system

Abstract

We provide the system. [Solution] A natural language processing method for analyzing communication data received from a user, An evaluation method for assigning an emotional score to the analyzed communication data, A selection means that filters communications based on an emotional rating value derived from user settings, A means of providing filtered communications to the user terminal, A learning method that improves filtering accuracy by updating the machine learning model based on adjustments to user selection settings, A receiving means that can identify and display communications that point out new areas for improvement, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] On social network services, defamatory and offensive comments are increasing, which causes problems for users to suffer from mental stress. In particular, for the brand images of famous people and companies with many followers, the impact of such negative comments is significant. On the other hand, since it is necessary to incorporate constructive criticism without overlooking it, simply hiding negative comments is not practical. Also, since the criteria for what each user feels "uncomfortable" with are different, individually customized filtering is required. As a result, there is an increasing need for a system that can reduce stress according to user-specific desires and obtain valuable feedback.

Means for Solving the Problems

[0005] This invention provides a natural language processing means for analyzing comment data received from users. This means includes a scoring means that understands the context of the comment data and assigns a score for the degree of negativity. Furthermore, it solves this problem by including a filtering means that automatically filters out comments that exceed a threshold of negativity acceptable to the user, based on user settings. The filtered comment data is then sent to the user's terminal, ultimately controlling the comments the user sees. In addition, the filtering accuracy is improved by collecting data on user-defined filtering settings and updating the machine learning model. As a result, it becomes possible to reduce the impact of negative comments according to the individual user's needs and alleviate stress on social media.

[0006] A "user" refers to an individual or organization that uses the system to configure comment filtering and receives content based on the results.

[0007] "Comment data" refers to a set of information that includes metadata such as text information posted on social networking services, the poster, and the time of posting.

[0008] "Natural language processing means" refers to methods that analyze received comment data and utilize techniques such as syntactic analysis and semantic analysis to understand its context and linguistic expression.

[0009] A "negativity score" is an indicator that numerically represents how offensive, offensive, or critical a comment is.

[0010] "Scoring method" refers to a technical means for assigning a score of negativity to comment data analyzed using natural language processing.

[0011] "Filtering method" refers to a means of selecting comment data based on a negativeness score, according to user-defined criteria, and deciding whether or not to display it visually.

[0012] "Transmission method" refers to the communication method used to send filtered comment data to the user's device.

[0013] A "machine learning model" refers to an algorithm or method used to analyze and predict data, and is a model that has the ability to improve its performance through user feedback and new data. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the language used in the following description will be explained.

[0017] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention is a filtering system for efficiently managing comments received by users on social networking services. Specific embodiments of this system are described below.

[0036] First, the user connects to the system via a dedicated terminal application and sets filtering criteria for incoming comments. These criteria include a threshold for negativity and specific expressions to be filtered.

[0037] Next, the server receives comment data flowing continuously from the SNS platform based on those criteria and passes the data to a dedicated natural language processing engine. This engine analyzes the context of the comments, specifically checking the structure and content of the language and the meaning of the words used.

[0038] The server then assigns a negativity score to the analyzed comment data. Here, it uses historical training data and machine learning models to evaluate how offensive or offensive each comment is considered to be.

[0039] Subsequently, the server's filtering mechanism refers to the criteria set by the user and filters out comments with high scores. Care is taken during this process to ensure that constructive critical comments are not affected.

[0040] The filtered results are sent from the server to the terminal, which then displays only the approved comments to the user. If the user feels that the comments do not match their filtering criteria, they are given the opportunity to adjust their settings.

[0041] Finally, the feedback is sent back to the server, and the machine learning model is updated using that information. This improves the overall filtering accuracy of the system, bringing it closer to the environment the user desires.

[0042] As a concrete example, consider a well-known company account that receives many comments about a particular product. This company wants to filter out negative expressions such as "terrible" or "bad," while receiving constructive feedback such as "there is room for improvement." This system would allow the company to eliminate unnecessary negative comments and utilize helpful feedback.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] Users launch a dedicated application on their device and set their own filtering criteria. These criteria include thresholds for the degree of negativity they want to filter out and specific keywords.

[0046] Step 2:

[0047] The server receives comment data flowing in real time from the social networking platform. The received data includes comment text and metadata.

[0048] Step 3:

[0049] The server sends the received comment data to a natural language processing engine. This engine analyzes the language syntax, vocabulary, and context, and converts the comments into a structured data format.

[0050] Step 4:

[0051] The server uses a machine learning model to assign negativity scores to structured comment data. This model quantifies aggression and unpleasantness based on training data.

[0052] Step 5:

[0053] The server filters comments based on the user's filtering criteria, referring to their negativity score. Comments exceeding the set threshold are filtered out, and unnecessary ones are excluded.

[0054] Step 6:

[0055] The server sends the allowed comment data from the filtered comments to the terminal.

[0056] Step 7:

[0057] The device displays received comments on the user's screen. Here, filtered negative comments are not displayed, and only comments deemed constructive are visualized.

[0058] Step 8:

[0059] Users review the displayed comments and make adjustments to their filtering settings if they feel they need to. This feedback is sent to the server via their device.

[0060] Step 9:

[0061] The server uses user feedback data to update its machine learning model and improve filtering accuracy.

[0062] (Example 1)

[0063] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0064] With the advancement of modern information and communication technology, users routinely receive large amounts of comments and information from various platforms. However, this information may include negative content or information unnecessary for the user, potentially impairing their comfortable user experience. Therefore, there is a need for means to appropriately filter the information users receive and eliminate unnecessary information.

[0065] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0066] In this invention, the server includes an information analysis means for analyzing data received from a user, an evaluation means for assigning an evaluation score to the analyzed information, and a selection means for selecting information based on the evaluation score according to user settings. This makes it possible to efficiently manage the information received by the user and accurately acquire only the important information.

[0067] "Information analysis means" refers to technologies that include the stage of analyzing data received from users and understanding its content and context.

[0068] "Evaluation methods" refer to techniques for quantifying the quality and attributes of analyzed information and assigning scores as evaluation criteria.

[0069] "Selection method" refers to a technology that uses assigned scores to select useful information and eliminate unnecessary information based on criteria set in advance by the user.

[0070] "Communication means" refers to the technologies and processes used to transmit selected information to a user's display device or other device.

[0071] "Learning methods" refer to technologies that enable a system to self-improve based on user feedback and setting changes, thereby improving the accuracy of information processing in subsequent instances.

[0072] The system for implementing this invention utilizes multiple electronic devices to provide information management as requested by the user. First, the user accesses the system through a dedicated terminal application and sets filtering criteria. These criteria include a threshold for the acceptable degree of negativity and the specification of specific expressions. The terminal application allows for easy modification and application of the criteria via a user interface.

[0073] The server receives the user's settings and continuously receives comment data from social media and other platforms. The received data is parsed using a natural language processing engine. This engine includes techniques to understand the context of the comments and analyze the content and nuances of the language. After analysis, the server uses a machine learning model to assign an appropriate negativeness score to the analysis results. This enables information filtering and allows comments to be filtered according to the user's settings.

[0074] Filtered information is sent from the server to the user's device, which displays only the selected information. During this process, if the user is dissatisfied with the displayed information, they can adjust the settings through feedback. This feedback is sent back to the server, and the machine learning model is updated based on that information. This improves the filtering accuracy for subsequent uses, providing the user with a more favorable information environment.

[0075] For example, when a user receives comments on a sports-related platform, they might want to avoid harsh criticism such as "XXX is the worst." By using a prompt in the format of "Please determine whether this comment is constructive according to user standards," the AI ​​model can determine whether the comment is constructive criticism or not and perform appropriate filtering.

[0076] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0077] Step 1:

[0078] The user sets the filtering criteria on their device.

[0079] Users use a dedicated application to set filtering criteria for incoming information. In this process, users input a threshold for negativity and specific expressions they wish to filter. The entered settings data is sent to the server and stored as the basis for the filtering process.

[0080] Step 2:

[0081] The server receives the comment data.

[0082] The server continuously retrieves comment data from social media and other platforms. This reception involves collecting large amounts of comment data in real time via APIs. The received data serves as input for subsequent processing and is stored in temporary storage on the server.

[0083] Step 3:

[0084] The server analyzes the data using a natural language processing engine.

[0085] The server passes the received comment data to a natural language processing engine. Here, the linguistic structure, context, and word meanings of the data are analyzed. For example, negative expressions and contexts within the text are identified. The analysis results are tagged and output as the basis for calculating evaluation scores.

[0086] Step 4:

[0087] The server evaluates the degree of negativity and assigns a score.

[0088] Based on the analysis results, the server uses a machine learning model to assign a negativity score to each comment. Using an algorithm trained on past data, it quantifies how negative each comment is. This score is then output as the comment's evaluation value.

[0089] Step 5:

[0090] The server filters comments based on the specified criteria.

[0091] Based on the assigned score, the server filters comments against the criteria set by the user. Negative comments exceeding the set threshold are excluded, and only positive or constructive feedback is selected. The filtered results are compiled into output data for the user.

[0092] Step 6:

[0093] The filtering results are sent from the server to the terminal.

[0094] The server sends filtered comments to the user's terminal. During transmission, data is transferred securely and quickly using a communication protocol. The comments are then output to the terminal as they should be displayed and provided to the user.

[0095] Step 7:

[0096] Users review the results and provide feedback.

[0097] Users review the comments displayed on their devices and provide feedback if they are dissatisfied with the filtering accuracy or wish to request adjustments. This feedback information is then sent back to the server and used as input data to improve the filtering criteria and machine learning models.

[0098] Step 8:

[0099] The server updates the machine learning model based on the feedback.

[0100] Based on the feedback, the server updates its machine learning model to improve filtering accuracy. This enhances its ability to handle new data analysis and filtering criteria, enabling more efficient information selection.

[0101] (Application Example 1)

[0102] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0103] With the increasing number of comments on online platforms, it has become difficult for administrators to efficiently filter out inappropriate or pointless negative comments and appropriately identify helpful improvement suggestions. In this situation, there is a need to accurately evaluate the nature of comments and implement filtering that meets the user's expectations.

[0104] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0105] In this invention, the server includes a natural language processing means for analyzing communication data received from a user, an evaluation means for assigning an emotional score to the analyzed communication data, and a selection means for filtering communications based on the emotional score evaluation value according to user settings. This improves inefficient comment processing and enables communication that emphasizes useful feedback.

[0106] "Natural language processing means" refers to technologies that analyze communication data received from users, understand its context and vocabulary, and structure it.

[0107] The "evaluation method" involves quantifying the degree of emotion and expression in the analyzed communication data to determine the degree of negativity or positivity of the comments.

[0108] A "selection method" is a technology that filters communication data according to the emotional intensity evaluation value based on criteria set by the user, and selects comments that exceed a specific standard.

[0109] "Transmission means" refers to the function of sending filtered communication data to the user's terminal, and is responsible for ensuring that only information useful to the user is provided.

[0110] A "learning tool" is a system that continuously updates machine learning models and improves filtering accuracy based on changes in user selection settings and feedback.

[0111] "Receiving means" refers to technology that identifies communications pointing out new areas for improvement and makes it easy for users to receive and view them.

[0112] The system for implementing this invention consists of a server and a user terminal. The server analyzes, evaluates, and sorts communication data using a natural language processing engine and a machine learning model. Specifically, it uses natural language processing libraries such as "spaCy" to perform language analysis and sentiment evaluation. For scoring sentiment levels, it uses machine learning frameworks such as "scikit-learn" and "TENSORFLOW®". Through these technologies, the server detects and sorts negative or positive comments.

[0113] Furthermore, the server is equipped with a means to update the learning model based on user feedback, thereby improving filtering accuracy. Specifically, comments that can be evaluated are identified based on user settings, and the selected comments are transmitted to the terminal. This makes it easy for users to access only the useful feedback from the comments they receive.

[0114] The user's device sends comment data to the server in real time via the SNS API. The server evaluates and filters the received comment data and returns the filtered results to the device. Users can further improve the accuracy of the filtering by adjusting their selection settings.

[0115] As a concrete example, a restaurant chain could use this system when launching a social media campaign for a new dish to filter out critical comments such as "it's not tasty" and receive specific improvement suggestions such as "it would be better if the spices were used sparingly." This would allow them to quickly provide services that meet customer needs. An example of a prompt to input into the generating AI model is: "Design a social media filtering app that efficiently manages comments received on a social media account. It should eliminate negative comments based on user settings and display only helpful feedback."

[0116] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0117] Step 1:

[0118] The user's terminal retrieves comment data through the API of a social networking service. The retrieved comment data is sent to the server in its original form. The input is comment data obtained from the SNS, and the output is data transfer to the server. Specifically, the system retrieves data using an HTTP request, analyzes the response, and extracts the necessary comments.

[0119] Step 2:

[0120] The server receives comment data sent from the user terminal. This data is input into a natural language processing engine for content analysis. The input is the comment data received from the user terminal, and the output is the analyzed comment data. Specifically, the "spaCy" library is used to analyze the grammatical structure and keywords of the comments.

[0121] Step 3:

[0122] The server assigns a sentiment score to each comment based on the analysis results. Machine learning tools such as "TensorFlow" and "scikit-learn" are used for this scoring. The input is the analyzed comment data, and the output is the comment data with the sentiment score assigned. Specifically, the scoring is performed using a model based on past training data.

[0123] Step 4:

[0124] The server filters comment data according to sentiment scores based on criteria pre-configured by the user. The input is scored comment data, and the output is filtered comments. Specifically, it excludes comments with scores above a set threshold and selects only permitted comments.

[0125] Step 5:

[0126] The filtered comment data is sent from the server to the user's terminal. The user's terminal receives this data and displays it on the screen. The input is filtered comment data, and the output is the display on the user's terminal. Specifically, the data is received in HTML or JSON format and rendered on the user interface.

[0127] Step 6:

[0128] Users review the display results of received comments and adjust the settings if they feel the filtering criteria are inappropriate. This feedback is sent to the server and used to update the model. The input is the user's feedback, and the output is the updated filtering criteria. Specifically, the criteria are adjusted from the settings screen and saved to the database.

[0129] Step 7:

[0130] The server updates its machine learning model based on user feedback, improving filtering accuracy. The input is the user's adjusted criteria and feedback, and the output is the updated model. Specifically, it uses the collected feedback data as new training data to retrain the generative AI model.

[0131] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0132] This invention is a filtering system for efficiently managing comments received by users when using social networking services, based on the user's emotional state. The system incorporates four main methods: natural language processing, scoring, filtering, and an emotion engine.

[0133] First, users can access the system through a dedicated terminal application and set filtering criteria that include their own emotional data. This setting utilizes real-time emotional state data obtained from the emotion engine. For example, users can set stricter filtering criteria when they are experiencing high stress levels.

[0134] Next, the server aggregates comment data received from the SNS platform in real time and analyzes it using a natural language processing engine. This analysis confirms the context and vocabulary of the comments and converts them into structured data.

[0135] The server assigns a negativity score to the analyzed comment data. This scoring process is performed by a machine learning model that accurately assesses the degree of negativity based on past data.

[0136] Subsequently, the emotion engine is utilized, and the server analyzes the user's emotional state and dynamically incorporates this data into the filtering criteria. This allows the filtering criteria to be automatically adjusted according to the user's current emotional state.

[0137] Filtered comments are sent from the server to the terminal, which then displays or hides the filtered content on the user's screen. The system prioritizes comments that are sensitive to the user's emotions. For example, when the user is emotionally positive, some critical comments are displayed; conversely, when the user is emotionally negative and easily stressed, potentially stressful comments are filtered more strictly.

[0138] In this way, users can comfortably use social media while taking their own emotional state into consideration. This invention protects users' mental health and ensures that only constructive opinions influence their lives.

[0139] The following describes the processing flow.

[0140] Step 1:

[0141] The user launches a dedicated application on their device and provides the system with their emotional state, either by inputting it or by providing emotional data obtained from sensors. This data is then used for comment filtering.

[0142] Step 2:

[0143] The server receives comment data flowing in from the social networking platform in real time. This data includes metadata such as the comments themselves and the person who posted them.

[0144] Step 3:

[0145] The server sends the received comment data to a natural language processing engine. This engine performs analysis and tokenization and part-of-speech tagging to understand the linguistic structure and content of the comments.

[0146] Step 4:

[0147] The server passes the analyzed comment data to a scoring system, where it assigns a score for negativity. This scoring is performed using a machine learning algorithm and is based on past training data.

[0148] Step 5:

[0149] The server uses an emotion engine to assess the user's emotional state and dynamically adjusts filtering criteria based on that data. If the user is in a negative emotional state, the filtering criteria are set to become stricter.

[0150] Step 6:

[0151] The server selects comments filtered based on scoring, removing those exceeding a threshold. Constructive criticism is retained according to user preferences.

[0152] Step 7:

[0153] The server sends the filtered comment data to the terminal. The terminal then selects only the comments to display on the user screen based on the filtered results.

[0154] Step 8:

[0155] Users can review the comments displayed on their device and provide feedback to change settings if they are dissatisfied with the filtering results. This feedback will be used to improve the filtering accuracy in the future.

[0156] Step 9:

[0157] The server collects user feedback and uses it to update machine learning models and sentiment engines, thereby improving the overall accuracy of the system.

[0158] (Example 2)

[0159] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0160] In today's social media environment, comments received by users can affect their mental health. For users who are particularly sensitive to negative comments, such comments can be a source of stress. The objective of this invention is to enable users to comfortably use social networks while protecting their mental health by filtering comments according to their emotional state.

[0161] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0162] In this invention, the server includes emotion analysis means for analyzing emotion data received from the user to determine the emotional state, criteria adjustment means for adjusting filtering criteria based on the determined emotional state, and natural language processing means for analyzing received comment data. This enables dynamic filtering according to the user's emotional state.

[0163] "Emotional analysis means" refers to a technological element that analyzes emotional data received from a user and determines the user's emotional state.

[0164] A "criteria adjustment means" is a technical element that dynamically adjusts the criteria for comment filtering based on the user's emotional state determined by the sentiment analysis means.

[0165] "Natural language processing means" refers to technical elements that analyze received comment data and extract context and emotional nuances.

[0166] A "scoring method" is a technical element that assigns a score of negativity to analyzed comment data and performs evaluation based on that score.

[0167] A "filtering mechanism" is a technical element that selects and determines whether or not to display comments based on adjusted filtering criteria.

[0168] "Transmission means" refers to the technical element that transfers filtered comments to the device used by the user.

[0169] This invention is a system that filters comments on social networks according to the user's emotional state, enabling the user to receive information comfortably.

[0170] The system first determines the user's emotional state using emotion analysis tools. Users can provide emotional data through a terminal application using biometric sensors and a user interface. This emotional data can be acquired, for example, from a heart rate monitor or facial recognition camera.

[0171] The sentiment analysis method utilizes machine learning algorithms and sentiment analysis libraries to analyze this data. The server uses the algorithm to classify the user's current emotional state into categories such as positive or negative.

[0172] Next, the criteria adjustment mechanism dynamically adjusts the filtering criteria based on the determined emotional state. This adjustment sets the strictness of filtering for negative comments to match the user's emotions.

[0173] The server analyzes comments collected from social media using natural language processing tools to detect the context and emotional nuances of the comments. This structures the comment data, and a negativeness score is assigned via a scoring mechanism. For example, a machine learning platform is used for evaluation.

[0174] The filtering mechanism selects which comments to display based on the negativity score and the user's emotional state. The filtered comments are then transmitted to the device via a transmission mechanism, where the device decides whether to display or hide them. This process ensures that the information the user receives is tailored to their emotional state at that time.

[0175] For example, when a user is relaxed, even slightly critical comments will be displayed without issue, but when they are under a lot of stress, comments that could cause stress will be strictly filtered out.

[0176] The prompts for the generating AI model are entered in the format of, for example, "Explain how to filter received comments when the user is in a stressful state." These prompts enable the system to implement specific filtering that takes the user's state into consideration.

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

[0178] Step 1:

[0179] Users input emotional data via their devices. Sensors such as heart rate monitors and facial recognition cameras are used for input. This allows the system to obtain the user's emotional state as numerical data. The server receives this data and stores it as initial data.

[0180] Step 2:

[0181] The server analyzes the received emotional data using emotion analysis tools. This analysis utilizes emotion analysis libraries and machine learning algorithms. Through this analysis, the emotional data is classified into categories such as "positive" and "negative." Based on these emotional states, the server sets filtering criteria for each user.

[0182] Step 3:

[0183] The server collects relevant comment data through the SNS API. The collected comments are stored in the database in their original form. This prepares the comment information for analysis.

[0184] Step 4:

[0185] The server uses natural language processing to analyze the collected comment data. This analysis extracts contextual and emotional nuances from vocabulary. The analysis results are output as structured data and passed on to the next scoring process.

[0186] Step 5:

[0187] The server assigns a negativity score to the analyzed comments using a scoring mechanism. Here, a predictive model based on historical data is applied using a generative AI model. The processed scoring data, with scores expressed in numerical format, is used as a filtering criterion.

[0188] Step 6:

[0189] The server applies filtering criteria based on the user's emotional state via a criteria adjustment mechanism. This dynamically filters the collected comments. Based on these criteria, comments with high negative scores are hidden, and only comments within an acceptable range are displayed.

[0190] Step 7:

[0191] The server delivers filtered comments to the user's terminal via a transmission method. The terminal receives these comments and displays them on the screen. For some hidden comments, the user is notified and given the option to display them if desired.

[0192] This series of steps allows users to view comments that take their current emotional state into consideration, enabling them to use social media in a way that respects their mental health.

[0193] (Application Example 2)

[0194] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0195] Traditional social networking services lack comment management that takes into account users' emotional states, potentially harming users' mental health. Furthermore, their filtering accuracy is fixed and inadequate, failing to adapt to real-time changes in users' emotions.

[0196] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0197] In this invention, the server includes an electronic computing means for analyzing user comment data, an evaluation means for assigning an evaluation to the analysis results, and an emotion analysis means for analyzing the user's emotional state and changing the filtering criteria. This enables comment management according to the user's emotional state, and allows for more flexible and improved filtering accuracy.

[0198] "Electronic computing means" refers to digital computing devices and their functions used to process and analyze information.

[0199] "Evaluation means" refers to a function or device that performs a process of assigning scores or ratings to data using specific criteria.

[0200] "Emotional analysis means" refers to a function or device that can analyze a user's emotional state in real time and dynamically change the criteria for data processing based on that analysis.

[0201] "Filtering means" refers to a function or device that selects information from data that meets specific criteria and displays or blocks it as needed.

[0202] "Communication means" refers to protocols, devices, or functions used to send and receive data.

[0203] The system implementing this invention provides a multi-stage process for managing comments in order to make the user's social networking service experience more comfortable. The system operates in a network environment that includes user terminals and servers.

[0204] The server receives comment data and uses electronic computing tools to analyze it. These tools employ natural language processing algorithms to analyze the context and vocabulary of the data. For example, technologies such as TensorFlow and NLTK can be used to effectively analyze text data.

[0205] Next, the server uses an evaluation tool to assign a negativity score to the analyzed comments. This scoring uses historical data and a learned model (e.g., a machine learning model) to assess the negativity of the comments.

[0206] The emotion analysis system analyzes the user's emotional state in real time. Using hardware such as voice sensors (e.g., Shure MOTIV) and video sensors (e.g., Intel RealSense camera), it detects the user's facial expressions and tone of voice to acquire their emotional state. This data is then incorporated as filtering criteria by a filtering system. This allows the server to filter comments based on criteria that reflect the user's real-time emotional state.

[0207] The filtered comments are ultimately sent to the device via communication and displayed to the user. Because this display takes the user's emotional state into consideration, it can provide a more stress-free social media experience.

[0208] For example, when a user is relaxed, many comments, including minor criticisms, will be displayed, but when they are tired, only positive comments will be prioritized and displayed.

[0209] An example of a prompt to input into the generating AI model is as follows: "We are developing a system that filters comments based on their emotional state. It will manage comments in real time based on user emotional data. Please provide helpful suggestions."

[0210] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0211] Step 1:

[0212] The server receives comment data from a social networking service. The input data consists of unparsed text comments, which are then sent to a natural language processing engine. The server uses this engine to analyze the context and vocabulary of the comments and generate structured data.

[0213] Step 2:

[0214] The server assigns a negativity score to the analyzed comment data using an evaluation method. The input data is the analyzed text, and the output data is the negativity score for each comment. This process is performed by calculating the negativity score from past data using a learning model.

[0215] Step 3:

[0216] The system acquires the user's emotional state in real time using voice and video sensors. The input is the user's voice and video data, and the output is the analyzed emotional state data. The server analyzes this data using emotion analysis tools and uses the results as filtering criteria for the next step.

[0217] Step 4:

[0218] The server sets filtering criteria based on the user's emotional state data and negativity score. The input data consists of the user's emotional state and the negativity score of the comments, which are combined and processed to determine the filtering criteria. The output is the filtering criteria.

[0219] Step 5:

[0220] The server uses filtering mechanisms to filter comments. The input consists of filtering criteria and parsed comments, and the output is the filtered comments. The server decides whether or not to display the comments based on the criteria and selects the filtered comments.

[0221] Step 6:

[0222] The server sends filtered comments to the terminal via a communication method. The input is the filtered comments, and the output is the comments displayed on the terminal. The terminal displays the received comments to the user, providing a stress-free SNS experience.

[0223] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0224] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0225] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0226] [Second Embodiment]

[0227] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0228] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0229] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0230] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0231] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0232] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0233] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0234] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0235] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0236] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0237] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0238] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0239] This invention is a filtering system for efficiently managing comments received by users on social networking services. Specific embodiments of this system are described below.

[0240] First, the user connects to the system via a dedicated terminal application and sets filtering criteria for incoming comments. These criteria include a threshold for negativity and specific expressions to be filtered.

[0241] Next, the server receives comment data flowing continuously from the SNS platform based on those criteria and passes the data to a dedicated natural language processing engine. This engine analyzes the context of the comments, specifically checking the structure and content of the language and the meaning of the words used.

[0242] The server then assigns a negativity score to the analyzed comment data. Here, it uses historical training data and machine learning models to evaluate how offensive or offensive each comment is considered to be.

[0243] Subsequently, the server's filtering mechanism refers to the criteria set by the user and filters out comments with high scores. Care is taken during this process to ensure that constructive critical comments are not affected.

[0244] The filtered results are sent from the server to the terminal, which then displays only the approved comments to the user. If the user feels that the comments do not match their filtering criteria, they are given the opportunity to adjust their settings.

[0245] Finally, the feedback is sent back to the server, and the machine learning model is updated using that information. This improves the overall filtering accuracy of the system, bringing it closer to the environment the user desires.

[0246] As a concrete example, consider a well-known company account that receives many comments about a particular product. This company wants to filter out negative expressions such as "terrible" or "bad," while receiving constructive feedback such as "there is room for improvement." This system would allow the company to eliminate unnecessary negative comments and utilize helpful feedback.

[0247] The following describes the processing flow.

[0248] Step 1:

[0249] Users launch a dedicated application on their device and set their own filtering criteria. These criteria include thresholds for the degree of negativity they want to filter out and specific keywords.

[0250] Step 2:

[0251] The server receives comment data flowing in real time from the social networking platform. The received data includes comment text and metadata.

[0252] Step 3:

[0253] The server sends the received comment data to a natural language processing engine. This engine analyzes the language syntax, vocabulary, and context, and converts the comments into a structured data format.

[0254] Step 4:

[0255] The server uses a machine learning model to assign negativity scores to structured comment data. This model quantifies aggression and unpleasantness based on training data.

[0256] Step 5:

[0257] The server filters comments based on the user's filtering criteria, referring to their negativity score. Comments exceeding the set threshold are filtered out, and unnecessary ones are excluded.

[0258] Step 6:

[0259] The server sends the allowed comment data from the filtered comments to the terminal.

[0260] Step 7:

[0261] The device displays received comments on the user's screen. Here, filtered negative comments are not displayed, and only comments deemed constructive are visualized.

[0262] Step 8:

[0263] Users review the displayed comments and make adjustments to their filtering settings if they feel they need to. This feedback is sent to the server via their device.

[0264] Step 9:

[0265] The server uses user feedback data to update its machine learning model and improve filtering accuracy.

[0266] (Example 1)

[0267] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0268] With the advancement of modern information and communication technology, users routinely receive large amounts of comments and information from various platforms. However, this information may include negative content or information unnecessary for the user, potentially impairing their comfortable user experience. Therefore, there is a need for means to appropriately filter the information users receive and eliminate unnecessary information.

[0269] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0270] In this invention, the server includes an information analysis means for analyzing data received from a user, an evaluation means for assigning an evaluation score to the analyzed information, and a selection means for selecting information based on the evaluation score according to user settings. This makes it possible to efficiently manage the information received by the user and accurately acquire only the important information.

[0271] "Information analysis means" refers to technologies that include the stage of analyzing data received from users and understanding its content and context.

[0272] "Evaluation methods" refer to techniques for quantifying the quality and attributes of analyzed information and assigning scores as evaluation criteria.

[0273] "Selection method" refers to a technology that uses assigned scores to select useful information and eliminate unnecessary information based on criteria set in advance by the user.

[0274] "Communication means" refers to the technologies and processes used to transmit selected information to a user's display device or other device.

[0275] "Learning methods" refer to technologies that enable a system to self-improve based on user feedback and setting changes, thereby improving the accuracy of information processing in subsequent instances.

[0276] The system for implementing this invention utilizes multiple electronic devices to provide information management as requested by the user. First, the user accesses the system through a dedicated terminal application and sets filtering criteria. These criteria include a threshold for the acceptable degree of negativity and the specification of specific expressions. The terminal application allows for easy modification and application of the criteria via a user interface.

[0277] The server receives the user's settings and continuously receives comment data from social media and other platforms. The received data is parsed using a natural language processing engine. This engine includes techniques to understand the context of the comments and analyze the content and nuances of the language. After analysis, the server uses a machine learning model to assign an appropriate negativeness score to the analysis results. This enables information filtering and allows comments to be filtered according to the user's settings.

[0278] Filtered information is sent from the server to the user's device, which displays only the selected information. During this process, if the user is dissatisfied with the displayed information, they can adjust the settings through feedback. This feedback is sent back to the server, and the machine learning model is updated based on that information. This improves the filtering accuracy for subsequent uses, providing the user with a more favorable information environment.

[0279] For example, when a user receives comments on a sports-related platform, they might want to avoid harsh criticism such as "XXX is the worst." By using a prompt in the format of "Please determine whether this comment is constructive according to user standards," the AI ​​model can determine whether the comment is constructive criticism or not and perform appropriate filtering.

[0280] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0281] Step 1:

[0282] The user sets the filtering criteria on their device.

[0283] The user sets the filtering criteria for the received information using a dedicated application. At this time, the user inputs the threshold of negativity and specific expressions to be filtered. The input setting data is sent to the server and saved as the criteria for the filtering process.

[0284] Step 2:

[0285] The server receives comment data.

[0286] The server continuously obtains comment data from SNS and other platforms. In this reception, a large amount of comment data is collected in real time via the API. The received data becomes the input for subsequent processing and is saved in the temporary storage within the server.

[0287] Step 3:

[0288] The server analyzes the data using a natural language processing engine.

[0289] The server passes the received comment data to the natural language processing engine. Here, the language structure, context, and meaning of the words in the data are analyzed. For example, negative expressions and contexts included in the text are identified. The result of the analysis is tagged and output as the basic data for calculating the evaluation score.

[0290] Step 4:

[0291] The server evaluates the degree of negativity and assigns a score.

[0292] Based on the analysis result, the server uses a machine learning model to assign a negativity score to the comment. Using an algorithm learned from past data, the degree to which each comment is negative is quantified. This score is output as the evaluation value of the comment.

[0293] Step 5:

[0294] The server filters comments based on the specified criteria.

[0295] Based on the assigned score, the server filters comments against the criteria set by the user. Negative comments exceeding the set threshold are excluded, and only positive or constructive feedback is selected. The filtered results are compiled into output data for the user.

[0296] Step 6:

[0297] The filtering results are sent from the server to the terminal.

[0298] The server sends filtered comments to the user's terminal. During transmission, data is transferred securely and quickly using a communication protocol. The comments are then output to the terminal as they should be displayed and provided to the user.

[0299] Step 7:

[0300] Users review the results and provide feedback.

[0301] Users review the comments displayed on their devices and provide feedback if they are dissatisfied with the filtering accuracy or wish to request adjustments. This feedback information is then sent back to the server and used as input data to improve the filtering criteria and machine learning models.

[0302] Step 8:

[0303] The server updates the machine learning model based on the feedback.

[0304] Based on the feedback, the server updates its machine learning model to improve filtering accuracy. This enhances its ability to handle new data analysis and filtering criteria, enabling more efficient information selection.

[0305] (Application Example 1)

[0306] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[0307] With the increase in comments on the online platform, it has become difficult for administrators to efficiently filter inappropriate or wasteful negative comments and appropriately pick up useful improvement suggestions. In such a situation, it is required to accurately evaluate the nature of the comments and realize the filtering desired by the users.

[0308] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0309] In this invention, the server includes natural language processing means for analyzing communication data received from a user, evaluation means for assigning a sentiment score to the analyzed communication data, and selection means for filtering communication based on the evaluation value of the sentiment according to user settings. Thereby, inefficient comment processing is improved, and communication that emphasizes useful feedback becomes possible.

[0310] The "natural language processing means" is a technology that analyzes communication data received from a user, understands its context and vocabulary, and structures it.

[0311] The "evaluation means" is to judge the negativity or positivity of a comment by quantifying the degree of sentiment and expression for the analyzed communication data.

[0312] The "selection means" is a technology that filters communication data according to the evaluation value of the sentiment based on the criteria set by the user and selects comments that exceed specific criteria.

[0313] "Transmission means" refers to the function of sending filtered communication data to the user's terminal, and is responsible for ensuring that only information useful to the user is provided.

[0314] A "learning tool" is a system that continuously updates machine learning models and improves filtering accuracy based on changes in user selection settings and feedback.

[0315] "Receiving means" refers to technology that identifies communications pointing out new areas for improvement and makes it easy for users to receive and view them.

[0316] The system for implementing this invention consists of a server and a user terminal. The server analyzes, evaluates, and sorts communication data using a natural language processing engine and a machine learning model. Specifically, it uses natural language processing libraries such as "spaCy" to perform language analysis and sentiment evaluation. Machine learning frameworks such as "scikit-learn" and "TensorFlow" are used for scoring sentiment levels. Through these technologies, the server detects and sorts negative or positive comments.

[0317] Furthermore, the server is equipped with a means to update the learning model based on user feedback, thereby improving filtering accuracy. Specifically, comments that can be evaluated are identified based on user settings, and the selected comments are transmitted to the terminal. This makes it easy for users to access only the useful feedback from the comments they receive.

[0318] The user's device sends comment data to the server in real time via the SNS API. The server evaluates and filters the received comment data and returns the filtered results to the device. Users can further improve the accuracy of the filtering by adjusting their selection settings.

[0319] As a concrete example, a restaurant chain could use this system when launching a social media campaign for a new dish to filter out critical comments such as "it's not tasty" and receive specific improvement suggestions such as "it would be better if the spices were used sparingly." This would allow them to quickly provide services that meet customer needs. An example of a prompt to input into the generating AI model is: "Design a social media filtering app that efficiently manages comments received on a social media account. It should eliminate negative comments based on user settings and display only helpful feedback."

[0320] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0321] Step 1:

[0322] The user's terminal retrieves comment data through the API of a social networking service. The retrieved comment data is sent to the server in its original form. The input is comment data obtained from the SNS, and the output is data transfer to the server. Specifically, the system retrieves data using an HTTP request, analyzes the response, and extracts the necessary comments.

[0323] Step 2:

[0324] The server receives comment data sent from the user terminal. This data is input into a natural language processing engine for content analysis. The input is the comment data received from the user terminal, and the output is the analyzed comment data. Specifically, the "spaCy" library is used to analyze the grammatical structure and keywords of the comments.

[0325] Step 3:

[0326] The server assigns a sentiment score to each comment based on the analysis results. Machine learning tools such as "TensorFlow" and "scikit-learn" are used for this scoring. The input is the analyzed comment data, and the output is the comment data with the sentiment score assigned. Specifically, the scoring is performed using a model based on past training data.

[0327] Step 4:

[0328] The server filters comment data according to sentiment scores based on criteria pre-configured by the user. The input is scored comment data, and the output is filtered comments. Specifically, it excludes comments with scores above a set threshold and selects only permitted comments.

[0329] Step 5:

[0330] The filtered comment data is sent from the server to the user's terminal. The user's terminal receives this data and displays it on the screen. The input is filtered comment data, and the output is the display on the user's terminal. Specifically, the data is received in HTML or JSON format and rendered on the user interface.

[0331] Step 6:

[0332] Users review the display results of received comments and adjust the settings if they feel the filtering criteria are inappropriate. This feedback is sent to the server and used to update the model. The input is the user's feedback, and the output is the updated filtering criteria. Specifically, the criteria are adjusted from the settings screen and saved to the database.

[0333] Step 7:

[0334] The server updates its machine learning model based on user feedback, improving filtering accuracy. The input is the user's adjusted criteria and feedback, and the output is the updated model. Specifically, it uses the collected feedback data as new training data to retrain the generative AI model.

[0335] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0336] This invention is a filtering system for efficiently managing comments received by users when using social networking services, based on the user's emotional state. The system incorporates four main methods: natural language processing, scoring, filtering, and an emotion engine.

[0337] First, users can access the system through a dedicated terminal application and set filtering criteria that include their own emotional data. This setting utilizes real-time emotional state data obtained from the emotion engine. For example, users can set stricter filtering criteria when they are experiencing high stress levels.

[0338] Next, the server aggregates comment data received from the SNS platform in real time and analyzes it using a natural language processing engine. This analysis confirms the context and vocabulary of the comments and converts them into structured data.

[0339] The server assigns a negativity score to the analyzed comment data. This scoring process is performed by a machine learning model that accurately assesses the degree of negativity based on past data.

[0340] Subsequently, the emotion engine is utilized, and the server analyzes the user's emotional state and dynamically incorporates this data into the filtering criteria. This allows the filtering criteria to be automatically adjusted according to the user's current emotional state.

[0341] Filtered comments are sent from the server to the terminal, which then displays or hides the filtered content on the user's screen. The system prioritizes comments that are sensitive to the user's emotions. For example, when the user is emotionally positive, some critical comments are displayed; conversely, when the user is emotionally negative and easily stressed, potentially stressful comments are filtered more strictly.

[0342] In this way, users can comfortably use social media while taking their own emotional state into consideration. This invention protects users' mental health and ensures that only constructive opinions influence their lives.

[0343] The following describes the processing flow.

[0344] Step 1:

[0345] The user launches a dedicated application on their device and provides the system with their emotional state, either by inputting it or by providing emotional data obtained from sensors. This data is then used for comment filtering.

[0346] Step 2:

[0347] The server receives comment data flowing in from the social networking platform in real time. This data includes metadata such as the comments themselves and the person who posted them.

[0348] Step 3:

[0349] The server sends the received comment data to a natural language processing engine. This engine performs analysis and tokenization and part-of-speech tagging to understand the linguistic structure and content of the comments.

[0350] Step 4:

[0351] The server passes the analyzed comment data to a scoring system, where it assigns a score for negativity. This scoring is performed using a machine learning algorithm and is based on past training data.

[0352] Step 5:

[0353] The server uses an emotion engine to assess the user's emotional state and dynamically adjusts filtering criteria based on that data. If the user is in a negative emotional state, the filtering criteria are set to become stricter.

[0354] Step 6:

[0355] The server selects comments filtered based on scoring, removing those exceeding a threshold. Constructive criticism is retained according to user preferences.

[0356] Step 7:

[0357] The server sends the filtered comment data to the terminal. The terminal then selects only the comments to display on the user screen based on the filtered results.

[0358] Step 8:

[0359] Users can review the comments displayed on their device and provide feedback to change settings if they are dissatisfied with the filtering results. This feedback will be used to improve the filtering accuracy in the future.

[0360] Step 9:

[0361] The server collects user feedback and uses it to update machine learning models and sentiment engines, thereby improving the overall accuracy of the system.

[0362] (Example 2)

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

[0364] In today's social media environment, comments received by users can affect their mental health. For users who are particularly sensitive to negative comments, such comments can be a source of stress. The objective of this invention is to enable users to comfortably use social networks while protecting their mental health by filtering comments according to their emotional state.

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

[0366] In this invention, the server includes emotion analysis means for analyzing emotion data received from the user to determine the emotional state, criteria adjustment means for adjusting filtering criteria based on the determined emotional state, and natural language processing means for analyzing received comment data. This enables dynamic filtering according to the user's emotional state.

[0367] "Emotional analysis means" refers to a technological element that analyzes emotional data received from a user and determines the user's emotional state.

[0368] A "criteria adjustment means" is a technical element that dynamically adjusts the criteria for comment filtering based on the user's emotional state determined by the sentiment analysis means.

[0369] "Natural language processing means" refers to technical elements that analyze received comment data and extract context and emotional nuances.

[0370] A "scoring method" is a technical element that assigns a score of negativity to analyzed comment data and performs evaluation based on that score.

[0371] A "filtering mechanism" is a technical element that selects and determines whether or not to display comments based on adjusted filtering criteria.

[0372] "Transmission means" refers to the technical element that transfers filtered comments to the device used by the user.

[0373] This invention is a system that filters comments on social networks according to the user's emotional state, enabling the user to receive information comfortably.

[0374] The system first determines the user's emotional state using emotion analysis tools. Users can provide emotional data through a terminal application using biometric sensors and a user interface. This emotional data can be acquired, for example, from a heart rate monitor or facial recognition camera.

[0375] The sentiment analysis method utilizes machine learning algorithms and sentiment analysis libraries to analyze this data. The server uses the algorithm to classify the user's current emotional state into categories such as positive or negative.

[0376] Next, the criteria adjustment mechanism dynamically adjusts the filtering criteria based on the determined emotional state. This adjustment sets the strictness of filtering for negative comments to match the user's emotions.

[0377] The server analyzes comments collected from social media using natural language processing tools to detect the context and emotional nuances of the comments. This structures the comment data, and a negativeness score is assigned via a scoring mechanism. For example, a machine learning platform is used for evaluation.

[0378] The filtering mechanism selects which comments to display based on the negativity score and the user's emotional state. The filtered comments are then transmitted to the device via a transmission mechanism, where the device decides whether to display or hide them. This process ensures that the information the user receives is tailored to their emotional state at that time.

[0379] For example, when a user is relaxed, even slightly critical comments will be displayed without issue, but when they are under a lot of stress, comments that could cause stress will be strictly filtered out.

[0380] The prompts for the generating AI model are entered in the format of, for example, "Explain how to filter received comments when the user is in a stressful state." These prompts enable the system to implement specific filtering that takes the user's state into consideration.

[0381] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0382] Step 1:

[0383] Users input emotional data via their devices. Sensors such as heart rate monitors and facial recognition cameras are used for input. This allows the system to obtain the user's emotional state as numerical data. The server receives this data and stores it as initial data.

[0384] Step 2:

[0385] The server analyzes the received emotional data using emotion analysis tools. This analysis utilizes emotion analysis libraries and machine learning algorithms. Through this analysis, the emotional data is classified into categories such as "positive" and "negative." Based on these emotional states, the server sets filtering criteria for each user.

[0386] Step 3:

[0387] The server collects relevant comment data through the SNS API. The collected comments are stored in the database in their original form. This prepares the comment information for analysis.

[0388] Step 4:

[0389] The server uses natural language processing to analyze the collected comment data. This analysis extracts contextual and emotional nuances from vocabulary. The analysis results are output as structured data and passed on to the next scoring process.

[0390] Step 5:

[0391] The server assigns a negativity score to the analyzed comments using a scoring mechanism. Here, a predictive model based on historical data is applied using a generative AI model. The processed scoring data, with scores expressed in numerical format, is used as a filtering criterion.

[0392] Step 6:

[0393] The server applies filtering criteria based on the user's emotional state via a criteria adjustment mechanism. This dynamically filters the collected comments. Based on these criteria, comments with high negative scores are hidden, and only comments within an acceptable range are displayed.

[0394] Step 7:

[0395] The server delivers filtered comments to the user's terminal via a transmission method. The terminal receives these comments and displays them on the screen. For some hidden comments, the user is notified and given the option to display them if desired.

[0396] This series of steps allows users to view comments that take their current emotional state into consideration, enabling them to use social media in a way that respects their mental health.

[0397] (Application Example 2)

[0398] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0399] Traditional social networking services lack comment management that takes into account users' emotional states, potentially harming users' mental health. Furthermore, their filtering accuracy is fixed and inadequate, failing to adapt to real-time changes in users' emotions.

[0400] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0401] In this invention, the server includes an electronic computing means for analyzing user comment data, an evaluation means for assigning an evaluation to the analysis results, and an emotion analysis means for analyzing the user's emotional state and changing the filtering criteria. This enables comment management according to the user's emotional state, and allows for more flexible and improved filtering accuracy.

[0402] "Electronic computing means" refers to digital computing devices and their functions used to process and analyze information.

[0403] "Evaluation means" refers to a function or device that performs a process of assigning scores or ratings to data using specific criteria.

[0404] "Emotional analysis means" refers to a function or device that can analyze a user's emotional state in real time and dynamically change the criteria for data processing based on that analysis.

[0405] "Filtering means" refers to a function or device that selects information from data that meets specific criteria and displays or blocks it as needed.

[0406] "Communication means" refers to protocols, devices, or functions used to send and receive data.

[0407] The system implementing this invention provides a multi-stage process for managing comments in order to make the user's social networking service experience more comfortable. The system operates in a network environment that includes user terminals and servers.

[0408] The server receives comment data and uses electronic computing tools to analyze it. These tools employ natural language processing algorithms to analyze the context and vocabulary of the data. For example, technologies such as TensorFlow and NLTK can be used to effectively analyze text data.

[0409] Next, the server uses an evaluation tool to assign a negativity score to the analyzed comments. This scoring uses historical data and a learned model (e.g., a machine learning model) to assess the negativity of the comments.

[0410] The emotion analysis system analyzes the user's emotional state in real time. Using hardware such as voice sensors (e.g., Shure MOTIV) and video sensors (e.g., Intel RealSense camera), it detects the user's facial expressions and tone of voice to acquire their emotional state. This data is then incorporated as filtering criteria by a filtering system. This allows the server to filter comments based on criteria that reflect the user's real-time emotional state.

[0411] The filtered comments are ultimately sent to the device via communication and displayed to the user. Because this display takes the user's emotional state into consideration, it can provide a more stress-free social media experience.

[0412] For example, when a user is relaxed, many comments, including minor criticisms, will be displayed, but when they are tired, only positive comments will be prioritized and displayed.

[0413] An example of a prompt to input into the generating AI model is as follows: "We are developing a system that filters comments based on their emotional state. It will manage comments in real time based on user emotional data. Please provide helpful suggestions."

[0414] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0415] Step 1:

[0416] The server receives comment data from a social networking service. The input data consists of unparsed text comments, which are then sent to a natural language processing engine. The server uses this engine to analyze the context and vocabulary of the comments and generate structured data.

[0417] Step 2:

[0418] The server assigns a negativity score to the analyzed comment data using an evaluation method. The input data is the analyzed text, and the output data is the negativity score for each comment. This process is performed by calculating the negativity score from past data using a learning model.

[0419] Step 3:

[0420] The system acquires the user's emotional state in real time using voice and video sensors. The input is the user's voice and video data, and the output is the analyzed emotional state data. The server analyzes this data using emotion analysis tools and uses the results as filtering criteria for the next step.

[0421] Step 4:

[0422] The server sets filtering criteria based on the user's emotional state data and negativity score. The input data consists of the user's emotional state and the negativity score of the comments, which are combined and processed to determine the filtering criteria. The output is the filtering criteria.

[0423] Step 5:

[0424] The server uses filtering mechanisms to filter comments. The input consists of filtering criteria and parsed comments, and the output is the filtered comments. The server decides whether or not to display the comments based on the criteria and selects the filtered comments.

[0425] Step 6:

[0426] The server sends filtered comments to the terminal via a communication method. The input is the filtered comments, and the output is the comments displayed on the terminal. The terminal displays the received comments to the user, providing a stress-free SNS experience.

[0427] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0428] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0429] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0430] [Third Embodiment]

[0431] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0432] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0433] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0434] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0435] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0436] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0437] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0438] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0439] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0440] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0441] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0442] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0443] This invention is a filtering system for efficiently managing comments received by users on social networking services. Specific embodiments of this system are described below.

[0444] First, the user connects to the system via a dedicated terminal application and sets filtering criteria for incoming comments. These criteria include a threshold for negativity and specific expressions to be filtered.

[0445] Next, the server receives comment data flowing continuously from the SNS platform based on those criteria and passes the data to a dedicated natural language processing engine. This engine analyzes the context of the comments, specifically checking the structure and content of the language and the meaning of the words used.

[0446] The server then assigns a negativity score to the analyzed comment data. Here, it uses historical training data and machine learning models to evaluate how offensive or offensive each comment is considered to be.

[0447] Subsequently, the server's filtering mechanism refers to the criteria set by the user and filters out comments with high scores. Care is taken during this process to ensure that constructive critical comments are not affected.

[0448] The filtered results are sent from the server to the terminal, which then displays only the approved comments to the user. If the user feels that the comments do not match their filtering criteria, they are given the opportunity to adjust their settings.

[0449] Finally, the feedback is sent back to the server, and the machine learning model is updated using that information. This improves the overall filtering accuracy of the system, bringing it closer to the environment the user desires.

[0450] As a concrete example, consider a well-known company account that receives many comments about a particular product. This company wants to filter out negative expressions such as "terrible" or "bad," while receiving constructive feedback such as "there is room for improvement." This system would allow the company to eliminate unnecessary negative comments and utilize helpful feedback.

[0451] The following describes the processing flow.

[0452] Step 1:

[0453] Users launch a dedicated application on their device and set their own filtering criteria. These criteria include thresholds for the degree of negativity they want to filter out and specific keywords.

[0454] Step 2:

[0455] The server receives comment data flowing in real time from the social networking platform. The received data includes comment text and metadata.

[0456] Step 3:

[0457] The server sends the received comment data to a natural language processing engine. This engine analyzes the language syntax, vocabulary, and context, and converts the comments into a structured data format.

[0458] Step 4:

[0459] The server uses a machine learning model to assign negativity scores to structured comment data. This model quantifies aggression and unpleasantness based on training data.

[0460] Step 5:

[0461] The server filters comments based on the user's filtering criteria, referring to their negativity score. Comments exceeding the set threshold are filtered out, and unnecessary ones are excluded.

[0462] Step 6:

[0463] The server sends the allowed comment data from the filtered comments to the terminal.

[0464] Step 7:

[0465] The device displays received comments on the user's screen. Here, filtered negative comments are not displayed, and only comments deemed constructive are visualized.

[0466] Step 8:

[0467] Users review the displayed comments and make adjustments to their filtering settings if they feel they need to. This feedback is sent to the server via their device.

[0468] Step 9:

[0469] The server uses user feedback data to update its machine learning model and improve filtering accuracy.

[0470] (Example 1)

[0471] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0472] With the advancement of modern information and communication technology, users routinely receive large amounts of comments and information from various platforms. However, this information may include negative content or information unnecessary for the user, potentially impairing their comfortable user experience. Therefore, there is a need for means to appropriately filter the information users receive and eliminate unnecessary information.

[0473] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0474] In this invention, the server includes an information analysis means for analyzing data received from a user, an evaluation means for assigning an evaluation score to the analyzed information, and a selection means for selecting information based on the evaluation score according to user settings. This makes it possible to efficiently manage the information received by the user and accurately acquire only the important information.

[0475] "Information analysis means" refers to technologies that include the stage of analyzing data received from users and understanding its content and context.

[0476] "Evaluation methods" refer to techniques for quantifying the quality and attributes of analyzed information and assigning scores as evaluation criteria.

[0477] "Selection method" refers to a technology that uses assigned scores to select useful information and eliminate unnecessary information based on criteria set in advance by the user.

[0478] "Communication means" refers to the technologies and processes used to transmit selected information to a user's display device or other device.

[0479] "Learning methods" refer to technologies that enable a system to self-improve based on user feedback and setting changes, thereby improving the accuracy of information processing in subsequent instances.

[0480] The system for implementing this invention utilizes multiple electronic devices to provide information management as requested by the user. First, the user accesses the system through a dedicated terminal application and sets filtering criteria. These criteria include a threshold for the acceptable degree of negativity and the specification of specific expressions. The terminal application allows for easy modification and application of the criteria via a user interface.

[0481] The server receives the user's settings and continuously receives comment data from social media and other platforms. The received data is parsed using a natural language processing engine. This engine includes techniques to understand the context of the comments and analyze the content and nuances of the language. After analysis, the server uses a machine learning model to assign an appropriate negativeness score to the analysis results. This enables information filtering and allows comments to be filtered according to the user's settings.

[0482] Filtered information is sent from the server to the user's device, which displays only the selected information. During this process, if the user is dissatisfied with the displayed information, they can adjust the settings through feedback. This feedback is sent back to the server, and the machine learning model is updated based on that information. This improves the filtering accuracy for subsequent uses, providing the user with a more favorable information environment.

[0483] For example, when a user receives comments on a sports-related platform, they might want to avoid harsh criticism such as "XXX is the worst." By using a prompt in the format of "Please determine whether this comment is constructive according to user standards," the AI ​​model can determine whether the comment is constructive criticism or not and perform appropriate filtering.

[0484] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0485] Step 1:

[0486] The user sets the filtering criteria on their device.

[0487] Users use a dedicated application to set filtering criteria for incoming information. In this process, users input a threshold for negativity and specific expressions they wish to filter. The entered settings data is sent to the server and stored as the basis for the filtering process.

[0488] Step 2:

[0489] The server receives the comment data.

[0490] The server continuously retrieves comment data from social media and other platforms. This reception involves collecting large amounts of comment data in real time via APIs. The received data serves as input for subsequent processing and is stored in temporary storage on the server.

[0491] Step 3:

[0492] The server analyzes the data using a natural language processing engine.

[0493] The server passes the received comment data to a natural language processing engine. Here, the linguistic structure, context, and word meanings of the data are analyzed. For example, negative expressions and contexts within the text are identified. The analysis results are tagged and output as the basis for calculating evaluation scores.

[0494] Step 4:

[0495] The server evaluates the degree of negativity and assigns a score.

[0496] Based on the analysis results, the server uses a machine learning model to assign a negativity score to each comment. Using an algorithm trained on past data, it quantifies how negative each comment is. This score is then output as the comment's evaluation value.

[0497] Step 5:

[0498] The server filters comments based on the specified criteria.

[0499] Based on the assigned score, the server filters comments against the criteria set by the user. Negative comments exceeding the set threshold are excluded, and only positive or constructive feedback is selected. The filtered results are compiled into output data for the user.

[0500] Step 6:

[0501] The filtering results are sent from the server to the terminal.

[0502] The server sends filtered comments to the user's terminal. During transmission, data is transferred securely and quickly using a communication protocol. The comments are then output to the terminal as they should be displayed and provided to the user.

[0503] Step 7:

[0504] Users review the results and provide feedback.

[0505] Users review the comments displayed on their devices and provide feedback if they are dissatisfied with the filtering accuracy or wish to request adjustments. This feedback information is then sent back to the server and used as input data to improve the filtering criteria and machine learning models.

[0506] Step 8:

[0507] The server updates the machine learning model based on the feedback.

[0508] Based on the feedback, the server updates its machine learning model to improve filtering accuracy. This enhances its ability to handle new data analysis and filtering criteria, enabling more efficient information selection.

[0509] (Application Example 1)

[0510] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0511] With the increasing number of comments on online platforms, it has become difficult for administrators to efficiently filter out inappropriate or pointless negative comments and appropriately identify helpful improvement suggestions. In this situation, there is a need to accurately evaluate the nature of comments and implement filtering that meets the user's expectations.

[0512] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0513] In this invention, the server includes a natural language processing means for analyzing communication data received from a user, an evaluation means for assigning an emotional score to the analyzed communication data, and a selection means for filtering communications based on the emotional score evaluation value according to user settings. This improves inefficient comment processing and enables communication that emphasizes useful feedback.

[0514] "Natural language processing means" refers to technologies that analyze communication data received from users, understand its context and vocabulary, and structure it.

[0515] The "evaluation method" involves quantifying the degree of emotion and expression in the analyzed communication data to determine the degree of negativity or positivity of the comments.

[0516] A "selection method" is a technology that filters communication data according to the emotional intensity evaluation value based on criteria set by the user, and selects comments that exceed a specific standard.

[0517] "Transmission means" refers to the function of sending filtered communication data to the user's terminal, and is responsible for ensuring that only information useful to the user is provided.

[0518] A "learning tool" is a system that continuously updates machine learning models and improves filtering accuracy based on changes in user selection settings and feedback.

[0519] "Receiving means" refers to technology that identifies communications pointing out new areas for improvement and makes it easy for users to receive and view them.

[0520] The system for implementing this invention consists of a server and a user terminal. The server analyzes, evaluates, and sorts communication data using a natural language processing engine and a machine learning model. Specifically, it uses natural language processing libraries such as "spaCy" to perform language analysis and sentiment evaluation. Machine learning frameworks such as "scikit-learn" and "TensorFlow" are used for scoring sentiment levels. Through these technologies, the server detects and sorts negative or positive comments.

[0521] Furthermore, the server is equipped with a means to update the learning model based on user feedback, thereby improving filtering accuracy. Specifically, comments that can be evaluated are identified based on user settings, and the selected comments are transmitted to the terminal. This makes it easy for users to access only the useful feedback from the comments they receive.

[0522] The user's device sends comment data to the server in real time via the SNS API. The server evaluates and filters the received comment data and returns the filtered results to the device. Users can further improve the accuracy of the filtering by adjusting their selection settings.

[0523] As a concrete example, a restaurant chain could use this system when launching a social media campaign for a new dish to filter out critical comments such as "it's not tasty" and receive specific improvement suggestions such as "it would be better if the spices were used sparingly." This would allow them to quickly provide services that meet customer needs. An example of a prompt to input into the generating AI model is: "Design a social media filtering app that efficiently manages comments received on a social media account. It should eliminate negative comments based on user settings and display only helpful feedback."

[0524] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0525] Step 1:

[0526] The user's terminal retrieves comment data through the API of a social networking service. The retrieved comment data is sent to the server in its original form. The input is comment data obtained from the SNS, and the output is data transfer to the server. Specifically, the system retrieves data using an HTTP request, analyzes the response, and extracts the necessary comments.

[0527] Step 2:

[0528] The server receives comment data sent from the user terminal. This data is input into a natural language processing engine for content analysis. The input is the comment data received from the user terminal, and the output is the analyzed comment data. Specifically, the "spaCy" library is used to analyze the grammatical structure and keywords of the comments.

[0529] Step 3:

[0530] The server assigns a sentiment score to each comment based on the analysis results. Machine learning tools such as "TensorFlow" and "scikit-learn" are used for this scoring. The input is the analyzed comment data, and the output is the comment data with the sentiment score assigned. Specifically, the scoring is performed using a model based on past training data.

[0531] Step 4:

[0532] The server filters comment data according to sentiment scores based on criteria pre-configured by the user. The input is scored comment data, and the output is filtered comments. Specifically, it excludes comments with scores above a set threshold and selects only permitted comments.

[0533] Step 5:

[0534] The filtered comment data is sent from the server to the user's terminal. The user's terminal receives this data and displays it on the screen. The input is filtered comment data, and the output is the display on the user's terminal. Specifically, the data is received in HTML or JSON format and rendered on the user interface.

[0535] Step 6:

[0536] Users review the display results of received comments and adjust the settings if they feel the filtering criteria are inappropriate. This feedback is sent to the server and used to update the model. The input is the user's feedback, and the output is the updated filtering criteria. Specifically, the criteria are adjusted from the settings screen and saved to the database.

[0537] Step 7:

[0538] The server updates its machine learning model based on user feedback, improving filtering accuracy. The input is the user's adjusted criteria and feedback, and the output is the updated model. Specifically, it uses the collected feedback data as new training data to retrain the generative AI model.

[0539] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0540] This invention is a filtering system for efficiently managing comments received by users when using social networking services, based on the user's emotional state. The system incorporates four main methods: natural language processing, scoring, filtering, and an emotion engine.

[0541] First, users can access the system through a dedicated terminal application and set filtering criteria that include their own emotional data. This setting utilizes real-time emotional state data obtained from the emotion engine. For example, users can set stricter filtering criteria when they are experiencing high stress levels.

[0542] Next, the server aggregates comment data received from the SNS platform in real time and analyzes it using a natural language processing engine. This analysis confirms the context and vocabulary of the comments and converts them into structured data.

[0543] The server assigns a negativity score to the analyzed comment data. This scoring process is performed by a machine learning model that accurately assesses the degree of negativity based on past data.

[0544] Subsequently, the emotion engine is utilized, and the server analyzes the user's emotional state and dynamically incorporates this data into the filtering criteria. This allows the filtering criteria to be automatically adjusted according to the user's current emotional state.

[0545] Filtered comments are sent from the server to the terminal, which then displays or hides the filtered content on the user's screen. The system prioritizes comments that are sensitive to the user's emotions. For example, when the user is emotionally positive, some critical comments are displayed; conversely, when the user is emotionally negative and easily stressed, potentially stressful comments are filtered more strictly.

[0546] In this way, users can comfortably use social media while taking their own emotional state into consideration. This invention protects users' mental health and ensures that only constructive opinions influence their lives.

[0547] The following describes the processing flow.

[0548] Step 1:

[0549] The user launches a dedicated application on their device and provides the system with their emotional state, either by inputting it or by providing emotional data obtained from sensors. This data is then used for comment filtering.

[0550] Step 2:

[0551] The server receives comment data flowing in from the social networking platform in real time. This data includes metadata such as the comments themselves and the person who posted them.

[0552] Step 3:

[0553] The server sends the received comment data to a natural language processing engine. This engine performs analysis and tokenization and part-of-speech tagging to understand the linguistic structure and content of the comments.

[0554] Step 4:

[0555] The server passes the analyzed comment data to a scoring system, where it assigns a score for negativity. This scoring is performed using a machine learning algorithm and is based on past training data.

[0556] Step 5:

[0557] The server uses an emotion engine to assess the user's emotional state and dynamically adjusts filtering criteria based on that data. If the user is in a negative emotional state, the filtering criteria are set to become stricter.

[0558] Step 6:

[0559] The server selects comments filtered based on scoring, removing those exceeding a threshold. Constructive criticism is retained according to user preferences.

[0560] Step 7:

[0561] The server sends the filtered comment data to the terminal. The terminal then selects only the comments to display on the user screen based on the filtered results.

[0562] Step 8:

[0563] Users can review the comments displayed on their device and provide feedback to change settings if they are dissatisfied with the filtering results. This feedback will be used to improve the filtering accuracy in the future.

[0564] Step 9:

[0565] The server collects user feedback and uses it to update machine learning models and sentiment engines, thereby improving the overall accuracy of the system.

[0566] (Example 2)

[0567] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0568] In today's social media environment, comments received by users can affect their mental health. For users who are particularly sensitive to negative comments, such comments can be a source of stress. The objective of this invention is to enable users to comfortably use social networks while protecting their mental health by filtering comments according to their emotional state.

[0569] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0570] In this invention, the server includes emotion analysis means for analyzing emotion data received from the user to determine the emotional state, criteria adjustment means for adjusting filtering criteria based on the determined emotional state, and natural language processing means for analyzing received comment data. This enables dynamic filtering according to the user's emotional state.

[0571] "Emotional analysis means" refers to a technological element that analyzes emotional data received from a user and determines the user's emotional state.

[0572] A "criteria adjustment means" is a technical element that dynamically adjusts the criteria for comment filtering based on the user's emotional state determined by the sentiment analysis means.

[0573] "Natural language processing means" refers to technical elements that analyze received comment data and extract context and emotional nuances.

[0574] A "scoring method" is a technical element that assigns a score of negativity to analyzed comment data and performs evaluation based on that score.

[0575] A "filtering mechanism" is a technical element that selects and determines whether or not to display comments based on adjusted filtering criteria.

[0576] "Transmission means" refers to the technical element that transfers filtered comments to the device used by the user.

[0577] This invention is a system that filters comments on social networks according to the user's emotional state, enabling the user to receive information comfortably.

[0578] The system first determines the user's emotional state using emotion analysis tools. Users can provide emotional data through a terminal application using biometric sensors and a user interface. This emotional data can be acquired, for example, from a heart rate monitor or facial recognition camera.

[0579] The sentiment analysis method utilizes machine learning algorithms and sentiment analysis libraries to analyze this data. The server uses the algorithm to classify the user's current emotional state into categories such as positive or negative.

[0580] Next, the criteria adjustment mechanism dynamically adjusts the filtering criteria based on the determined emotional state. This adjustment sets the strictness of filtering for negative comments to match the user's emotions.

[0581] The server analyzes comments collected from social media using natural language processing tools to detect the context and emotional nuances of the comments. This structures the comment data, and a negativeness score is assigned via a scoring mechanism. For example, a machine learning platform is used for evaluation.

[0582] The filtering mechanism selects which comments to display based on the negativity score and the user's emotional state. The filtered comments are then transmitted to the device via a transmission mechanism, where the device decides whether to display or hide them. This process ensures that the information the user receives is tailored to their emotional state at that time.

[0583] For example, when a user is relaxed, even slightly critical comments will be displayed without issue, but when they are under a lot of stress, comments that could cause stress will be strictly filtered out.

[0584] The prompts for the generating AI model are entered in the format of, for example, "Explain how to filter received comments when the user is in a stressful state." These prompts enable the system to implement specific filtering that takes the user's state into consideration.

[0585] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0586] Step 1:

[0587] Users input emotional data via their devices. Sensors such as heart rate monitors and facial recognition cameras are used for input. This allows the system to obtain the user's emotional state as numerical data. The server receives this data and stores it as initial data.

[0588] Step 2:

[0589] The server analyzes the received emotional data using emotion analysis tools. This analysis utilizes emotion analysis libraries and machine learning algorithms. Through this analysis, the emotional data is classified into categories such as "positive" and "negative." Based on these emotional states, the server sets filtering criteria for each user.

[0590] Step 3:

[0591] The server collects relevant comment data through the SNS API. The collected comments are stored in the database in their original form. This prepares the comment information for analysis.

[0592] Step 4:

[0593] The server uses natural language processing to analyze the collected comment data. This analysis extracts contextual and emotional nuances from vocabulary. The analysis results are output as structured data and passed on to the next scoring process.

[0594] Step 5:

[0595] The server assigns a negativity score to the analyzed comments using a scoring mechanism. Here, a predictive model based on historical data is applied using a generative AI model. The processed scoring data, with scores expressed in numerical format, is used as a filtering criterion.

[0596] Step 6:

[0597] The server applies filtering criteria based on the user's emotional state via a criteria adjustment mechanism. This dynamically filters the collected comments. Based on these criteria, comments with high negative scores are hidden, and only comments within an acceptable range are displayed.

[0598] Step 7:

[0599] The server delivers filtered comments to the user's terminal via a transmission method. The terminal receives these comments and displays them on the screen. For some hidden comments, the user is notified and given the option to display them if desired.

[0600] This series of steps allows users to view comments that take their current emotional state into consideration, enabling them to use social media in a way that respects their mental health.

[0601] (Application Example 2)

[0602] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0603] Traditional social networking services lack comment management that takes into account users' emotional states, potentially harming users' mental health. Furthermore, their filtering accuracy is fixed and inadequate, failing to adapt to real-time changes in users' emotions.

[0604] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0605] In this invention, the server includes an electronic computing means for analyzing user comment data, an evaluation means for assigning an evaluation to the analysis results, and an emotion analysis means for analyzing the user's emotional state and changing the filtering criteria. This enables comment management according to the user's emotional state, and allows for more flexible and improved filtering accuracy.

[0606] "Electronic computing means" refers to digital computing devices and their functions used to process and analyze information.

[0607] "Evaluation means" refers to a function or device that performs a process of assigning scores or ratings to data using specific criteria.

[0608] "Emotional analysis means" refers to a function or device that can analyze a user's emotional state in real time and dynamically change the criteria for data processing based on that analysis.

[0609] "Filtering means" refers to a function or device that selects information from data that meets specific criteria and displays or blocks it as needed.

[0610] "Communication means" refers to protocols, devices, or functions used to send and receive data.

[0611] The system implementing this invention provides a multi-stage process for managing comments in order to make the user's social networking service experience more comfortable. The system operates in a network environment that includes user terminals and servers.

[0612] The server receives comment data and uses electronic computing tools to analyze it. These tools employ natural language processing algorithms to analyze the context and vocabulary of the data. For example, technologies such as TensorFlow and NLTK can be used to effectively analyze text data.

[0613] Next, the server uses an evaluation tool to assign a negativity score to the analyzed comments. This scoring uses historical data and a learned model (e.g., a machine learning model) to assess the negativity of the comments.

[0614] The emotion analysis system analyzes the user's emotional state in real time. Using hardware such as voice sensors (e.g., Shure MOTIV) and video sensors (e.g., Intel RealSense camera), it detects the user's facial expressions and tone of voice to acquire their emotional state. This data is then incorporated as filtering criteria by a filtering system. This allows the server to filter comments based on criteria that reflect the user's real-time emotional state.

[0615] The filtered comments are ultimately sent to the device via communication and displayed to the user. Because this display takes the user's emotional state into consideration, it can provide a more stress-free social media experience.

[0616] For example, when a user is relaxed, many comments, including minor criticisms, will be displayed, but when they are tired, only positive comments will be prioritized and displayed.

[0617] An example of a prompt to input into the generating AI model is as follows: "We are developing a system that filters comments based on their emotional state. It will manage comments in real time based on user emotional data. Please provide helpful suggestions."

[0618] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0619] Step 1:

[0620] The server receives comment data from a social networking service. The input data consists of unparsed text comments, which are then sent to a natural language processing engine. The server uses this engine to analyze the context and vocabulary of the comments and generate structured data.

[0621] Step 2:

[0622] The server assigns a negativity score to the analyzed comment data using an evaluation method. The input data is the analyzed text, and the output data is the negativity score for each comment. This process is performed by calculating the negativity score from past data using a learning model.

[0623] Step 3:

[0624] The system acquires the user's emotional state in real time using voice and video sensors. The input is the user's voice and video data, and the output is the analyzed emotional state data. The server analyzes this data using emotion analysis tools and uses the results as filtering criteria for the next step.

[0625] Step 4:

[0626] The server sets filtering criteria based on the user's emotional state data and negativity score. The input data consists of the user's emotional state and the negativity score of the comments, which are combined and processed to determine the filtering criteria. The output is the filtering criteria.

[0627] Step 5:

[0628] The server uses filtering mechanisms to filter comments. The input consists of filtering criteria and parsed comments, and the output is the filtered comments. The server decides whether or not to display the comments based on the criteria and selects the filtered comments.

[0629] Step 6:

[0630] The server sends filtered comments to the terminal via a communication method. The input is the filtered comments, and the output is the comments displayed on the terminal. The terminal displays the received comments to the user, providing a stress-free SNS experience.

[0631] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0632] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0633] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0634] [Fourth Embodiment]

[0635] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0636] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0637] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0638] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0639] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0640] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0641] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0642] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0643] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0644] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0645] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0646] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0647] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0648] This invention is a filtering system for efficiently managing comments received by users on social networking services. Specific embodiments of this system are described below.

[0649] First, the user connects to the system via a dedicated terminal application and sets filtering criteria for incoming comments. These criteria include a threshold for negativity and specific expressions to be filtered.

[0650] Next, the server receives comment data flowing continuously from the SNS platform based on those criteria and passes the data to a dedicated natural language processing engine. This engine analyzes the context of the comments, specifically checking the structure and content of the language and the meaning of the words used.

[0651] The server then assigns a negativity score to the analyzed comment data. Here, it uses historical training data and machine learning models to evaluate how offensive or offensive each comment is considered to be.

[0652] Subsequently, the server's filtering mechanism refers to the criteria set by the user and filters out comments with high scores. Care is taken during this process to ensure that constructive critical comments are not affected.

[0653] The filtered results are sent from the server to the terminal, which then displays only the approved comments to the user. If the user feels that the comments do not match their filtering criteria, they are given the opportunity to adjust their settings.

[0654] Finally, the feedback is sent back to the server, and the machine learning model is updated using that information. This improves the overall filtering accuracy of the system, bringing it closer to the environment the user desires.

[0655] As a concrete example, consider a well-known company account that receives many comments about a particular product. This company wants to filter out negative expressions such as "terrible" or "bad," while receiving constructive feedback such as "there is room for improvement." This system would allow the company to eliminate unnecessary negative comments and utilize helpful feedback.

[0656] The following describes the processing flow.

[0657] Step 1:

[0658] Users launch a dedicated application on their device and set their own filtering criteria. These criteria include thresholds for the degree of negativity they want to filter out and specific keywords.

[0659] Step 2:

[0660] The server receives comment data flowing in real time from the social networking platform. The received data includes comment text and metadata.

[0661] Step 3:

[0662] The server sends the received comment data to a natural language processing engine. This engine analyzes the language syntax, vocabulary, and context, and converts the comments into a structured data format.

[0663] Step 4:

[0664] The server uses a machine learning model to assign negativity scores to structured comment data. This model quantifies aggression and unpleasantness based on training data.

[0665] Step 5:

[0666] The server filters comments based on the user's filtering criteria, referring to their negativity score. Comments exceeding the set threshold are filtered out, and unnecessary ones are excluded.

[0667] Step 6:

[0668] The server sends the allowed comment data from the filtered comments to the terminal.

[0669] Step 7:

[0670] The device displays received comments on the user's screen. Here, filtered negative comments are not displayed, and only comments deemed constructive are visualized.

[0671] Step 8:

[0672] Users review the displayed comments and make adjustments to their filtering settings if they feel they need to. This feedback is sent to the server via their device.

[0673] Step 9:

[0674] The server uses user feedback data to update its machine learning model and improve filtering accuracy.

[0675] (Example 1)

[0676] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0677] With the advancement of modern information and communication technology, users routinely receive large amounts of comments and information from various platforms. However, this information may include negative content or information unnecessary for the user, potentially impairing their comfortable user experience. Therefore, there is a need for means to appropriately filter the information users receive and eliminate unnecessary information.

[0678] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0679] In this invention, the server includes an information analysis means for analyzing data received from a user, an evaluation means for assigning an evaluation score to the analyzed information, and a selection means for selecting information based on the evaluation score according to user settings. This makes it possible to efficiently manage the information received by the user and accurately acquire only the important information.

[0680] "Information analysis means" refers to technologies that include the stage of analyzing data received from users and understanding its content and context.

[0681] "Evaluation methods" refer to techniques for quantifying the quality and attributes of analyzed information and assigning scores as evaluation criteria.

[0682] "Selection method" refers to a technology that uses assigned scores to select useful information and eliminate unnecessary information based on criteria set in advance by the user.

[0683] "Communication means" refers to the technologies and processes used to transmit selected information to a user's display device or other device.

[0684] "Learning methods" refer to technologies that enable a system to self-improve based on user feedback and setting changes, thereby improving the accuracy of information processing in subsequent instances.

[0685] The system for implementing this invention utilizes multiple electronic devices to provide information management as requested by the user. First, the user accesses the system through a dedicated terminal application and sets filtering criteria. These criteria include a threshold for the acceptable degree of negativity and the specification of specific expressions. The terminal application allows for easy modification and application of the criteria via a user interface.

[0686] The server receives the user's settings and continuously receives comment data from social media and other platforms. The received data is parsed using a natural language processing engine. This engine includes techniques to understand the context of the comments and analyze the content and nuances of the language. After analysis, the server uses a machine learning model to assign an appropriate negativeness score to the analysis results. This enables information filtering and allows comments to be filtered according to the user's settings.

[0687] Filtered information is sent from the server to the user's device, which displays only the selected information. During this process, if the user is dissatisfied with the displayed information, they can adjust the settings through feedback. This feedback is sent back to the server, and the machine learning model is updated based on that information. This improves the filtering accuracy for subsequent uses, providing the user with a more favorable information environment.

[0688] For example, when a user receives comments on a sports-related platform, they might want to avoid harsh criticism such as "XXX is the worst." By using a prompt in the format of "Please determine whether this comment is constructive according to user standards," the AI ​​model can determine whether the comment is constructive criticism or not and perform appropriate filtering.

[0689] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0690] Step 1:

[0691] The user sets the filtering criteria on their device.

[0692] Users use a dedicated application to set filtering criteria for incoming information. In this process, users input a threshold for negativity and specific expressions they wish to filter. The entered settings data is sent to the server and stored as the basis for the filtering process.

[0693] Step 2:

[0694] The server receives the comment data.

[0695] The server continuously retrieves comment data from social media and other platforms. This reception involves collecting large amounts of comment data in real time via APIs. The received data serves as input for subsequent processing and is stored in temporary storage on the server.

[0696] Step 3:

[0697] The server analyzes the data using a natural language processing engine.

[0698] The server passes the received comment data to a natural language processing engine. Here, the linguistic structure, context, and word meanings of the data are analyzed. For example, negative expressions and contexts within the text are identified. The analysis results are tagged and output as the basis for calculating evaluation scores.

[0699] Step 4:

[0700] The server evaluates the degree of negativity and assigns a score.

[0701] Based on the analysis results, the server uses a machine learning model to assign a negativity score to each comment. Using an algorithm trained on past data, it quantifies how negative each comment is. This score is then output as the comment's evaluation value.

[0702] Step 5:

[0703] The server filters comments based on the specified criteria.

[0704] Based on the assigned score, the server filters comments against the criteria set by the user. Negative comments exceeding the set threshold are excluded, and only positive or constructive feedback is selected. The filtered results are compiled into output data for the user.

[0705] Step 6:

[0706] The filtering results are sent from the server to the terminal.

[0707] The server sends filtered comments to the user's terminal. During transmission, data is transferred securely and quickly using a communication protocol. The comments are then output to the terminal as they should be displayed and provided to the user.

[0708] Step 7:

[0709] Users review the results and provide feedback.

[0710] Users review the comments displayed on their devices and provide feedback if they are dissatisfied with the filtering accuracy or wish to request adjustments. This feedback information is then sent back to the server and used as input data to improve the filtering criteria and machine learning models.

[0711] Step 8:

[0712] The server updates the machine learning model based on the feedback.

[0713] Based on the feedback, the server updates its machine learning model to improve filtering accuracy. This enhances its ability to handle new data analysis and filtering criteria, enabling more efficient information selection.

[0714] (Application Example 1)

[0715] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0716] With the increasing number of comments on online platforms, it has become difficult for administrators to efficiently filter out inappropriate or pointless negative comments and appropriately identify helpful improvement suggestions. In this situation, there is a need to accurately evaluate the nature of comments and implement filtering that meets the user's expectations.

[0717] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0718] In this invention, the server includes a natural language processing means for analyzing communication data received from a user, an evaluation means for assigning an emotional score to the analyzed communication data, and a selection means for filtering communications based on the emotional score evaluation value according to user settings. This improves inefficient comment processing and enables communication that emphasizes useful feedback.

[0719] "Natural language processing means" refers to technologies that analyze communication data received from users, understand its context and vocabulary, and structure it.

[0720] The "evaluation method" involves quantifying the degree of emotion and expression in the analyzed communication data to determine the degree of negativity or positivity of the comments.

[0721] A "selection method" is a technology that filters communication data according to the emotional intensity evaluation value based on criteria set by the user, and selects comments that exceed a specific standard.

[0722] "Transmission means" refers to the function of sending filtered communication data to the user's terminal, and is responsible for ensuring that only information useful to the user is provided.

[0723] A "learning tool" is a system that continuously updates machine learning models and improves filtering accuracy based on changes in user selection settings and feedback.

[0724] "Receiving means" refers to technology that identifies communications pointing out new areas for improvement and makes it easy for users to receive and view them.

[0725] The system for implementing this invention consists of a server and a user terminal. The server analyzes, evaluates, and sorts communication data using a natural language processing engine and a machine learning model. Specifically, it uses natural language processing libraries such as "spaCy" to perform language analysis and sentiment evaluation. Machine learning frameworks such as "scikit-learn" and "TensorFlow" are used for scoring sentiment levels. Through these technologies, the server detects and sorts negative or positive comments.

[0726] Furthermore, the server is equipped with a means to update the learning model based on user feedback, thereby improving filtering accuracy. Specifically, comments that can be evaluated are identified based on user settings, and the selected comments are transmitted to the terminal. This makes it easy for users to access only the useful feedback from the comments they receive.

[0727] The user's device sends comment data to the server in real time via the SNS API. The server evaluates and filters the received comment data and returns the filtered results to the device. Users can further improve the accuracy of the filtering by adjusting their selection settings.

[0728] As a concrete example, a restaurant chain could use this system when launching a social media campaign for a new dish to filter out critical comments such as "it's not tasty" and receive specific improvement suggestions such as "it would be better if the spices were used sparingly." This would allow them to quickly provide services that meet customer needs. An example of a prompt to input into the generating AI model is: "Design a social media filtering app that efficiently manages comments received on a social media account. It should eliminate negative comments based on user settings and display only helpful feedback."

[0729] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0730] Step 1:

[0731] The user's terminal retrieves comment data through the API of a social networking service. The retrieved comment data is sent to the server in its original form. The input is comment data obtained from the SNS, and the output is data transfer to the server. Specifically, the system retrieves data using an HTTP request, analyzes the response, and extracts the necessary comments.

[0732] Step 2:

[0733] The server receives comment data sent from the user terminal. This data is input into a natural language processing engine for content analysis. The input is the comment data received from the user terminal, and the output is the analyzed comment data. Specifically, the "spaCy" library is used to analyze the grammatical structure and keywords of the comments.

[0734] Step 3:

[0735] The server assigns a sentiment score to each comment based on the analysis results. Machine learning tools such as "TensorFlow" and "scikit-learn" are used for this scoring. The input is the analyzed comment data, and the output is the comment data with the sentiment score assigned. Specifically, the scoring is performed using a model based on past training data.

[0736] Step 4:

[0737] The server filters comment data according to sentiment scores based on criteria pre-configured by the user. The input is scored comment data, and the output is filtered comments. Specifically, it excludes comments with scores above a set threshold and selects only permitted comments.

[0738] Step 5:

[0739] The filtered comment data is sent from the server to the user's terminal. The user's terminal receives this data and displays it on the screen. The input is filtered comment data, and the output is the display on the user's terminal. Specifically, the data is received in HTML or JSON format and rendered on the user interface.

[0740] Step 6:

[0741] Users review the display results of received comments and adjust the settings if they feel the filtering criteria are inappropriate. This feedback is sent to the server and used to update the model. The input is the user's feedback, and the output is the updated filtering criteria. Specifically, the criteria are adjusted from the settings screen and saved to the database.

[0742] Step 7:

[0743] The server updates its machine learning model based on user feedback, improving filtering accuracy. The input is the user's adjusted criteria and feedback, and the output is the updated model. Specifically, it uses the collected feedback data as new training data to retrain the generative AI model.

[0744] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0745] This invention is a filtering system for efficiently managing comments received by users when using social networking services, based on the user's emotional state. The system incorporates four main methods: natural language processing, scoring, filtering, and an emotion engine.

[0746] First, users can access the system through a dedicated terminal application and set filtering criteria that include their own emotional data. This setting utilizes real-time emotional state data obtained from the emotion engine. For example, users can set stricter filtering criteria when they are experiencing high stress levels.

[0747] Next, the server aggregates comment data received from the SNS platform in real time and analyzes it using a natural language processing engine. This analysis confirms the context and vocabulary of the comments and converts them into structured data.

[0748] The server assigns a negativity score to the analyzed comment data. This scoring process is performed by a machine learning model that accurately assesses the degree of negativity based on past data.

[0749] Subsequently, the emotion engine is utilized, and the server analyzes the user's emotional state and dynamically incorporates this data into the filtering criteria. This allows the filtering criteria to be automatically adjusted according to the user's current emotional state.

[0750] Filtered comments are sent from the server to the terminal, which then displays or hides the filtered content on the user's screen. The system prioritizes comments that are sensitive to the user's emotions. For example, when the user is emotionally positive, some critical comments are displayed; conversely, when the user is emotionally negative and easily stressed, potentially stressful comments are filtered more strictly.

[0751] In this way, users can comfortably use social media while taking their own emotional state into consideration. This invention protects users' mental health and ensures that only constructive opinions influence their lives.

[0752] The following describes the processing flow.

[0753] Step 1:

[0754] The user launches a dedicated application on their device and provides the system with their emotional state, either by inputting it or by providing emotional data obtained from sensors. This data is then used for comment filtering.

[0755] Step 2:

[0756] The server receives comment data flowing in from the social networking platform in real time. This data includes metadata such as the comments themselves and the person who posted them.

[0757] Step 3:

[0758] The server sends the received comment data to a natural language processing engine. This engine performs analysis and tokenization and part-of-speech tagging to understand the linguistic structure and content of the comments.

[0759] Step 4:

[0760] The server passes the analyzed comment data to a scoring system, where it assigns a score for negativity. This scoring is performed using a machine learning algorithm and is based on past training data.

[0761] Step 5:

[0762] The server uses an emotion engine to assess the user's emotional state and dynamically adjusts filtering criteria based on that data. If the user is in a negative emotional state, the filtering criteria are set to become stricter.

[0763] Step 6:

[0764] The server selects comments filtered based on scoring, removing those exceeding a threshold. Constructive criticism is retained according to user preferences.

[0765] Step 7:

[0766] The server sends the filtered comment data to the terminal. The terminal then selects only the comments to display on the user screen based on the filtered results.

[0767] Step 8:

[0768] Users can review the comments displayed on their device and provide feedback to change settings if they are dissatisfied with the filtering results. This feedback will be used to improve the filtering accuracy in the future.

[0769] Step 9:

[0770] The server collects user feedback and uses it to update machine learning models and sentiment engines, thereby improving the overall accuracy of the system.

[0771] (Example 2)

[0772] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0773] In today's social media environment, comments received by users can affect their mental health. For users who are particularly sensitive to negative comments, such comments can be a source of stress. The objective of this invention is to enable users to comfortably use social networks while protecting their mental health by filtering comments according to their emotional state.

[0774] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0775] In this invention, the server includes emotion analysis means for analyzing emotion data received from the user to determine the emotional state, criteria adjustment means for adjusting filtering criteria based on the determined emotional state, and natural language processing means for analyzing received comment data. This enables dynamic filtering according to the user's emotional state.

[0776] "Emotional analysis means" refers to a technological element that analyzes emotional data received from a user and determines the user's emotional state.

[0777] A "criteria adjustment means" is a technical element that dynamically adjusts the criteria for comment filtering based on the user's emotional state determined by the sentiment analysis means.

[0778] "Natural language processing means" refers to technical elements that analyze received comment data and extract context and emotional nuances.

[0779] A "scoring method" is a technical element that assigns a score of negativity to analyzed comment data and performs evaluation based on that score.

[0780] A "filtering mechanism" is a technical element that selects and determines whether or not to display comments based on adjusted filtering criteria.

[0781] "Transmission means" refers to the technical element that transfers filtered comments to the device used by the user.

[0782] This invention is a system that filters comments on social networks according to the user's emotional state, enabling the user to receive information comfortably.

[0783] The system first determines the user's emotional state using emotion analysis tools. Users can provide emotional data through a terminal application using biometric sensors and a user interface. This emotional data can be acquired, for example, from a heart rate monitor or facial recognition camera.

[0784] The sentiment analysis method utilizes machine learning algorithms and sentiment analysis libraries to analyze this data. The server uses the algorithm to classify the user's current emotional state into categories such as positive or negative.

[0785] Next, the criteria adjustment mechanism dynamically adjusts the filtering criteria based on the determined emotional state. This adjustment sets the strictness of filtering for negative comments to match the user's emotions.

[0786] The server analyzes comments collected from social media using natural language processing tools to detect the context and emotional nuances of the comments. This structures the comment data, and a negativeness score is assigned via a scoring mechanism. For example, a machine learning platform is used for evaluation.

[0787] The filtering mechanism selects which comments to display based on the negativity score and the user's emotional state. The filtered comments are then transmitted to the device via a transmission mechanism, where the device decides whether to display or hide them. This process ensures that the information the user receives is tailored to their emotional state at that time.

[0788] For example, when a user is relaxed, even slightly critical comments will be displayed without issue, but when they are under a lot of stress, comments that could cause stress will be strictly filtered out.

[0789] The prompts for the generating AI model are entered in the format of, for example, "Explain how to filter received comments when the user is in a stressful state." These prompts enable the system to implement specific filtering that takes the user's state into consideration.

[0790] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0791] Step 1:

[0792] Users input emotional data via their devices. Sensors such as heart rate monitors and facial recognition cameras are used for input. This allows the system to obtain the user's emotional state as numerical data. The server receives this data and stores it as initial data.

[0793] Step 2:

[0794] The server analyzes the received emotional data using emotion analysis tools. This analysis utilizes emotion analysis libraries and machine learning algorithms. Through this analysis, the emotional data is classified into categories such as "positive" and "negative." Based on these emotional states, the server sets filtering criteria for each user.

[0795] Step 3:

[0796] The server collects relevant comment data through the SNS API. The collected comments are stored in the database in their original form. This prepares the comment information for analysis.

[0797] Step 4:

[0798] The server uses natural language processing to analyze the collected comment data. This analysis extracts contextual and emotional nuances from vocabulary. The analysis results are output as structured data and passed on to the next scoring process.

[0799] Step 5:

[0800] The server assigns a negativity score to the analyzed comments using a scoring mechanism. Here, a predictive model based on historical data is applied using a generative AI model. The processed scoring data, with scores expressed in numerical format, is used as a filtering criterion.

[0801] Step 6:

[0802] The server applies filtering criteria based on the user's emotional state via a criteria adjustment mechanism. This dynamically filters the collected comments. Based on these criteria, comments with high negative scores are hidden, and only comments within an acceptable range are displayed.

[0803] Step 7:

[0804] The server delivers filtered comments to the user's terminal via a transmission method. The terminal receives these comments and displays them on the screen. For some hidden comments, the user is notified and given the option to display them if desired.

[0805] This series of steps allows users to view comments that take their current emotional state into consideration, enabling them to use social media in a way that respects their mental health.

[0806] (Application Example 2)

[0807] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0808] Traditional social networking services lack comment management that takes into account users' emotional states, potentially harming users' mental health. Furthermore, their filtering accuracy is fixed and inadequate, failing to adapt to real-time changes in users' emotions.

[0809] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0810] In this invention, the server includes an electronic computing means for analyzing user comment data, an evaluation means for assigning an evaluation to the analysis results, and an emotion analysis means for analyzing the user's emotional state and changing the filtering criteria. This enables comment management according to the user's emotional state, and allows for more flexible and improved filtering accuracy.

[0811] "Electronic computing means" refers to digital computing devices and their functions used to process and analyze information.

[0812] "Evaluation means" refers to a function or device that performs a process of assigning scores or ratings to data using specific criteria.

[0813] "Emotional analysis means" refers to a function or device that can analyze a user's emotional state in real time and dynamically change the criteria for data processing based on that analysis.

[0814] "Filtering means" refers to a function or device that selects information from data that meets specific criteria and displays or blocks it as needed.

[0815] "Communication means" refers to protocols, devices, or functions used to send and receive data.

[0816] The system implementing this invention provides a multi-stage process for managing comments in order to make the user's social networking service experience more comfortable. The system operates in a network environment that includes user terminals and servers.

[0817] The server receives comment data and uses electronic computing tools to analyze it. These tools employ natural language processing algorithms to analyze the context and vocabulary of the data. For example, technologies such as TensorFlow and NLTK can be used to effectively analyze text data.

[0818] Next, the server uses an evaluation tool to assign a negativity score to the analyzed comments. This scoring uses historical data and a learned model (e.g., a machine learning model) to assess the negativity of the comments.

[0819] The emotion analysis system analyzes the user's emotional state in real time. Using hardware such as voice sensors (e.g., Shure MOTIV) and video sensors (e.g., Intel RealSense camera), it detects the user's facial expressions and tone of voice to acquire their emotional state. This data is then incorporated as filtering criteria by a filtering system. This allows the server to filter comments based on criteria that reflect the user's real-time emotional state.

[0820] The filtered comments are ultimately sent to the device via communication and displayed to the user. Because this display takes the user's emotional state into consideration, it can provide a more stress-free social media experience.

[0821] For example, when a user is relaxed, many comments, including minor criticisms, will be displayed, but when they are tired, only positive comments will be prioritized and displayed.

[0822] An example of a prompt to input into the generating AI model is as follows: "We are developing a system that filters comments based on their emotional state. It will manage comments in real time based on user emotional data. Please provide helpful suggestions."

[0823] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0824] Step 1:

[0825] The server receives comment data from a social networking service. The input data consists of unparsed text comments, which are then sent to a natural language processing engine. The server uses this engine to analyze the context and vocabulary of the comments and generate structured data.

[0826] Step 2:

[0827] The server assigns a negativity score to the analyzed comment data using an evaluation method. The input data is the analyzed text, and the output data is the negativity score for each comment. This process is performed by calculating the negativity score from past data using a learning model.

[0828] Step 3:

[0829] The system acquires the user's emotional state in real time using voice and video sensors. The input is the user's voice and video data, and the output is the analyzed emotional state data. The server analyzes this data using emotion analysis tools and uses the results as filtering criteria for the next step.

[0830] Step 4:

[0831] The server sets filtering criteria based on the user's emotional state data and negativity score. The input data consists of the user's emotional state and the negativity score of the comments, which are combined and processed to determine the filtering criteria. The output is the filtering criteria.

[0832] Step 5:

[0833] The server uses filtering mechanisms to filter comments. The input consists of filtering criteria and parsed comments, and the output is the filtered comments. The server decides whether or not to display the comments based on the criteria and selects the filtered comments.

[0834] Step 6:

[0835] The server sends filtered comments to the terminal via a communication method. The input is the filtered comments, and the output is the comments displayed on the terminal. The terminal displays the received comments to the user, providing a stress-free SNS experience.

[0836] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0837] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0838] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0839] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0840] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0841] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0842] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0843] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0844] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0845] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0846] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0847] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0848] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0849] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0850] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0851] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0852] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0853] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0854] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0855] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0856] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0857] The following is further disclosed regarding the embodiments described above.

[0858] (Claim 1)

[0859] A natural language processing method for analyzing comment data received from users,

[0860] A scoring method that assigns a negativity score to the analyzed comment data,

[0861] A filtering method that filters comments based on a negativeness score based on user settings,

[0862] A transmission means for sending filtered comments to the user's terminal,

[0863] A system that includes this.

[0864] (Claim 2)

[0865] The system according to claim 1, which uses a negativity score assigned by a scoring means to hide comments that exceed a threshold acceptable to the user.

[0866] (Claim 3)

[0867] The system according to claim 1, which improves filtering accuracy by updating a machine learning model based on user adjustments to filtering settings.

[0868] "Example 1"

[0869] (Claim 1)

[0870] Information analysis means for analyzing data received from users,

[0871] A rating system that assigns evaluation scores to the analyzed information,

[0872] A selection method that sorts information based on evaluation scores derived from user settings,

[0873] A communication means for transmitting the selected information to a user display device,

[0874] A learning method for updating machine learning models based on user information,

[0875] A system that includes this.

[0876] (Claim 2)

[0877] The system according to claim 1, which uses an evaluation score assigned by an evaluation means to hide information that exceeds a standard acceptable to the user.

[0878] (Claim 3)

[0879] The system according to claim 1, which improves sorting accuracy by updating the machine learning model based on adjustments made by the user to the sorting settings.

[0880] "Application Example 1"

[0881] (Claim 1)

[0882] A natural language processing method for analyzing communication data received from a user,

[0883] An evaluation method for assigning an emotional score to the analyzed communication data,

[0884] A selection means that filters communications based on an emotional rating value derived from user settings,

[0885] A means of providing filtered communications to the user terminal,

[0886] A learning method that improves filtering accuracy by updating the machine learning model based on adjustments to user selection settings,

[0887] A receiving means that can identify and display communications that point out new areas for improvement,

[0888] A system that includes this.

[0889] (Claim 2)

[0890] The system according to claim 1, which uses an evaluation value of the emotional level assigned by an evaluation means to hide communications that exceed a threshold value acceptable to the user.

[0891] (Claim 3)

[0892] The system according to claim 1, which is capable of receiving and displaying new improvement suggestions based on criteria customized by the user via a terminal device.

[0893] "Example 2 of combining an emotion engine"

[0894] (Claim 1)

[0895] A sentiment analysis method that analyzes emotional data received from a user to determine their emotional state,

[0896] A criteria adjustment means that adjusts the filtering criteria based on the determined emotional state,

[0897] A natural language processing method for analyzing received comment data,

[0898] A scoring method that assigns a negativity score to the analyzed comment data,

[0899] A filtering means that filters comments based on adjusted filtering criteria,

[0900] A transmission means for sending filtered comments to a user device,

[0901] A system that includes this.

[0902] (Claim 2)

[0903] The system according to claim 1, which dynamically changes the filtering criteria according to the user's emotional state.

[0904] (Claim 3)

[0905] The system according to claim 1, which improves filtering accuracy by providing feedback on adjusting filtering settings based on the user's emotional state and updating the machine learning model.

[0906] "Application example 2 when combining with an emotional engine"

[0907] (Claim 1)

[0908] An electronic computing means for analyzing comment data received from users,

[0909] An evaluation method for assigning a negativity score to the analyzed comment data,

[0910] A sentiment analysis method that analyzes the user's emotional state in real time and dynamically adjusts filtering criteria using that data,

[0911] A filtering means that filters comments based on a score of negativity, based on user settings and sentiment analysis means,

[0912] A communication means for sending filtered comments to an output device,

[0913] A system that includes this.

[0914] (Claim 2)

[0915] The system according to claim 1, which uses a negativity score assigned by an evaluation means to hide comments that exceed a threshold acceptable to the user.

[0916] (Claim 3)

[0917] The system according to claim 1, which improves filtering accuracy by updating the learning model based on user adjustments to filtering settings. [Explanation of Symbols]

[0918] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A natural language processing method for analyzing communication data received from a user, An evaluation method for assigning an emotional score to the analyzed communication data, A selection means that filters communications based on an emotional rating value derived from user settings, A means of providing filtered communications to the user terminal, A learning method that improves filtering accuracy by updating the machine learning model based on adjustments to user selection settings, A receiving means capable of identifying and displaying communications that point out new areas for improvement, A system that includes this.

2. The system according to claim 1, which uses an evaluation value of the emotional level assigned by an evaluation means to hide communications that exceed a threshold value acceptable to the user.

3. The system according to claim 1, which is capable of receiving and displaying new improvement suggestions based on criteria customized by the user via a terminal device.

Images