressions. This allows more appropriate detection by adjusting the detection criteria for inappropriate content based on the user's emotion. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the detection criteria for inappropriate content.

[0098] The detection unit can analyze the context of the content at the time of detection, and detect inappropriate content with higher accuracy. AI can be used for analyzing the context. For example, the detection unit inputs the context of the content using an AI model, and detects inappropriate expressions based on the context. The detection unit can also analyze the frequency of occurrence of keywords in the content, and identify inappropriate content based on the context. Furthermore, the detection unit can analyze the tone and nuance of the content, and detect inappropriate content based on the context. As a result, by analyzing the context of the content, it is possible to detect inappropriate content with higher accuracy. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the context data of the content into an AI model, and the AI ​​model detects inappropriate content.

[0099] The detection unit can optimize the detection algorithm by referring to the past detection history at the time of detection. The detection unit, for example, learns the pattern of inappropriate content detected in the past and optimizes the detection algorithm. AI can be used for optimization. For example, the detection unit inputs the past detection history using an AI model and optimizes the detection algorithm. The detection unit can also analyze the tendency of inappropriate content posted by a specific user based on the past detection history. Furthermore, the detection unit can also build a feedback loop for improving the accuracy of the detection algorithm by referring to the past detection history. As a result, the accuracy of the detection algorithm can be improved by referring to the past detection history. A part or all of the above-mentioned processing in the detection unit may be performed using AI, or may be performed without using AI. For example, the detection unit inputs the past detection history data into an AI model, and the AI ​​model optimizes the detection algorithm.

[0100] The deletion unit can estimate the user's emotion and adjust the timing of deletion based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the deletion unit can delete inappropriate content immediately when the user is stressed. In addition, when the user is relaxed, the deletion unit can delete inappropriate content at a normal timing. Furthermore, when the user is excited, the deletion unit can delete inappropriate content before it causes an emotional reaction. In this way, by adjusting the timing of deletion based on the user's emotion, inappropriate content can be deleted at a more appropriate timing. A part or all of the above-mentioned processing in the deletion unit may be performed using AI, or may be performed without using AI. For example, the deletion unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the timing of deletion.

[0101] The deletion unit can evaluate the influence of the content at the time of deletion, and can delete the content with high influence preferentially. For example, the deletion unit preferentially deletes the content viewed by many users. AI can be used for the evaluation of the influence. For example, the deletion unit inputs the influence of the content using an AI model, and identifies the content with high influence. The deletion unit can also preferentially delete the inappropriate content posted by an influential user. Furthermore, the deletion unit can evaluate the diffusion degree of the content, and can also preferentially delete the content with high influence. In this way, by evaluating the influence of the content, it is possible to preferentially delete the inappropriate content with high influence. A part or all of the above-mentioned processing in the deletion unit may be performed using AI, or may be performed without using AI. For example, the deletion unit inputs the influence data of the content into an AI model, and the AI ​​model identifies the content with high influence.

[0102] The deletion unit can optimize the deletion algorithm by referring to the past deletion history at the time of deletion. For example, the deletion unit learns the pattern of inappropriate content deleted in the past and optimizes the deletion algorithm. AI can be used for optimization. For example, the deletion unit inputs the past deletion history using an AI model and optimizes the deletion algorithm. The deletion unit can also analyze the tendency of inappropriate content posted by a specific user based on the past deletion history. Furthermore, the deletion unit can also build a feedback loop for improving the accuracy of the deletion algorithm by referring to the past deletion history. As a result, the accuracy of the deletion algorithm can be improved by referring to the past deletion history. A part or all of the above-mentioned processing in the deletion unit may be performed using AI, or may be performed without using AI. For example, the deletion unit inputs the past deletion history data into an AI model, and the AI ​​model optimizes the deletion algorithm.

[0103] The identification unit can estimate the user's emotions and adjust the identification criteria for spam content based on the estimated user's emotions. An emotion engine or a generative AI can be used to estimate emotions. The generative AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the identification unit can identify spam content with stricter criteria when the user is stressed. Also, when the user is relaxed, the identification unit can identify spam content with normal criteria. Furthermore, when the user is excited, the identification unit can preferentially identify spam content that is likely to cause an emotional response. This allows for more appropriate identification by adjusting the identification criteria for spam content based on the user's emotions. Part or all of the above-mentioned processing in the identification unit may be performed using AI or may be performed without using AI. For example, the identification unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the identification criteria for spam content.

[0104] The identification unit can analyze the context of the content during identification and identify spam content with higher accuracy. The identification unit, for example, analyzes the context of the content and identifies spam content based on the context. AI can be used for analyzing the context. For example, the identification unit inputs the context of the content using an AI model and identifies spam content based on the context. The identification unit can also analyze the frequency of appearance of keywords in the content and identify spam content based on the context. Furthermore, the identification unit can analyze the tone and nuance of the content and identify spam content based on the context. As a result, by analyzing the context of the content, it is possible to identify spam content with higher accuracy. A part or all of the above-mentioned processing in the identification unit may be performed using AI, or may be performed without using AI. For example, the identification unit inputs the context data of the content to an AI model, and the AI ​​model identifies spam content.

[0105] The identification unit can optimize the identification algorithm by referring to the past identification history at the time of identification. For example, the identification unit learns the patterns of spam content identified in the past and optimizes the identification algorithm. AI can be used for the optimization. For example, the identification unit inputs the past identification history using an AI model and optimizes the identification algorithm. The identification unit can also analyze the tendency of spam content posted by a specific user based on the past identification history. Furthermore, the identification unit can also build a feedback loop for improving the accuracy of the identification algorithm by referring to the past identification history. As a result, the accuracy of the identification algorithm can be improved by referring to the past identification history. A part or all of the above-mentioned processing in the identification unit may be performed using AI, or may be performed without using AI. For example, the identification unit inputs the past identification history data into an AI model, and the AI ​​model optimizes the identification algorithm.

[0106] The blocking unit can estimate the user's emotions and adjust the timing of the block based on the estimated user's emotions. An emotion engine or a generation AI can be used to estimate the emotions. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the blocking unit immediately blocks spam content when the user is stressed. Also, when the user is relaxed, the blocking unit can block spam content at a normal timing. Furthermore, when the user is excited, the blocking unit can block spam content before it causes an emotional reaction. This allows the spam content to be blocked at a more appropriate timing by adjusting the timing of the block based on the user's emotions. A part or all of the above-mentioned processing in the blocking unit may be performed using AI, or may be performed without using AI. For example, the blocking unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the timing of the block.

[0107] The blocking unit can evaluate the influence of the content at the time of blocking, and block the content with high influence preferentially. For example, the blocking unit blocks spam content viewed by many users preferentially. AI can be used for the evaluation of the influence. For example, the blocking unit inputs the influence of the content using an AI model, and identifies the content with high influence. The blocking unit can also preferentially block spam content posted by an influential user. Furthermore, the blocking unit can evaluate the diffusion degree of the content, and preferentially block spam content with high influence. In this way, by evaluating the influence of the content, it is possible to block spam content with high influence preferentially. A part or all of the above-mentioned processing in the blocking unit may be performed using AI, or may be performed without using AI. For example, the blocking unit inputs the influence data of the content into an AI model, and the AI ​​model identifies the content with high influence.

[0108] The blocking unit can optimize the blocking algorithm by referring to the past blocking history at the time of blocking. For example, the blocking unit learns the patterns of spam content blocked in the past and optimizes the blocking algorithm. AI can be used for optimization. For example, the blocking unit inputs the past blocking history using an AI model and optimizes the blocking algorithm. The blocking unit can also analyze the tendency of spam content posted by a specific user based on the past blocking history. Furthermore, the blocking unit can also build a feedback loop for improving the accuracy of the blocking algorithm by referring to the past blocking history. As a result, the accuracy of the blocking algorithm can be improved by referring to the past blocking history. A part or all of the above-mentioned processing in the blocking unit may be performed using AI, or may be performed without using AI. For example, the blocking unit inputs the past blocking history data into an AI model, and the AI ​​model optimizes the blocking algorithm.

[0109] The classification unit can estimate the user's emotion and adjust the classification criteria based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the classification unit classifies content with stricter criteria when the user is stressed. Also, when the user is relaxed, the classification unit can classify content with normal criteria. Furthermore, when the user is excited, the classification unit can preferentially classify content that is likely to cause an emotional response. This allows for more appropriate classification by adjusting the classification criteria based on the user's emotion. A part or all of the above-mentioned processing in the classification unit may be performed using AI, or may be performed without using AI. For example, the classification unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the classification criteria.

[0110] The classification unit can analyze the context of the content at the time of classification and perform classification with higher accuracy. The classification unit, for example, analyzes the context of the content and classifies the content based on the context. AI can be used for analyzing the context. For example, the classification unit inputs the context of the content using an AI model and classifies the content based on the context. The classification unit can also analyze the frequency of appearance of keywords in the content and identify the content based on the context. Furthermore, the classification unit can analyze the tone and nuance of the content and classify the content based on the context. As a result, more accurate classification is possible by analyzing the context of the content. A part or all of the above-mentioned processing in the classification unit may be performed using AI, or may be performed without using AI. For example, the classification unit inputs the context data of the content to an AI model, and the AI ​​model classifies the content.

[0111] The classification unit can optimize the classification algorithm by referring to the past classification history at the time of classification. For example, the classification unit learns the patterns of the contents classified in the past and optimizes the classification algorithm. AI can be used for optimization. For example, the classification unit inputs the past classification history using an AI model and optimizes the classification algorithm. The classification unit can also analyze the tendency of the contents posted by a specific user based on the past classification history. Furthermore, the classification unit can also build a feedback loop for improving the accuracy of the classification algorithm by referring to the past classification history. As a result, the accuracy of the classification algorithm can be improved by referring to the past classification history. A part or all of the above-mentioned processing in the classification unit may be performed using AI, or may be performed without using AI. For example, the classification unit inputs the past classification history data into an AI model, and the AI ​​model optimizes the classification algorithm.

[0112] The tagging unit can estimate the user's emotion and adjust the tagging criteria based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the tagging unit performs tagging with stricter criteria when the user is stressed. Also, when the user is relaxed, tagging can be performed with normal criteria. Furthermore, when the user is excited, content that is likely to cause an emotional response can be preferentially tagged. This allows more appropriate tagging by adjusting the tagging criteria based on the user's emotion. A part or all of the above-mentioned processing in the tagging unit may be performed using AI or may be performed without using AI. For example, the tagging unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the tagging criteria.

[0113] The tagging unit can analyze the context of the content at the time of tagging and perform tagging with higher accuracy. The tagging unit, for example, analyzes the context of the content and performs tagging based on the context. AI can be used for analyzing the context. For example, the tagging unit inputs the context of the content using an AI model and performs tagging based on the context. The tagging unit can also analyze the frequency of appearance of keywords in the content and perform tagging based on the context. Furthermore, the tagging unit can analyze the tone and nuance of the content and perform tagging based on the context. This enables tagging with higher accuracy by analyzing the context of the content. A part or all of the above-mentioned processing in the tagging unit may be performed using AI, or may be performed without using AI. For example, the tagging unit inputs the context data of the content to an AI model, and the AI ​​model performs tagging.

[0114] The tagging unit can optimize the tagging algorithm by referring to the past tagging history at the time of tagging. For example, the tagging unit learns the patterns of the content tagged in the past and optimizes the tagging algorithm. AI can be used for optimization. For example, the tagging unit inputs the past tagging history using an AI model and optimizes the tagging algorithm. The tagging unit can also analyze the tendency of the content posted by a specific user based on the past tagging history. Furthermore, the tagging unit can also build a feedback loop for improving the accuracy of the tagging algorithm by referring to the past tagging history. As a result, the accuracy of the tagging algorithm can be improved by referring to the past tagging history. A part or all of the above-mentioned processing in the tagging unit may be performed using AI or may be performed without using AI. For example, the tagging unit inputs the past tagging history data into an AI model, and the AI ​​model optimizes the tagging algorithm.

[0115] The guidance unit can estimate the user's emotion and adjust the content of the guidance based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the guidance unit provides concise and easy-to-understand guidance when the user is stressed. In addition, the guidance unit can provide detailed guidance when the user is relaxed. Furthermore, in the case of an excited user, guidance for calming the user can be provided. This allows more appropriate guidance to be provided by adjusting the content of the guidance based on the user's emotion. A part or all of the above-mentioned processing in the guidance unit may be performed using AI, or may be performed without using AI. For example, the guidance unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the content of the guidance.

[0116] The guidance unit can provide optimal guidance by referring to the user's past behavior history when providing guidance. AI can be used to refer to the behavior history. For example, the guidance unit inputs the past behavior history using an AI model to generate optimal guidance. The guidance unit can also predict the required guidance content from the user's past behavior history. Furthermore, the guidance unit can analyze the user's past behavior history and provide the most effective guidance method. As a result, optimal guidance can be provided by referring to the user's past behavior history. A part or all of the above-mentioned processing in the guidance unit may be performed using AI, or may be performed without using AI. For example, the guidance unit inputs past behavior history data into an AI model, and the AI ​​model generates optimal guidance.

[0117] The guidance unit can provide customized guidance based on the user's field of interest at the time of guidance. For example, the guidance unit preferentially guides information related to the user's field of interest. AI can be used to identify the field of interest. For example, the guidance unit inputs the user's field of interest using an AI model and generates customized guidance. The guidance unit can also provide customized guidance content based on the user's field of interest. Furthermore, the guidance unit can analyze the user's field of interest and provide an optimal guidance method. As a result, more effective guidance can be provided by providing customized guidance based on the user's field of interest. A part or all of the above-mentioned processing in the guidance unit may be performed using AI, or may be performed without using AI. For example, the guidance unit inputs the user's field of interest data into an AI model, and the AI ​​model generates customized guidance.

[0118] The prediction unit can estimate the user's emotions and adjust the prediction criteria for inappropriate behavior based on the estimated user's emotions. An emotion engine or a generation AI can be used to estimate emotions. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the prediction unit predicts inappropriate behavior with stricter criteria when the user is stressed. Also, when the user is relaxed, it can predict inappropriate behavior with normal criteria. Furthermore, when the user is excited, it can predict preferentially behavior that is likely to cause an emotional reaction. This allows for more appropriate prediction by adjusting the prediction criteria for inappropriate behavior based on the user's emotions. Part or all of the above-mentioned processing in the prediction unit may be performed using AI or may be performed without using AI. For example, the prediction unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the prediction criteria for inappropriate behavior.

[0119] The prediction unit can optimize the prediction algorithm by referring to the past behavior history at the time of prediction. The prediction unit, for example, learns the pattern of inappropriate behavior predicted in the past and optimizes the prediction algorithm. AI can be used for optimization. For example, the prediction unit inputs the past behavior history using an AI model and optimizes the prediction algorithm. The prediction unit can also analyze the tendency of inappropriate behavior performed by a specific user based on the past behavior history. Furthermore, the prediction unit can also build a feedback loop for improving the accuracy of the prediction algorithm by referring to the past behavior history. As a result, the accuracy of the prediction algorithm can be improved by referring to the past behavior history. A part or all of the above-mentioned processing in the prediction unit may be performed using AI, or may be performed without using AI. For example, the prediction unit inputs the past behavior history data into an AI model, and the AI ​​model optimizes the prediction algorithm.

[0120] The prediction unit can analyze the context of the content at the time of prediction and predict inappropriate behavior with higher accuracy. The prediction unit, for example, analyzes the context of the content and predicts inappropriate behavior based on the context. AI can be used for analyzing the context. For example, the prediction unit inputs the context of the content using an AI model and predicts inappropriate behavior based on the context. The prediction unit can also analyze the frequency of occurrence of keywords in the content and identify inappropriate behavior based on the context. Furthermore, the prediction unit can analyze the tone and nuance of the content and predict inappropriate behavior based on the context. As a result, by analyzing the context of the content, it is possible to predict inappropriate behavior with higher accuracy. A part or all of the above-mentioned processing in the prediction unit may be performed using AI, or may be performed without using AI. For example, the prediction unit inputs context data of the content into an AI model, and the AI ​​model predicts inappropriate behavior.

[0121] The warning unit can estimate the user's emotion and adjust the content of the warning based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the warning unit issues a warning in a gentle tone when the user is stressed. Also, when the user is relaxed, the warning unit can issue a warning in a normal tone. Furthermore, when the user is excited, the warning unit can issue a warning in a calm tone. This allows for a more appropriate warning by adjusting the content of the warning based on the user's emotion. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the content of the warning.

[0122] The warning unit can optimize the warning algorithm by referring to the past warning history at the time of warning. For example, the warning unit learns the pattern of behavior that has been warned in the past and optimizes the warning algorithm. AI can be used for optimization. For example, the warning unit inputs the past warning history using an AI model and optimizes the warning algorithm. The warning unit can also analyze the tendency of inappropriate behavior performed by a specific user based on the past warning history. Furthermore, the warning unit can also build a feedback loop for improving the accuracy of the warning algorithm by referring to the past warning history. As a result, the accuracy of the warning algorithm can be improved by referring to the past warning history. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit inputs past warning history data into an AI model, and the AI ​​model optimizes the warning algorithm.

[0123] The warning unit can analyze the user's behavior history and select the optimal warning timing at the time of warning. For example, the warning unit analyzes the timing of inappropriate behavior performed by the user in the past and selects the optimal warning timing. AI can be used to analyze the behavior history. For example, the warning unit inputs past behavior history using an AI model and selects the optimal warning timing. The warning unit can also adjust the warning timing based on the user's behavior history. Furthermore, the warning unit can analyze the user's behavior pattern and select the most effective warning timing. As a result, a warning can be issued at the optimal timing by analyzing the user's behavior history. A part or all of the above-mentioned processing in the warning unit may be performed using AI, or may be performed without using AI. For example, the warning unit inputs past behavior history data into an AI model, and the AI ​​model selects the optimal warning timing.

[0124] The emotion analysis unit can estimate the emotion of the user and adjust the criteria of the emotion analysis based on the estimated emotion of the user. An emotion engine or a generative AI can be used to estimate the emotion. The generative AI can be a text generation AI (e.g., LLM) or a multimodal generative AI. For example, the emotion analysis unit performs emotion analysis with stricter criteria when the user is stressed. Also, when the user is relaxed, the emotion analysis unit can perform emotion analysis with normal criteria. Furthermore, when the user is excited, content that is likely to cause an emotional response can be preferentially analyzed. This enables more appropriate emotion analysis by adjusting the criteria of the emotion analysis based on the emotion of the user. A part or all of the above-mentioned processing in the emotion analysis unit may be performed using AI or may be performed without using AI. For example, the emotion analysis unit inputs the emotion data of the user into an AI model, and the AI ​​model adjusts the criteria of the emotion analysis.

[0125] The sentiment analysis unit can analyze the context of the content during sentiment analysis and perform sentiment analysis with higher accuracy. The sentiment analysis unit, for example, analyzes the context of the content and performs sentiment analysis based on the context. AI can be used for analyzing the context. For example, the sentiment analysis unit inputs the context of the content using an AI model and performs sentiment analysis based on the context. The sentiment analysis unit can also analyze the frequency of appearance of keywords in the content and identify emotions based on the context. Furthermore, the sentiment analysis unit can analyze the tone and nuance of the content and perform sentiment analysis based on the context. This enables more accurate sentiment analysis by analyzing the context of the content. A part or all of the above-mentioned processing in the sentiment analysis unit may be performed using AI, or may be performed without using AI. For example, the sentiment analysis unit inputs context data of the content into an AI model, and the AI ​​model identifies emotions.

[0126] The sentiment analysis unit can optimize the sentiment analysis algorithm by referring to the past sentiment analysis history during sentiment analysis. The sentiment analysis unit, for example, learns the patterns of emotions analyzed in the past and optimizes the sentiment analysis algorithm. AI can be used for optimization. For example, the sentiment analysis unit inputs the past sentiment analysis history using an AI model and optimizes the sentiment analysis algorithm. The sentiment analysis unit can also analyze the tendency of emotions expressed by a specific user based on the past sentiment analysis history. Furthermore, the sentiment analysis unit can also refer to the past sentiment analysis history and build a feedback loop for improving the accuracy of the sentiment analysis algorithm. As a result, the accuracy of the sentiment analysis algorithm can be improved by referring to the past sentiment analysis history. A part or all of the above-mentioned processing in the sentiment analysis unit may be performed using AI, or may be performed without using AI. For example, the sentiment analysis unit inputs the past sentiment analysis history data into an AI model, and the AI ​​model optimizes the sentiment analysis algorithm.

[0127] The counseling unit can estimate the user's emotions and adjust the content of counseling based on the estimated user's emotions. An emotion engine or a generation AI can be used to estimate the emotions. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the counseling unit provides counseling for stress reduction when the user is stressed. Also, when the user is relaxed, it can provide normal counseling. Furthermore, when the user is excited, it can provide counseling for calming the emotions. This enables more appropriate counseling by adjusting the content of counseling based on the user's emotions. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the content of counseling.

[0128] The counseling unit can optimize the counseling algorithm by referring to the past counseling history during counseling. The counseling unit, for example, learns the patterns of counseling provided in the past and optimizes the counseling algorithm. AI can be used for optimization. For example, the counseling unit inputs the past counseling history using an AI model and optimizes the counseling algorithm. The counseling unit can also analyze counseling trends for a specific user based on the past counseling history. Furthermore, the counseling unit can also build a feedback loop for improving the accuracy of the counseling algorithm by referring to the past counseling history. As a result, the accuracy of the counseling algorithm can be improved by referring to the past counseling history. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs past counseling history data into an AI model, and the AI ​​model optimizes the counseling algorithm.

[0129] The counseling unit can analyze the user's behavior history during counseling and select the optimal counseling timing. The counseling unit, for example, analyzes the timing of actions taken by the user in the past and selects the optimal counseling timing. AI can be used to analyze the behavior history. For example, the counseling unit inputs the past behavior history using an AI model and selects the optimal counseling timing. The counseling unit can also adjust the timing of counseling based on the user's behavior history. Furthermore, the counseling unit can analyze the user's behavior pattern and select the most effective counseling timing. As a result, counseling can be provided at the optimal timing by analyzing the user's behavior history. A part or all of the above-mentioned processing in the counseling unit may be performed using AI, or may be performed without using AI. For example, the counseling unit inputs past behavior history data into an AI model, and the AI ​​model selects the optimal counseling timing.

[0130] The evaluation aggregation unit can estimate the user's emotions and adjust the criteria for evaluation aggregation based on the estimated user's emotions. An emotion engine or a generation AI can be used to estimate emotions. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, the evaluation aggregation unit performs evaluation aggregation with stricter criteria when the user is stressed. Also, when the user is relaxed, the evaluation aggregation unit can perform evaluation aggregation with normal criteria. Furthermore, when the user is excited, the evaluation aggregation unit can prioritize evaluation aggregation of content that is likely to cause an emotional response. This enables more appropriate evaluation aggregation by adjusting the criteria for evaluation aggregation based on the user's emotions. Part or all of the above-mentioned processing in the evaluation aggregation unit may be performed using AI, or may be performed without using AI. For example, the evaluation aggregation unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the criteria for evaluation aggregation.

[0131] The evaluation aggregation unit can analyze the context of the content when aggregating the evaluations, and perform more accurate evaluation aggregation. The evaluation aggregation unit, for example, analyzes the context of the content and performs evaluation aggregation based on the context. AI can be used for analyzing the context. For example, the evaluation aggregation unit inputs the context of the content using an AI model, and performs evaluation aggregation based on the context. The evaluation aggregation unit can also analyze the frequency of appearance of keywords in the content, and identify the evaluation based on the context. Furthermore, the evaluation aggregation unit can analyze the tone and nuance of the content, and perform evaluation aggregation based on the context. As a result, more accurate evaluation aggregation is possible by analyzing the context of the content. A part or all of the above-mentioned processing in the evaluation aggregation unit may be performed using AI, or may be performed without using AI. For example, the evaluation aggregation unit inputs the context data of the content into an AI model, and the AI ​​model identifies the evaluation.

[0132] The evaluation aggregation unit can optimize the evaluation aggregation algorithm by referring to the past evaluation aggregation history when collating the evaluations. The evaluation aggregation unit, for example, learns the patterns of evaluations collected in the past and optimizes the evaluation aggregation algorithm. AI can be used for the optimization. For example, the evaluation aggregation unit inputs the past evaluation aggregation history using an AI model and optimizes the evaluation aggregation algorithm. The evaluation aggregation unit can also analyze the tendency of evaluations made by a specific user based on the past evaluation aggregation history. Furthermore, the evaluation aggregation unit can also refer to the past evaluation aggregation history and build a feedback loop for improving the accuracy of the evaluation aggregation algorithm. As a result, the accuracy of the evaluation aggregation algorithm can be improved by referring to the past evaluation aggregation history. A part or all of the above-mentioned processing in the evaluation aggregation unit may be performed using AI, or may be performed without using AI. For example, the evaluation aggregation unit inputs the past evaluation aggregation history data into an AI model, and the AI ​​model optimizes the evaluation aggregation algorithm.

[0133] The display order adjustment unit can estimate the user's emotion and adjust the display order adjustment criterion based on the estimated user's emotion. An emotion engine or a generation AI can be used to estimate the emotion. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI. For example, when the user is stressed, the display order adjustment unit preferentially displays content for reducing stress. In addition, when the user is relaxed, the display order adjustment unit can also adjust the display order based on the normal standard. Furthermore, when the user is excited, the display order adjustment unit can also preferentially display content that is likely to cause an emotional response. This makes it possible to adjust the display order more appropriately by adjusting the display order adjustment criterion based on the user's emotion. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the user's emotion data into an AI model, and the AI ​​model adjusts the display order adjustment criterion.

[0134] The display order adjustment unit can analyze the context of the content when adjusting the display order, and adjust the display order more accurately. The display order adjustment unit, for example, analyzes the context of the content, and adjusts the display order based on the context. AI can be used for analyzing the context. For example, the display order adjustment unit inputs the context of the content using an AI model, and adjusts the display order based on the context. The display order adjustment unit can also analyze the frequency of appearance of keywords in the content, and adjust the display order based on the context. Furthermore, the display order adjustment unit can analyze the tone and nuance of the content, and adjust the display order based on the context. As a result, by analyzing the context of the content, it is possible to adjust the display order more accurately. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the context data of the content into an AI model, and the AI ​​model adjusts the display order.

[0135] The display order adjustment unit can optimize the display order adjustment algorithm by referring to the past display order adjustment history when adjusting the display order. The display order adjustment unit, for example, learns the pattern of the display order adjusted in the past and optimizes the display order adjustment algorithm. AI can be used for optimization. For example, the display order adjustment unit inputs the past display order adjustment history using an AI model and optimizes the display order adjustment algorithm. The display order adjustment unit can also analyze the tendency of the content preferred by a specific user based on the past display order adjustment history. Furthermore, the display order adjustment unit can also build a feedback loop for improving the accuracy of the display order adjustment algorithm by referring to the past display order adjustment history. As a result, the accuracy of the display order adjustment algorithm can be improved by referring to the past display order adjustment history. A part or all of the above-mentioned processing in the display order adjustment unit may be performed using AI, or may be performed without using AI. For example, the display order adjustment unit inputs the past display order adjustment history data into an AI model, and the AI ​​model optimizes the display order adjustment algorithm.

[0136] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0137] The online community moderator system may further include a personalized feedback unit based on the user's behavioral history. The personalized feedback unit analyzes the user's past behavioral history and provides individual feedback. For example, if the user has made an inappropriate post in the past, specific improvements to that behavior may be presented. Also, if the user has made a positive contribution, a message praising the contribution may be sent. Furthermore, the user's behavioral patterns may be analyzed and advice for improving future behavior may be provided. This allows the user to reflect on their own behavior and aim for better community participation.

[0138] The detection unit may provide real-time feedback to the user's posts. For example, if a user is creating a post and finds that it contains inappropriate content, the detection unit may immediately display a warning and prompt the user to correct the content. If the post is positive, the detection unit may display an encouraging message. If the post is neutral, the detection unit may provide additional information related to a particular topic. This allows the user to improve the post in real time and provide more appropriate content.

[0139] The identification unit can have a content evaluation function for the contents posted by users. For example, it can not only evaluate whether the posted contents are spam, but also evaluate the quality of the contents. AI can be used for the evaluation. For example, the identification unit inputs the posted contents using an AI model and evaluates the quality of the contents. It can also provide feedback to the user based on the evaluation results. Furthermore, it can adjust the display order of the posted contents based on the evaluation results. This allows high-quality content to be displayed preferentially.

[0140] The prediction unit can not only predict future behavior based on the user's behavioral history, but also analyze the motivation behind the behavior. For example, if a user frequently posts inappropriate content on a specific topic, the motivation behind the posting can be identified. AI can be used to analyze the motivation. For example, the prediction unit inputs the behavioral history using an AI model to identify the motivation. In addition, the prediction unit can provide appropriate feedback and advice to the user based on the motivation. This can fundamentally improve the user's behavior.

[0141] The warning unit can customize the contents of the warning based on the user's behavior history. For example, a stricter warning can be issued to a user who has engaged in similar inappropriate behavior in the past. Also, a gentler warning can be issued to a user who has engaged in inappropriate behavior for the first time. Furthermore, the warning unit can analyze the user's behavioral patterns and select the most effective warning method. This can effectively improve the user's behavior.

[0142] The sentiment analysis unit can analyze not only the content posted by users, but also the comments and reactions. For example, it can identify how users feel about other members' posts. It can also analyze the emotions in the comments and reactions and respond based on the specific emotions. Furthermore, it can provide appropriate feedback and support to users based on the results of the sentiment analysis. This makes it possible to understand the flow of emotions within the community and respond appropriately.

[0143] The counseling unit can create an individual counseling plan based on the result of the user's emotion analysis. For example, if the user feels stressed, a specific action plan for stress reduction can be provided. Also, if the user feels anxious, resources and support for reducing anxiety can be provided. Furthermore, the frequency and content of counseling can be adjusted according to the user's emotional state. This can support the user's mental health.

[0144] The evaluation aggregation unit can evaluate the reliability of the evaluations based on the user's evaluation history. For example, it can give more weight to evaluations from users who have consistently given high ratings in the past. It can also disregard evaluations from users who have given biased ratings in the past. It can also analyze the evaluation history to identify the tendency of evaluations made by a particular user. This improves the reliability of the evaluations and enables more accurate evaluation aggregation.

[0145] The display order adjustment unit can customize the display order based on the user's browsing history. For example, it can preferentially display content related to topics that the user has frequently browsed in the past. It can also preferentially display content related to specific keywords in which the user is interested. It can also analyze the user's browsing history and display content that is most likely to attract the user's attention. This allows the display order to be adjusted based on the user's interests, making it possible to provide content more effectively.

[0146] The detection unit can estimate the user's emotion and adjust the detection criteria for inappropriate content based on the estimated emotion, as well as provide feedback according to the user's emotion. For example, if the user is feeling stressed, advice for reducing stress can be provided. Also, if the user is relaxed, content for maintaining relaxation can be suggested. Furthermore, if the user is excited, resources for calming the user can be provided. In this way, appropriate feedback can be provided based on the user's emotion, and user satisfaction can be improved.

[0147] The process flow of the second embodiment will be briefly described below.

[0148] Step 1: The detection unit detects inappropriate content from among the content posted to the online community. For example, it detects content that contains discriminatory or violent language. The detection unit automatically detects this content using AI. Step 2: The deletion unit deletes the inappropriate content detected by the detection unit. The deletion unit can also perform the deletion automatically using AI. Step 3: The identification unit identifies whether the content posted to the online community is spam content or content containing unwanted links. The identification unit identifies such content using a machine learning model. Step 4: The blocking unit blocks the posting of content that is identified by the identification unit as spam content or content containing unnecessary links. The blocking unit can also perform blocking automatically using AI. Step 5: The classifier categorizes the content posted to the online community into relevant topics. The classifier can also use AI to categorize these contents. Step 6: The tagger tagges the content classified by the classifier. The tagger can also tag automatically using AI. Step 7: The information section provides the new members of the online community with the community guide and rules. The information section can also provide automatic guidance using AI.

[0149] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice 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 voice data.

[0150] 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<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) are input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may also be an AI agent. In addition, when the processes of each of the above-mentioned parts are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples.In addition, processing performed by AI including the generation AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI including the generation AI.

[0151] Furthermore, the above-described processing by the data processing system 10 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 required for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or an external device.

[0152] For example, the detection unit detects content posted to the online community using the camera 42 or microphone 38B of the smart device 14, and identifies inappropriate content by the control unit 46A. The deletion unit deletes inappropriate content detected by the specific processing unit 290 of the data processing device 12. The identification unit identifies spam content and unnecessary links by the specific processing unit 290 of the data processing device 12. The blocking unit blocks the posting of spam content identified by the control unit 46A of the smart device 14. The classification unit classifies posts into related topics by the specific processing unit 290 of the data processing device 12. The tagging unit tags the content classified by the control unit 46A of the smart device 14. The guidance unit guides new members to the community guide and rules by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0153] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0154] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0155] 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, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0156] 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.

[0157] 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 the voice according to instructions from the processor 46.

[0158] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0159] 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 performed in a secure state.

[0160] Fig. 4 shows an example of 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.

[0161] 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.

[0162] 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0163] In the smart glasses 214, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. 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 glasses 214 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 also perform processing similar to that of the specific processing unit 290 using these models.

[0164] 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 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0165] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the 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.

[0166] 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0167] 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 be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, 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, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device.

[0168] For example, the detection unit detects content posted to the online community using the camera 42 and microphone 238 of the smart glasses 214, and identifies inappropriate content by the control unit 46A. The deletion unit deletes inappropriate content detected by the specific processing unit 290 of the data processing device 12. The identification unit identifies spam content and unnecessary links by the specific processing unit 290 of the data processing device 12. The blocking unit blocks the posting of spam content identified by the control unit 46A of the smart glasses 214. The classification unit classifies posts into related topics by the specific processing unit 290 of the data processing device 12. The tagging unit tags the content classified by the control unit 46A of the smart glasses 214. The guidance unit guides new members to the community guide and rules by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0169] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0170] 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.

[0171] 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, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0172] 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.

[0173] 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 the voice according to instructions from the processor 46.

[0174] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0175] 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 performed in a secure state.

[0176] 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.

[0177] 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.

[0178] 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0179] In the headset type terminal 314, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50, and executes the read specific 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 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.

[0180] 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 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0181] 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 voice indicating a user input for 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.

[0182] 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0183] 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 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. In addition, 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, and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device.

[0184] For example, the detection unit detects content posted to the online community using the camera 42 or microphone 238 of the headset type terminal 314, and identifies inappropriate content by the control unit 46A. The deletion unit deletes inappropriate content detected by the specific processing unit 290 of the data processing device 12. The identification unit identifies spam content and unnecessary links by the specific processing unit 290 of the data processing device 12. The blocking unit blocks the posting of spam content identified by the control unit 46A of the headset type terminal 314. The classification unit classifies posts into related topics by the specific processing unit 290 of the data processing device 12. The tagging unit tags the content classified by the control unit 46A of the headset type terminal 314. The guidance unit guides new members to the community guide and rules by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0185] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0186] 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.

[0187] 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, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0188] 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.

[0189] 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 the voice according to instructions from the processor 46.

[0190] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of a typical healthy person).

[0191] 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 performed in a secure state.

[0192] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and legs. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, and legs. 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.

[0193] 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.

[0194] 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.

[0195] 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 the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0196] In the robot 414, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads out the specific program 60 from the storage 50, and executes the specific 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 program 60 executed on the RAM 48. The robot 414 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.

[0197] 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 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0198] 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 a voice indicating a user input for 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.

[0199] 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while 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, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0200] 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 be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, 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.

[0201] For example, the detection unit detects content posted to the online community using the camera 42 and microphone 238 of the robot 414, and identifies inappropriate content by the control unit 46A. The deletion unit deletes inappropriate content detected by the specific processing unit 290 of the data processing device 12. The identification unit identifies spam content and unnecessary links by the specific processing unit 290 of the data processing device 12. The blocking unit blocks the posting of spam content identified by the control unit 46A of the robot 414. The classification unit classifies posts into related topics by the specific processing unit 290 of the data processing device 12. The tagging unit tags the content classified by the control unit 46A of the robot 414. The guidance unit guides new members to the community guide and rules by the specific processing unit 290 of the data processing device 12. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0202] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0203] FIG. 9 is a diagram showing 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. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0204] These emotions are distributed in the 3 o'clock direction of emotion map 400 and usually 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.

[0205] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out on emotion map 400 you go, the more visible the emotions become (the more they are expressed in behavior).

[0206] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0207] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0208] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the 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, "relief," "calm," and "encouraging," have similar emotion values.

[0209] 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.

[0210] 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 Universal Serial Bus (USB) 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.

[0211] In addition, 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 upon request from the data processing device 12.

[0212] 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.

[0213] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing a specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[0214] The hardware resource that executes the specific process may be one of these various processors, or may be 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.

[0215] As an example of a configuration using one 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 configuration in which a processor is used that realizes the functions of the 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.

[0216] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[0217] In the above example, the first to fourth embodiments are described separately, but some or all of these embodiments may be combined. The smart device 14, the smart glasses 214, the headset terminal 314, and the robot 414 are only examples, and they may be combined with each other, or may be other devices. In the above example, the first and second embodiments are described separately, but they may be combined.

[0218] The above description and illustrations are detailed descriptions 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, function, action, and effect is an example of the configuration, function, action, and effect 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 replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require particular explanation in enabling the implementation of the technology of the present disclosure.

[0219] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0220] (Appendix 1) a detection unit that detects content that is determined to be inappropriate from among content posted to an online community; a deletion unit that deletes the content that is determined to be inappropriate and detected by the detection unit; an identification unit that determines whether content posted to the online community is spam content or content including unnecessary links; a blocking unit that blocks posting of content determined by the identification unit to be spam content or content including unnecessary links; a categorization unit for categorizing content posted to the online community into related topics; a tagging unit that tags the content classified by the classification unit; A guidance unit that provides a community guide and rules to new members of the online community. A system characterized by: (Appendix 2) The detection unit is Detecting content including discriminatory language and content including violent language as content determined to be inappropriate 2. The system of claim 1, (Appendix 3) The identification unit is Using a machine learning model to identify whether content posted to the online community is the spam content and the content containing the unwanted links. 2. The system of claim 1, (Appendix 4) In the online community, A prediction unit that analyzes the behavioral history of members of the online community and predicts behavior that is deemed inappropriate. 2. The system of claim 1, (Appendix 5) a warning unit that warns of the behavior that is predicted by the prediction unit and is determined to be inappropriate; 4. The system of claim 3. (Appendix 6) In the online community, A sentiment analysis unit is provided for analyzing sentiment in the content posted by members of the online community. 2. The system of claim 1, (Appendix 7) Based on the specific emotion detected by the emotion analysis unit, a counseling department that provides counseling to said members; 6. The system of claim 5, (Appendix 8) In the online community, Equipped with a rating aggregation unit that aggregates ratings of posted content 2. The system of claim 1, (Appendix 9) Based on the evaluations tallied by the evaluation tallying unit, Equipped with a display order adjustment unit that adjusts the display order of posted content 8. The system of claim 7, (Appendix 10) The detection unit is Inferring user sentiment and adjusting inappropriate content detection criteria based on the estimated user sentiment 2. The system of claim 1, (Appendix 11) The detection unit is Upon detection, the context of the content is analyzed to provide more accurate detection of inappropriate content. 2. The system of claim 1, (Appendix 12) The detection unit is When detecting, the detection algorithm is optimized by referring to past detection history. 2. The system of claim 1, (Appendix 13) The deletion unit is Estimate user sentiment and adjust the timing of deletion based on the estimated user sentiment 2. The system of claim 1, (Appendix 14) The deletion unit is When deleting content, evaluate its impact and prioritize deleting content with the greatest impact. 2. The system of claim 1, (Appendix 15) The deletion unit is When deleting, the deletion algorithm is optimized by referring to the past deletion history. 2. The system of claim 1, (Appendix 16) The identification unit is Inferring user sentiment and adjusting spam content identification criteria based on the estimated user sentiment 2. The system of claim 1, (Appendix 17) The identification unit is During identification, the context of the content is analyzed to more accurately identify spam content. 2. The system of claim 1, (Appendix 18) The identification unit is During identification, the identification algorithm is optimized by referring to past identification history. 2. The system of claim 1, (Appendix 19) The block portion is Estimate the user's emotions and adjust the timing of blocks based on the estimated user emotions. 2. The system of claim 1, (Appendix 20) The block portion is When blocking, the impact of the content is evaluated and the content with the highest impact is blocked first. 2. The system of claim 1, (Appendix 21) The block portion is When blocking, the blocking algorithm is optimized by referring to past block history. 2. The system of claim 1, (Appendix 22) The classification unit includes: Inferring user emotion and adjusting classification criteria based on the estimated user emotion 2. The system of claim 1, (Appendix 23) The classification unit includes: During classification, the context of the content is analyzed to provide a more accurate classification. 2. The system of claim 1, (Appendix 24) The classification unit includes: During classification, the classification algorithm is optimized by referring to past classification history. 2. The system of claim 1, (Appendix 25) The tagging unit includes: Infer user sentiment and adjust tagging criteria based on the inferred user sentiment 2. The system of claim 1, (Appendix 26) The tagging unit includes: When tagging, the context of the content is analyzed to make tagging more accurate. 2. The system of claim 1, (Appendix 27) The tagging unit includes: When tagging, look at past tagging history to optimize the tagging algorithm 2. The system of claim 1, (Appendix 28) The guide unit is Estimate the user's emotions and adjust the guidance content based on the estimated user's emotions 2. The system of claim 1, (Appendix 29) The guide unit is When providing guidance, the system provides optimal guidance by referring to the user's past behavior history. 2. The system of claim 1, (Appendix 30) The guide unit is When navigating, provide customized guidance based on the user's areas of interest 2. The system of claim 1, (Appendix 31) The prediction unit is Inferring a user's emotions and adjusting a prediction criterion for inappropriate behavior based on the inferred user's emotions 2. The system of claim 1, (Appendix 32) The prediction unit is When making predictions, the prediction algorithm is optimized by referring to past behavioral history. 2. The system of claim 1, (Appendix 33) The prediction unit is When making predictions, the context of the content is analyzed to make more accurate predictions of inappropriate behavior. 2. The system of claim 1, (Appendix 34) The warning unit is Infer user emotions and adjust warning content based on the estimated user emotions 2. The system of claim 1, (Appendix 35) The warning unit is When a warning occurs, the warning algorithm is optimized by referring to past warning history. 2. The system of claim 1, (Appendix 36) The warning unit is When issuing a warning, the user's behavior history is analyzed to determine the optimal warning timing. 2. The system of claim 1, (Appendix 37) The emotion analysis unit is Inferring a user's sentiment and adjusting sentiment analysis criteria based on the inferred user's sentiment 2. The system of claim 1, (Appendix 38) The emotion analysis unit is When analyzing sentiment, the context of the content is analyzed to perform more accurate sentiment analysis. 2. The system of claim 1, (Appendix 39) The emotion analysis unit is When analyzing sentiment, optimize the sentiment analysis algorithm by referring to past sentiment analysis history. 2. The system of claim 1, (Appendix 40) The counseling department: To estimate a user's emotion and adjust counseling contents based on the estimated user's emotion. 2. The system of claim 1, (Appendix 41) The counseling department: During counseling, the counseling algorithm is optimized by referring to past counseling history. 2. The system of claim 1, (Appendix 42) The counseling department: During counseling, analyze the user's behavioral history and select the optimal timing for counseling. 2. The system of claim 1, (Appendix 43) The evaluation tabulation unit includes: Inferring user sentiment and adjusting criteria for rating aggregation based on the estimated user sentiment 2. The system of claim 1, (Appendix 44) The evaluation tabulation unit includes: When collating ratings, the context of the content is analyzed to allow for more accurate compilation of ratings. 2. The system of claim 1, (Appendix 45) The evaluation tabulation unit includes: When collecting ratings, the rating aggregation algorithm is optimized by referencing past rating aggregation history. 2. The system of claim 1, (Appendix 46) The display order adjustment unit is A user's emotion is estimated, and a display ranking adjustment criterion is adjusted based on the estimated user's emotion. 2. The system of claim 1, (Appendix 47) The display order adjustment unit is When adjusting display ranking, the context of the content is analyzed to adjust the display ranking more accurately. 2. The system of claim 1, (Appendix 48) The display order adjustment unit is When adjusting display rank, optimize the display rank adjustment algorithm by referring to the history of past display rank adjustments. 2. The system of claim 1, [Explanation of symbols]

[0221] 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 detection unit that detects content that is determined to be inappropriate from among content posted to an online community; a deletion unit that deletes the content that is determined to be inappropriate and detected by the detection unit; an identification unit that determines whether content posted to the online community is spam content or content including unnecessary links; a blocking unit that blocks posting of content determined by the identification unit to be the spam content or the content including the unnecessary link; a categorization unit for categorizing content posted to the online community into related topics; a tagging unit that tags the content classified by the classification unit; A guidance unit that provides a community guide and rules to new members of the online community. A system characterized by:

2. The detection unit is Detecting content including discriminatory language and content including violent language as content determined to be inappropriate 2. The system of claim 1 .

3. The identification unit is Using a machine learning model to identify whether content posted to the online community is the spam content and the content containing the unwanted links.

2. The system of claim 1 .

4. A prediction unit that analyzes the behavioral history of the members of the online community and predicts behavior that is determined to be inappropriate.

2. The system of claim 1 .

5. a warning unit that warns of the behavior that is predicted by the prediction unit and is determined to be inappropriate; 5. The system of claim 4.

6. A sentiment analysis unit that analyzes sentiment of the content posted by the members of the online community in the online community.

2. The system of claim 1 .

7. A counseling unit that provides counseling to the member based on the specific emotion detected by the emotion analysis unit.

7. The system of claim 6.

8. A rating aggregation unit that aggregates ratings of content posted in the online community.

2. The system of claim 1 .

Images