system

Abstract

The system according to this embodiment aims to visualize only highly reliable reviews. [Solution] The system according to the embodiment comprises a comparison unit, a score calculation unit, and a filtering unit. The comparison unit automatically compares review comments. The score calculation unit calculates a credibility score based on the reviews evaluated by the comparison unit. The filtering unit filters the reviews based on the credibility score calculated by the score calculation unit.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, there are many low-reliability reviews and spams, which may mislead the user's judgment.

[0005] The system according to the embodiment aims to visualize only highly reliable reviews.

Means for Solving the Problems

[0006] The system according to the embodiment includes a comparison unit, a score calculation unit, and a filtering unit. The comparison unit automatically compares review comments. The score calculation unit calculates a credibility score based on the reviews evaluated by the comparison unit. The filtering unit filters the reviews based on the credibility score calculated by the score calculation unit. [Effects of the Invention]

[0007] The system according to this embodiment can visualize only highly reliable reviews. [Brief explanation of the drawing]

[0008] [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. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

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

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0013] 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.

[0014] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna. 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).

[0015] 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 only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

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

[0017] As shown in FIG. 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.

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

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

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice 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 unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (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.

[0022] 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.

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

[0024] 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.

[0025] 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. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

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

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

[0028] (Example of form 1) An AI agent according to an embodiment of the present invention is a system that eliminates unreliable reviews, spam, and misinformation, and visualizes high-quality reviews. This system automatically compares reviews with other review comments and evaluates them through sentiment analysis and duplicate checks. Next, it calculates a credibility score based on the reviewer's history and the characteristics of the text. Finally, it introduces a filtering system that visualizes only reviews with high credibility. For example, the AI ​​agent automatically compares reviews with other review comments. In this process, it uses natural language processing technology to evaluate the consistency, specificity, and emotional bias of the review content. For example, it detects reviews with the same content or those that are nearly identical and gives them a low rating. It also gives a low rating to reviews that contain extreme emotional expressions. This makes it possible to eliminate unreliable reviews. Next, the AI ​​agent calculates a credibility score based on the reviewer's history and the characteristics of the text. Specifically, it evaluates the quality, consistency, and specificity of the reviewer's past posts and calculates a credibility score in real time. For example, it gives a high rating to reviews with specific examples and abundant information, and a low rating to reviews that contain extreme expressions or duplicate content. This makes it possible to identify highly reliable reviews. Furthermore, the AI ​​agent implements a filtering system that visualizes only highly credible reviews. Specifically, it displays only reviews with a credibility score exceeding a certain standard, restricting the display of less reliable reviews. This allows users to make decisions based on high-quality information. This improves the reliability of reviews and eliminates misinformation. It also makes it easier to evaluate the reliability of reviewers and ensures fairness. For example, spam and misinformation are eliminated, making it easier for users to find truly useful information. In addition, extreme emotional expressions are eliminated, making it easier to grasp the overall picture. This AI agent utilizes natural language processing and machine learning technologies, and continuously learns and improves to respond to review trends and new spam patterns. This improves the accuracy of the system and maintains a reliable review environment. In this way, the AI ​​agent can improve the reliability of reviews, eliminate misinformation, and ensure fairness.

[0029] The AI ​​agent according to this embodiment comprises a comparison unit, a score calculation unit, and a filtering unit. The comparison unit automatically compares review comments. The comparison unit evaluates the consistency, specificity, and emotional bias of the review content, for example, using natural language processing technology. For example, the comparison unit detects reviews with the same content or those that are copies of almost the same content and gives them a low rating. The comparison unit also gives a low rating to reviews that contain extreme emotional expressions. The score calculation unit calculates a credibility score based on the reviews evaluated by the comparison unit. The score calculation unit calculates the credibility score in real time, for example, based on the reviewer's history and the characteristics of the text. For example, the score calculation unit gives a high rating to reviews with specific examples and abundant information, and a low rating to reviews that contain extreme expressions or repetitive content. The filtering unit filters the reviews based on the credibility score calculated by the score calculation unit. For example, the filtering unit displays only reviews whose credibility score exceeds a certain standard and restricts the display of reviews with low credibility. As a result, the AI ​​agent according to this embodiment can improve the reliability of reviews, eliminate misinformation, and ensure fairness.

[0030] The comparison unit automatically compares review comments. For example, it uses natural language processing techniques to evaluate the consistency, specificity, and emotional bias of the review content. Specifically, it employs natural language processing techniques such as tokenization, morphological analysis, grammatical analysis, and semantic analysis. This allows it to accurately understand the context and meaning of the reviews and detect reviews with identical or nearly identical content. For example, the comparison unit analyzes the frequency of keywords and phrases in the review text and calculates a similarity score. A high similarity score indicates that the review is identical or similar and is given a low rating. The comparison unit also uses sentiment analysis techniques to evaluate the emotional expressions in the reviews. For example, reviews containing extreme emotional expressions (such as being extremely angry or extremely happy) are judged to have emotional bias and are given a low rating. Furthermore, to evaluate the specificity of the reviews, the comparison unit checks whether they contain specific examples and detailed explanations. Reviews containing specific examples and detailed explanations receive a high rating, while reviews with many abstract and ambiguous expressions receive a low rating. This allows the comparison unit to evaluate the content of the reviews from multiple perspectives and select highly reliable reviews.

[0031] The scoring unit calculates a credibility score based on the reviews evaluated by the comparison unit. The scoring unit calculates the credibility score in real time, for example, based on the reviewer's history and the characteristics of the review text. Specifically, it considers information such as the reviewer's past review history, the reviewer's rating, and the number and frequency of reviews the reviewer has posted. For example, reviews from reviewers who have posted many highly reliable reviews in the past will be given a high rating, while reviews from reviewers who have posted many low-rated reviews in the past will be given a low rating. The scoring unit also analyzes the characteristics of the review text and gives a high rating to reviews that contain specific examples and abundant information. For example, if the review text includes specific examples of product use and detailed explanations, that review will be given a high rating. On the other hand, reviews that contain extreme expressions or repetitive content will be given a low rating. For example, reviews that contain extreme expressions such as "This product is the best!" or "Absolutely do not buy this!" will be judged to have an emotional bias and will be given a low rating. Furthermore, the scoring unit also includes the length of the review and the accuracy of the grammar as evaluation criteria. Long, detailed reviews and grammatically accurate reviews receive high ratings, while short, vague reviews and reviews with many grammatical errors receive low ratings. This allows the scoring unit to comprehensively evaluate the reliability of the reviews and calculate a credibility score.

[0032] The filtering unit filters reviews based on the credibility score calculated by the scoring unit. For example, the filtering unit displays only reviews with a credibility score exceeding a certain threshold, restricting the display of less reliable reviews. Specifically, it prioritizes the display of reviews with high credibility scores, ensuring users access reliable information. For instance, it displays only reviews with a credibility score of 80 or higher, hiding those below that level. The filtering unit can also adjust display criteria according to user settings. For example, if a user selects "Show only reliable reviews," it displays only reviews with a credibility score of 90 or higher. Conversely, if the user selects "Show all reviews," it displays all reviews regardless of their credibility score. Furthermore, the filtering unit can filter reviews based on categories and tags. For example, it can filter reviews by category, such as "Reviews on product quality" or "Reviews on customer service," making it easier for users to access specific information. This allows the filtering unit to ensure users access reliable reviews and eliminate misinformation. The filtering unit also updates filtering results in real time, displaying reviews based on the latest information. This ensures that the filtering unit always provides the most up-to-date and reliable reviews, guaranteeing fairness.

[0033] The comparison unit can evaluate the consistency, specificity, and emotional bias of the review content using natural language processing techniques. For example, the comparison unit can evaluate the consistency of the review content using morphological analysis. For example, it can evaluate the coherence of context and the logical flow. The comparison unit can also evaluate the specificity of the review content using grammatical analysis. For example, it can evaluate the presence or absence of specific examples and detailed explanations. The comparison unit can also evaluate emotional bias using semantic analysis. For example, it can perform sentiment analysis and determine whether the sentiment is positive or negative. In this way, the comparison unit can eliminate unreliable reviews by evaluating the consistency, specificity, and emotional bias of the review content. Some or all of the above processing in the comparison unit may be performed using, for example, generative AI, or without generative AI. For example, the comparison unit can input the review content into a generative AI, which can then evaluate the consistency, specificity, and emotional bias of the review content.

[0034] The score calculation unit can calculate a credibility score in real time based on the reviewer's history and the characteristics of the text. For example, the score calculation unit calculates the credibility score by referring to the reviewer's past number of posts and evaluation history. For example, the score calculation unit will give a higher credibility score to reviewers with a large number of past posts. The score calculation unit can also give a higher credibility score to reviewers with a good evaluation history. Furthermore, the score calculation unit can calculate the credibility score using specific criteria and methods for evaluating the characteristics of the text. For example, the score calculation unit will evaluate the frequency of keyword occurrences and the structure of the text. In this way, the score calculation unit can identify highly reliable reviews by calculating a credibility score based on the reviewer's history and the characteristics of the text. Some or all of the above processing in the score calculation unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the score calculation unit can input the reviewer's history and the characteristics of the text into a generative AI, and the generative AI can calculate the credibility score.

[0035] The filtering unit can display only reviews whose credibility score exceeds a certain standard. For example, the filtering unit may prioritize displaying reviews with high credibility scores. For instance, it may display only reviews with a credibility score of 80 or higher. The filtering unit can also choose not to display reviews with low credibility scores. For example, it may not display reviews with a credibility score below 50. By displaying only reviews with a credibility score exceeding a certain standard, the filtering unit enables users to make decisions based on high-quality information. Some or all of the above processing in the filtering unit may be performed using, for example, a generative AI, or not. For example, the filtering unit may input credibility scores into a generative AI, which can then select the reviews to display.

[0036] The comparison unit can detect reviews with the same content or those that are copies of nearly identical content. For example, the comparison unit can detect reviews with the same content by calculating text similarity. For example, the comparison unit can calculate the similarity of reviews using cosine similarity. The comparison unit can also detect copies of nearly identical content using a duplicate detection algorithm. For example, the comparison unit can determine copied content using a partial match detection method. In this way, the comparison unit can eliminate duplicate reviews by detecting reviews with the same content or those that are copies of nearly identical content. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the review content into a generative AI, and the generative AI can calculate the similarity of the reviews.

[0037] The scoring unit can give high ratings to reviews with specific examples and abundant information, and low ratings to reviews containing extreme expressions or repetitive content. For example, the scoring unit evaluates the number and level of detail of specific examples. For example, the scoring unit gives high ratings to reviews with many specific examples. The scoring unit can also calculate a credibility score using specific criteria and methods for evaluating abundant information. For example, the scoring unit evaluates the comprehensiveness and reliability of the information. The scoring unit can also calculate a credibility score using specific criteria and methods for evaluating extreme expressions. For example, the scoring unit determines the presence or absence of emotional expressions and exaggerations. In this way, the scoring unit can identify highly reliable reviews by giving high ratings to reviews with specific examples and abundant information, and low ratings to reviews containing extreme expressions or repetitive content. Some or all of the above processing in the scoring unit may be performed using, for example, a generative AI, or without a generative AI. For example, the scoring unit can input the review content into a generative AI, which can evaluate specific examples and abundant information.

[0038] The comparison unit can improve the accuracy of its evaluation by referring to the reviewer's past posting history when assessing the consistency and specificity of reviews. For example, the comparison unit can analyze the quality of reviews previously posted by the reviewer and give higher ratings to consistent reviewers. For example, the comparison unit can give higher ratings to reviewers who tend to provide specific information based on the reviewer's past posting history. The comparison unit can also give lower ratings to unreliable reviewers based on the reviewer's past posting history. In this way, the comparison unit improves the accuracy of its review evaluation by referring to the reviewer's past posting history. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the comparison unit can input the reviewer's past posting history into a generative AI, which can then evaluate the consistency and specificity of the reviews.

[0039] The comparison unit can evaluate review content by considering the frequency of occurrence of specific keywords or phrases. For example, the comparison unit can detect frequently used positive keywords in a review and give them a higher rating. For example, the comparison unit can detect frequently used negative keywords in a review and give them a lower rating. The comparison unit can also add the reliability of a phrase to its evaluation criteria if that phrase is used repeatedly in the review. This improves the accuracy of the review evaluation by considering the frequency of occurrence of specific keywords or phrases. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the review content into a generative AI, which can then evaluate the frequency of occurrence of specific keywords or phrases.

[0040] The comparison unit can evaluate review content while considering the reviewer's geographical location. For example, the comparison unit will rate a review higher if the reviewer is close to the location being reviewed. For example, the comparison unit will rate a review lower if the reviewer is far from the location being reviewed. The comparison unit can also rate reviews containing region-specific information higher based on the reviewer's geographical location. As a result, by considering the reviewer's geographical location, the comparison unit can improve the rating of reviews containing region-specific information. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the reviewer's geographical location information into a generative AI, which can then evaluate the reviews.

[0041] The comparison unit can improve the accuracy of its evaluations by analyzing the reviewer's social media activity when evaluating review content. For example, if the reviewer's social media activity is active and highly reliable, the comparison unit will give a higher rating to that review. For example, if the reviewer's social media activity is low, the comparison unit will give a lower rating to that review. The comparison unit can also evaluate the reliability of a review based on the reviewer's number of followers and engagement rate on social media. In this way, the comparison unit improves the accuracy of its review evaluations by analyzing the reviewer's social media activity. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the reviewer's social media activity into a generative AI, and the generative AI can perform the review evaluation.

[0042] The score calculation unit can optimize the score by referring to the quality and consistency of the reviewer's past posts when calculating the credibility score. For example, the score calculation unit will give a higher credibility score to a reviewer if the quality of the reviews they have previously posted is high. For example, the score calculation unit will give a higher credibility score to a reviewer if the content of their past posts is consistent. The score calculation unit may also give a higher credibility score to a reviewer if their past posts contain specific and detailed information. In this way, the accuracy of the credibility score can be improved by the score calculation unit referring to the quality and consistency of the reviewer's past posts. Some or all of the above processing in the score calculation unit may be performed using, for example, a generating AI, or not using a generating AI. For example, the score calculation unit can input the reviewer's past posts into a generating AI, and the generating AI can calculate the credibility score.

[0043] The score calculation unit can adjust the score when calculating the credibility score by taking into account the reviewer's posting frequency and activity history. For example, if a reviewer posts frequently, the score calculation unit will give that reviewer a higher credibility score. For example, if a reviewer has a long activity history, the score calculation unit will give that reviewer a higher credibility score. Conversely, if a reviewer has a short activity history, the score calculation unit can also give that reviewer a lower credibility score. In this way, the accuracy of the credibility score can be improved by the score calculation unit taking into account the reviewer's posting frequency and activity history. Some or all of the above processing in the score calculation unit may be performed using, for example, a generating AI, or it may be performed without using a generating AI. For example, the score calculation unit can input the reviewer's posting frequency and activity history into a generating AI, and the generating AI can calculate the credibility score.

[0044] The score calculation unit can calculate the credibility score by considering the reviewer's geographical location information. For example, if the reviewer is close to the location being reviewed, the score calculation unit will give that reviewer a higher credibility score. For example, if the reviewer is far from the location being reviewed, the score calculation unit will give that reviewer a lower credibility score. The score calculation unit can also give a higher credibility score to reviews that include region-specific information, based on the reviewer's geographical location information. In this way, by considering the reviewer's geographical location information, the score calculation unit can improve the credibility score of reviews that include region-specific information. Some or all of the above processing in the score calculation unit may be performed using, for example, a generating AI, or without using a generating AI. For example, the score calculation unit can input the reviewer's geographical location information into a generating AI, and the generating AI can calculate the credibility score.

[0045] The score calculation unit can improve the accuracy of the score by analyzing the reviewer's social media activity when calculating the credibility score. For example, if a reviewer's social media activity is active and highly reliable, the score calculation unit will give that reviewer a higher credibility score. For example, if a reviewer's social media activity is low, the score calculation unit will give that reviewer a lower credibility score. The score calculation unit can also calculate the credibility score based on the number of followers and engagement rate of the reviewer on social media. In this way, the score calculation unit can improve the accuracy of the credibility score by analyzing the reviewer's social media activity. Some or all of the above processing in the score calculation unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the score calculation unit can input the reviewer's social media activity into a generative AI, and the generative AI can calculate the credibility score.

[0046] The filtering unit can update the displayed content in real time to reflect changes in the credibility score during filtering. For example, if the credibility score of a review changes, the filtering unit updates the displayed content in real time. For example, if the credibility score of a reviewer changes, the filtering unit updates the displayed content in real time. The filtering unit can also update the displayed content in real time if the evaluation criteria for a review changes. As a result, the filtering unit can display reviews based on the latest information by reflecting changes in the credibility score in real time. Some or all of the above processing in the filtering unit may be performed using a generation AI, for example, or without a generation AI. For example, the filtering unit can input data on changes in the credibility score into a generation AI, and the generation AI can update the displayed content.

[0047] The filtering unit can provide customization features to prioritize the display of posts from specific reviewers during filtering. For example, the filtering unit can prioritize the display of posts from highly reliable reviewers. For example, the filtering unit can prioritize the display of posts from reviewers that the user follows. The filtering unit can also prioritize the display of posts from reviewers that the user has previously given high ratings to. In this way, the filtering unit can prioritize the display of posts from specific reviewers, thereby allowing users to value the opinions of reviewers they trust. Some or all of the above processing in the filtering unit may be performed using, for example, a generating AI, or not using a generating AI. For example, the filtering unit can input post data from a specific reviewer into a generating AI, which can then select posts to display preferentially.

[0048] The filtering unit can adjust the displayed content when filtering, taking into account the reviewer's geographical location information. For example, if the reviewer is close to the location being reviewed, the filtering unit will prioritize displaying that review. For example, if the reviewer is far from the location being reviewed, the filtering unit will not display that review. The filtering unit can also prioritize displaying reviews that contain region-specific information based on the reviewer's geographical location information. As a result, by considering the reviewer's geographical location information, the filtering unit improves the display of reviews that contain region-specific information. Some or all of the above processing in the filtering unit may be performed using, for example, a generating AI, or without a generating AI. For example, the filtering unit can input the reviewer's geographical location information into the generating AI, and the generating AI can adjust the displayed content.

[0049] The filtering unit can improve the accuracy of displayed content by analyzing the reviewer's social media activity during filtering. For example, if a reviewer's social media activity is active and highly reliable, the filtering unit will prioritize displaying that review. For example, if a reviewer's social media activity is low, the filtering unit will not display that review. The filtering unit can also prioritize displaying highly reliable reviews based on the reviewer's number of followers and engagement rate on social media. In this way, the filtering unit improves the accuracy of review display by analyzing the reviewer's social media activity. Some or all of the above processing in the filtering unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the filtering unit can input the reviewer's social media activity into a generative AI, which can then adjust the displayed content.

[0050] The filtering unit can adjust the displayed content when filtering, taking into account the reviewer's posting frequency. For example, if a reviewer posts frequently, the filtering unit will prioritize displaying their reviews. For example, if a reviewer posts infrequently, the filtering unit will not display their reviews. The filtering unit can also prioritize displaying highly reliable reviews based on the reviewer's posting frequency. As a result, the filtering unit improves the display of highly reliable reviews by considering the reviewer's posting frequency. Some or all of the above processing in the filtering unit may be performed using, for example, a generating AI, or without using a generating AI. For example, the filtering unit can input reviewer posting frequency data into a generating AI, and the generating AI can adjust the displayed content.

[0051] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0052] The comparison section can consider the reviewer's expertise and qualifications when evaluating review content. For example, the comparison section will give a higher rating to a review if the reviewer has expertise or qualifications in a particular field. It can also give a higher rating to a review if the reviewer has experience in the relevant industry. This allows the comparison section to identify more reliable reviews by considering the reviewer's expertise and qualifications.

[0053] The scoring unit can incorporate the diversity of reviewers' posts as an evaluation criterion. For example, the scoring unit will give a higher credibility score to reviewers who post on a variety of topics. It can also give a higher credibility score to reviewers who offer different perspectives and opinions. This allows the scoring unit to evaluate the diversity of reviewers' posts, enabling a more balanced review evaluation.

[0054] The filtering function can adjust the display order of reviews by considering the user's browsing history. For example, the filtering function analyzes the trends of reviews the user has viewed in the past and prioritizes the display of similar reviews. It can also prioritize the display of posts from reviewers that the user has previously given high ratings to. In this way, the filtering function, by considering the user's browsing history, can display reviews that are optimal for the user.

[0055] The comparison section can evaluate review content by considering the reviewer's posting frequency and activity history. For example, the comparison section will give a higher rating to reviews from reviewers with a high posting frequency. It can also give a higher rating to reviews from reviewers with a long activity history. This allows the comparison section to identify highly reliable reviews by considering the reviewer's posting frequency and activity history.

[0056] The scoring unit can calculate a credibility score by analyzing a reviewer's social media activity. For example, if a reviewer's social media activity is active and highly reliable, the scoring unit will assign a higher credibility score to that reviewer. The scoring unit can also calculate a credibility score based on the reviewer's number of followers and engagement rate on social media. This allows the scoring unit to improve the accuracy of its credibility score by analyzing the reviewer's social media activity.

[0057] The following briefly describes the processing flow for example form 1.

[0058] Step 1: The comparison unit automatically compares review comments. The comparison unit uses, for example, natural language processing technology to evaluate the consistency, specificity, and emotional bias of the review content. For example, the comparison unit detects reviews with the same content or nearly identical content and assigns a low rating to them. The comparison unit also assigns a low rating to reviews that contain extreme emotional expressions. Step 2: The score calculation unit calculates a credibility score based on the reviews evaluated by the comparison unit. The score calculation unit calculates the credibility score in real time, for example, based on the reviewer's history and the characteristics of the text. For example, the score calculation unit highly rates reviews with specific examples and abundant information, and lowers the rate of reviews that contain extreme expressions or repetitive content. Step 3: The filtering unit filters the reviews based on the credibility score calculated by the score calculation unit. The filtering unit, for example, displays only reviews whose credibility score exceeds a certain standard, and restricts the display of unreliable reviews. This allows the AI ​​agent according to the embodiment to improve the reliability of reviews, eliminate misinformation, and ensure fairness.

[0059] (Example of form 2) An AI agent according to an embodiment of the present invention is a system that eliminates unreliable reviews, spam, and misinformation, and visualizes high-quality reviews. This system automatically compares reviews with other review comments and evaluates them through sentiment analysis and duplicate checks. Next, it calculates a credibility score based on the reviewer's history and the characteristics of the text. Finally, it introduces a filtering system that visualizes only reviews with high credibility. For example, the AI ​​agent automatically compares reviews with other review comments. In this process, it uses natural language processing technology to evaluate the consistency, specificity, and emotional bias of the review content. For example, it detects reviews with the same content or those that are nearly identical and gives them a low rating. It also gives a low rating to reviews that contain extreme emotional expressions. This makes it possible to eliminate unreliable reviews. Next, the AI ​​agent calculates a credibility score based on the reviewer's history and the characteristics of the text. Specifically, it evaluates the quality, consistency, and specificity of the reviewer's past posts and calculates a credibility score in real time. For example, it gives a high rating to reviews with specific examples and abundant information, and a low rating to reviews that contain extreme expressions or duplicate content. This makes it possible to identify highly reliable reviews. Furthermore, the AI ​​agent implements a filtering system that visualizes only highly credible reviews. Specifically, it displays only reviews with a credibility score exceeding a certain standard, restricting the display of less reliable reviews. This allows users to make decisions based on high-quality information. This improves the reliability of reviews and eliminates misinformation. It also makes it easier to evaluate the reliability of reviewers and ensures fairness. For example, spam and misinformation are eliminated, making it easier for users to find truly useful information. In addition, extreme emotional expressions are eliminated, making it easier to grasp the overall picture. This AI agent utilizes natural language processing and machine learning technologies, and continuously learns and improves to respond to review trends and new spam patterns. This improves the accuracy of the system and maintains a reliable review environment. In this way, the AI ​​agent can improve the reliability of reviews, eliminate misinformation, and ensure fairness.

[0060] The AI ​​agent according to this embodiment comprises a comparison unit, a score calculation unit, and a filtering unit. The comparison unit automatically compares review comments. The comparison unit evaluates the consistency, specificity, and emotional bias of the review content, for example, using natural language processing technology. For example, the comparison unit detects reviews with the same content or those that are copies of almost the same content and gives them a low rating. The comparison unit also gives a low rating to reviews that contain extreme emotional expressions. The score calculation unit calculates a credibility score based on the reviews evaluated by the comparison unit. The score calculation unit calculates the credibility score in real time, for example, based on the reviewer's history and the characteristics of the text. For example, the score calculation unit gives a high rating to reviews with specific examples and abundant information, and a low rating to reviews that contain extreme expressions or repetitive content. The filtering unit filters the reviews based on the credibility score calculated by the score calculation unit. For example, the filtering unit displays only reviews whose credibility score exceeds a certain standard and restricts the display of reviews with low credibility. As a result, the AI ​​agent according to this embodiment can improve the reliability of reviews, eliminate misinformation, and ensure fairness.

[0061] The comparison unit automatically compares review comments. For example, it uses natural language processing techniques to evaluate the consistency, specificity, and emotional bias of the review content. Specifically, it employs natural language processing techniques such as tokenization, morphological analysis, grammatical analysis, and semantic analysis. This allows it to accurately understand the context and meaning of the reviews and detect reviews with identical or nearly identical content. For example, the comparison unit analyzes the frequency of keywords and phrases in the review text and calculates a similarity score. A high similarity score indicates that the review is identical or similar and is given a low rating. The comparison unit also uses sentiment analysis techniques to evaluate the emotional expressions in the reviews. For example, reviews containing extreme emotional expressions (such as being extremely angry or extremely happy) are judged to have emotional bias and are given a low rating. Furthermore, to evaluate the specificity of the reviews, the comparison unit checks whether they contain specific examples and detailed explanations. Reviews containing specific examples and detailed explanations receive a high rating, while reviews with many abstract and ambiguous expressions receive a low rating. This allows the comparison unit to evaluate the content of the reviews from multiple perspectives and select highly reliable reviews.

[0062] The scoring unit calculates a credibility score based on the reviews evaluated by the comparison unit. The scoring unit calculates the credibility score in real time, for example, based on the reviewer's history and the characteristics of the review text. Specifically, it considers information such as the reviewer's past review history, the reviewer's rating, and the number and frequency of reviews the reviewer has posted. For example, reviews from reviewers who have posted many highly reliable reviews in the past will be given a high rating, while reviews from reviewers who have posted many low-rated reviews in the past will be given a low rating. The scoring unit also analyzes the characteristics of the review text and gives a high rating to reviews that contain specific examples and abundant information. For example, if the review text includes specific examples of product use and detailed explanations, that review will be given a high rating. On the other hand, reviews that contain extreme expressions or repetitive content will be given a low rating. For example, reviews that contain extreme expressions such as "This product is the best!" or "Absolutely do not buy this!" will be judged to have an emotional bias and will be given a low rating. Furthermore, the scoring unit also includes the length of the review and the accuracy of the grammar as evaluation criteria. Long, detailed reviews and grammatically accurate reviews receive high ratings, while short, vague reviews and reviews with many grammatical errors receive low ratings. This allows the scoring unit to comprehensively evaluate the reliability of the reviews and calculate a credibility score.

[0063] The filtering unit filters reviews based on the credibility score calculated by the scoring unit. For example, the filtering unit displays only reviews with a credibility score exceeding a certain threshold, restricting the display of less reliable reviews. Specifically, it prioritizes the display of reviews with high credibility scores, ensuring users access reliable information. For instance, it displays only reviews with a credibility score of 80 or higher, hiding those below that level. The filtering unit can also adjust display criteria according to user settings. For example, if a user selects "Show only reliable reviews," it displays only reviews with a credibility score of 90 or higher. Conversely, if the user selects "Show all reviews," it displays all reviews regardless of their credibility score. Furthermore, the filtering unit can filter reviews based on categories and tags. For example, it can filter reviews by category, such as "Reviews on product quality" or "Reviews on customer service," making it easier for users to access specific information. This allows the filtering unit to ensure users access reliable reviews and eliminate misinformation. The filtering unit also updates filtering results in real time, displaying reviews based on the latest information. This ensures that the filtering unit always provides the most up-to-date and reliable reviews, guaranteeing fairness.

[0064] The comparison unit can evaluate the consistency, specificity, and emotional bias of the review content using natural language processing techniques. For example, the comparison unit can evaluate the consistency of the review content using morphological analysis. For example, it can evaluate the coherence of context and the logical flow. The comparison unit can also evaluate the specificity of the review content using grammatical analysis. For example, it can evaluate the presence or absence of specific examples and detailed explanations. The comparison unit can also evaluate emotional bias using semantic analysis. For example, it can perform sentiment analysis and determine whether the sentiment is positive or negative. In this way, the comparison unit can eliminate unreliable reviews by evaluating the consistency, specificity, and emotional bias of the review content. Some or all of the above processing in the comparison unit may be performed using, for example, generative AI, or without generative AI. For example, the comparison unit can input the review content into a generative AI, which can then evaluate the consistency, specificity, and emotional bias of the review content.

[0065] The score calculation unit can calculate a credibility score in real time based on the reviewer's history and the characteristics of the text. For example, the score calculation unit calculates the credibility score by referring to the reviewer's past number of posts and evaluation history. For example, the score calculation unit will give a higher credibility score to reviewers with a large number of past posts. The score calculation unit can also give a higher credibility score to reviewers with a good evaluation history. Furthermore, the score calculation unit can calculate the credibility score using specific criteria and methods for evaluating the characteristics of the text. For example, the score calculation unit will evaluate the frequency of keyword occurrences and the structure of the text. In this way, the score calculation unit can identify highly reliable reviews by calculating a credibility score based on the reviewer's history and the characteristics of the text. Some or all of the above processing in the score calculation unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the score calculation unit can input the reviewer's history and the characteristics of the text into a generative AI, and the generative AI can calculate the credibility score.

[0066] The filtering unit can display only reviews whose credibility score exceeds a certain standard. For example, the filtering unit may prioritize displaying reviews with high credibility scores. For instance, it may display only reviews with a credibility score of 80 or higher. The filtering unit can also choose not to display reviews with low credibility scores. For example, it may not display reviews with a credibility score below 50. By displaying only reviews with a credibility score exceeding a certain standard, the filtering unit enables users to make decisions based on high-quality information. Some or all of the above processing in the filtering unit may be performed using, for example, a generative AI, or not. For example, the filtering unit may input credibility scores into a generative AI, which can then select the reviews to display.

[0067] The comparison unit can detect reviews with the same content or those that are copies of nearly identical content. For example, the comparison unit can detect reviews with the same content by calculating text similarity. For example, the comparison unit can calculate the similarity of reviews using cosine similarity. The comparison unit can also detect copies of nearly identical content using a duplicate detection algorithm. For example, the comparison unit can determine copied content using a partial match detection method. In this way, the comparison unit can eliminate duplicate reviews by detecting reviews with the same content or those that are copies of nearly identical content. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the review content into a generative AI, and the generative AI can calculate the similarity of the reviews.

[0068] The scoring unit can give high ratings to reviews with specific examples and abundant information, and low ratings to reviews containing extreme expressions or repetitive content. For example, the scoring unit evaluates the number and level of detail of specific examples. For example, the scoring unit gives high ratings to reviews with many specific examples. The scoring unit can also calculate a credibility score using specific criteria and methods for evaluating abundant information. For example, the scoring unit evaluates the comprehensiveness and reliability of the information. The scoring unit can also calculate a credibility score using specific criteria and methods for evaluating extreme expressions. For example, the scoring unit determines the presence or absence of emotional expressions and exaggerations. In this way, the scoring unit can identify highly reliable reviews by giving high ratings to reviews with specific examples and abundant information, and low ratings to reviews containing extreme expressions or repetitive content. Some or all of the above processing in the scoring unit may be performed using, for example, a generative AI, or without a generative AI. For example, the scoring unit can input the review content into a generative AI, which can evaluate specific examples and abundant information.

[0069] The comparison unit can estimate the user's emotions and adjust the evaluation criteria for the review content based on the estimated user emotions. For example, the comparison unit estimates the user's emotions using an emotion analysis algorithm. For instance, the comparison unit classifies the user's emotions as positive, negative, or neutral. Furthermore, the comparison unit adjusts the evaluation criteria for the review content using specific methods and criteria based on the estimated user emotions. For example, if the user has positive emotions, the unit emphasizes the specificity of the review to mitigate emotional bias. If the user has negative emotions, the unit can exclude emotional expressions from the evaluation criteria and emphasize factual content. Additionally, if the user has neutral emotions, the unit can evaluate the consistency and specificity of the review in a balanced way. This allows the comparison unit to provide more appropriate review evaluations by adjusting the evaluation criteria based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the comparison unit may be performed using, for example, a generative AI, or without using a generative AI. For example, the comparison unit can input user emotion data into a generative AI, which can then estimate the emotion.

[0070] The comparison unit can improve the accuracy of its evaluation by referring to the reviewer's past posting history when assessing the consistency and specificity of reviews. For example, the comparison unit can analyze the quality of reviews previously posted by the reviewer and give higher ratings to consistent reviewers. For example, the comparison unit can give higher ratings to reviewers who tend to provide specific information based on the reviewer's past posting history. The comparison unit can also give lower ratings to unreliable reviewers based on the reviewer's past posting history. In this way, the comparison unit improves the accuracy of its review evaluation by referring to the reviewer's past posting history. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the comparison unit can input the reviewer's past posting history into a generative AI, which can then evaluate the consistency and specificity of the reviews.

[0071] The comparison unit can evaluate review content by considering the frequency of occurrence of specific keywords or phrases. For example, the comparison unit can detect frequently used positive keywords in a review and give them a higher rating. For example, the comparison unit can detect frequently used negative keywords in a review and give them a lower rating. The comparison unit can also add the reliability of a phrase to its evaluation criteria if that phrase is used repeatedly in the review. This improves the accuracy of the review evaluation by considering the frequency of occurrence of specific keywords or phrases. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the review content into a generative AI, which can then evaluate the frequency of occurrence of specific keywords or phrases.

[0072] The comparison unit can estimate the user's emotions and adjust the order in which review ratings are displayed based on the estimated emotions. For example, the comparison unit might use an emotion analysis algorithm to estimate the user's emotions. For instance, it might classify the user's emotions as positive, negative, or neutral. Furthermore, the comparison unit adjusts the order in which review ratings are displayed using specific methods and criteria based on the estimated emotions. For example, if a user has positive emotions, positive reviews will be displayed preferentially. If a user has negative emotions, negative reviews will be displayed preferentially. Additionally, if a user has neutral emotions, highly-rated reviews will be displayed in a balanced manner. This allows the comparison unit to display reviews optimally for the user by adjusting the order in which review ratings are displayed based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the comparison unit may be performed using, for example, a generative AI, or without using a generative AI. For example, the comparison unit can input user emotion data into a generative AI, which can then estimate the emotion.

[0073] The comparison unit can evaluate review content while considering the reviewer's geographical location. For example, the comparison unit will rate a review higher if the reviewer is close to the location being reviewed. For example, the comparison unit will rate a review lower if the reviewer is far from the location being reviewed. The comparison unit can also rate reviews containing region-specific information higher based on the reviewer's geographical location. As a result, by considering the reviewer's geographical location, the comparison unit can improve the rating of reviews containing region-specific information. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the reviewer's geographical location information into a generative AI, which can then evaluate the reviews.

[0074] The comparison unit can improve the accuracy of its evaluations by analyzing the reviewer's social media activity when evaluating review content. For example, if the reviewer's social media activity is active and highly reliable, the comparison unit will give a higher rating to that review. For example, if the reviewer's social media activity is low, the comparison unit will give a lower rating to that review. The comparison unit can also evaluate the reliability of a review based on the reviewer's number of followers and engagement rate on social media. In this way, the comparison unit improves the accuracy of its review evaluations by analyzing the reviewer's social media activity. Some or all of the above processing in the comparison unit may be performed using, for example, a generative AI, or without a generative AI. For example, the comparison unit can input the reviewer's social media activity into a generative AI, and the generative AI can perform the review evaluation.

[0075] The scoring unit can estimate the user's emotions and adjust the method for calculating the credit score based on the estimated user emotions. For example, the scoring unit estimates the user's emotions using an emotion analysis algorithm. For instance, the scoring unit classifies the user's emotions as positive, negative, or neutral. Furthermore, the scoring unit adjusts the method for calculating the credit score using specific methods and criteria based on the estimated user emotions. For example, if a user has positive emotions, positive reviews will receive a higher credit score. Similarly, if a user has negative emotions, negative reviews will receive a higher credit score. Additionally, if a user has neutral emotions, the credit score can be calculated with an emphasis on the consistency and specificity of the review. This allows the scoring unit to calculate a more appropriate credit score by adjusting the method based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the score calculation unit may be performed using, for example, a generative AI, or without using a generative AI. For example, the score calculation unit can input user emotion data into a generative AI, which can then estimate the emotion.

[0076] The score calculation unit can optimize the score by referring to the quality and consistency of the reviewer's past posts when calculating the credibility score. For example, the score calculation unit will give a higher credibility score to a reviewer if the quality of the reviews they have previously posted is high. For example, the score calculation unit will give a higher credibility score to a reviewer if the content of their past posts is consistent. The score calculation unit may also give a higher credibility score to a reviewer if their past posts contain specific and detailed information. In this way, the accuracy of the credibility score can be improved by the score calculation unit referring to the quality and consistency of the reviewer's past posts. Some or all of the above processing in the score calculation unit may be performed using, for example, a generating AI, or not using a generating AI. For example, the score calculation unit can input the reviewer's past posts into a generating AI, and the generating AI can calculate the credibility score.

[0077] The score calculation unit can adjust the score when calculating the credibility score by taking into account the reviewer's posting frequency and activity history. For example, if a reviewer posts frequently, the score calculation unit will give that reviewer a higher credibility score. For example, if a reviewer has a long activity history, the score calculation unit will give that reviewer a higher credibility score. Conversely, if a reviewer has a short activity history, the score calculation unit can also give that reviewer a lower credibility score. In this way, the accuracy of the credibility score can be improved by the score calculation unit taking into account the reviewer's posting frequency and activity history. Some or all of the above processing in the score calculation unit may be performed using, for example, a generating AI, or it may be performed without using a generating AI. For example, the score calculation unit can input the reviewer's posting frequency and activity history into a generating AI, and the generating AI can calculate the credibility score.

[0078] The scoring unit can estimate the user's emotions and adjust the display method of the credit score based on the estimated user emotions. For example, the scoring unit estimates the user's emotions using an emotion analysis algorithm. For instance, the scoring unit classifies the user's emotions as positive, negative, or neutral. Furthermore, the scoring unit adjusts the display method of the credit score using specific methods and criteria based on the estimated user emotions. For example, if a user has positive emotions, the credit scores of positive reviews can be displayed more prominently. Similarly, if a user has negative emotions, the credit scores of negative reviews can be displayed more prominently. Moreover, if a user has neutral emotions, the credit scores of all reviews can be displayed in a balanced manner. In this way, the scoring unit can optimize the display for the user by adjusting the display method of the credit score based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the score calculation unit may be performed using, for example, a generative AI, or without using a generative AI. For example, the score calculation unit can input user emotion data into a generative AI, which can then estimate the emotion.

[0079] The score calculation unit can calculate the credibility score by considering the reviewer's geographical location information. For example, if the reviewer is close to the location being reviewed, the score calculation unit will give that reviewer a higher credibility score. For example, if the reviewer is far from the location being reviewed, the score calculation unit will give that reviewer a lower credibility score. The score calculation unit can also give a higher credibility score to reviews that include region-specific information, based on the reviewer's geographical location information. In this way, by considering the reviewer's geographical location information, the score calculation unit can improve the credibility score of reviews that include region-specific information. Some or all of the above processing in the score calculation unit may be performed using, for example, a generating AI, or without using a generating AI. For example, the score calculation unit can input the reviewer's geographical location information into a generating AI, and the generating AI can calculate the credibility score.

[0080] The score calculation unit can improve the accuracy of the score by analyzing the reviewer's social media activity when calculating the credibility score. For example, if a reviewer's social media activity is active and highly reliable, the score calculation unit will give that reviewer a higher credibility score. For example, if a reviewer's social media activity is low, the score calculation unit will give that reviewer a lower credibility score. The score calculation unit can also calculate the credibility score based on the number of followers and engagement rate of the reviewer on social media. In this way, the score calculation unit can improve the accuracy of the credibility score by analyzing the reviewer's social media activity. Some or all of the above processing in the score calculation unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the score calculation unit can input the reviewer's social media activity into a generative AI, and the generative AI can calculate the credibility score.

[0081] The filtering unit can estimate the user's emotions and adjust the filtering criteria based on the estimated emotions. For example, the filtering unit might use an emotion analysis algorithm to estimate the user's emotions. For instance, it might classify the user's emotions as positive, negative, or neutral. The filtering unit also adjusts the filtering criteria using specific methods and criteria based on the estimated emotions. For example, if a user has positive emotions, positive reviews are prioritized. If a user has negative emotions, negative reviews are prioritized. Furthermore, if a user has neutral emotions, highly-rated reviews are displayed in a balanced manner. This allows the filtering unit to display reviews optimally for the user by adjusting the filtering criteria based on their emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the filtering unit may be performed using, for example, generative AI, or without using generative AI. For example, the filtering unit can input user emotion data into a generating AI, which can then estimate the emotion.

[0082] The filtering unit can update the displayed content in real time to reflect changes in the credibility score during filtering. For example, if the credibility score of a review changes, the filtering unit updates the displayed content in real time. For example, if the credibility score of a reviewer changes, the filtering unit updates the displayed content in real time. The filtering unit can also update the displayed content in real time if the evaluation criteria for a review changes. As a result, the filtering unit can display reviews based on the latest information by reflecting changes in the credibility score in real time. Some or all of the above processing in the filtering unit may be performed using a generation AI, for example, or without a generation AI. For example, the filtering unit can input data on changes in the credibility score into a generation AI, and the generation AI can update the displayed content.

[0083] The filtering unit can provide customization features to prioritize the display of posts from specific reviewers during filtering. For example, the filtering unit can prioritize the display of posts from highly reliable reviewers. For example, the filtering unit can prioritize the display of posts from reviewers that the user follows. The filtering unit can also prioritize the display of posts from reviewers that the user has previously given high ratings to. In this way, the filtering unit can prioritize the display of posts from specific reviewers, thereby allowing users to value the opinions of reviewers they trust. Some or all of the above processing in the filtering unit may be performed using, for example, a generating AI, or not using a generating AI. For example, the filtering unit can input post data from a specific reviewer into a generating AI, which can then select posts to display preferentially.

[0084] The filtering unit can estimate the user's emotions and adjust the display method of the filtering results based on the estimated user emotions. For example, the filtering unit estimates the user's emotions using an emotion analysis algorithm. For example, the filtering unit classifies the user's emotions as positive, negative, or neutral. The filtering unit also adjusts the display method of the filtering results using specific methods and criteria that adjust the display method of the filtering results based on the estimated user emotions. For example, if the user has positive emotions, positive reviews can be displayed prominently. If the user has negative emotions, negative reviews can be displayed prominently. Furthermore, if the user has neutral emotions, all reviews can be displayed in a balanced manner. In this way, the filtering unit can display reviews optimally for the user by adjusting the display method of the filtering results based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above processing in the filtering unit may be performed using a generative AI, for example, or without a generative AI. For example, the filtering unit can input user emotion data into a generating AI, which can then estimate the emotion.

[0085] The filtering unit can adjust the displayed content when filtering, taking into account the reviewer's geographical location information. For example, if the reviewer is close to the location being reviewed, the filtering unit will prioritize displaying that review. For example, if the reviewer is far from the location being reviewed, the filtering unit will not display that review. The filtering unit can also prioritize displaying reviews that contain region-specific information based on the reviewer's geographical location information. As a result, by considering the reviewer's geographical location information, the filtering unit improves the display of reviews that contain region-specific information. Some or all of the above processing in the filtering unit may be performed using, for example, a generating AI, or without a generating AI. For example, the filtering unit can input the reviewer's geographical location information into the generating AI, and the generating AI can adjust the displayed content.

[0086] The filtering unit can improve the accuracy of displayed content by analyzing the reviewer's social media activity during filtering. For example, if a reviewer's social media activity is active and highly reliable, the filtering unit will prioritize displaying that review. For example, if a reviewer's social media activity is low, the filtering unit will not display that review. The filtering unit can also prioritize displaying highly reliable reviews based on the reviewer's number of followers and engagement rate on social media. In this way, the filtering unit improves the accuracy of review display by analyzing the reviewer's social media activity. Some or all of the above processing in the filtering unit may be performed using, for example, a generative AI, or not using a generative AI. For example, the filtering unit can input the reviewer's social media activity into a generative AI, which can then adjust the displayed content.

[0087] The filtering unit can adjust the displayed content when filtering, taking into account the reviewer's posting frequency. For example, if a reviewer posts frequently, the filtering unit will prioritize displaying their reviews. For example, if a reviewer posts infrequently, the filtering unit will not display their reviews. The filtering unit can also prioritize displaying highly reliable reviews based on the reviewer's posting frequency. As a result, the filtering unit improves the display of highly reliable reviews by considering the reviewer's posting frequency. Some or all of the above processing in the filtering unit may be performed using, for example, a generating AI, or without using a generating AI. For example, the filtering unit can input reviewer posting frequency data into a generating AI, and the generating AI can adjust the displayed content.

[0088] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0089] The comparison section can consider the reviewer's expertise and qualifications when evaluating review content. For example, the comparison section will give a higher rating to a review if the reviewer has expertise or qualifications in a particular field. It can also give a higher rating to a review if the reviewer has experience in the relevant industry. This allows the comparison section to identify more reliable reviews by considering the reviewer's expertise and qualifications.

[0090] The scoring unit can incorporate the diversity of reviewers' posts as an evaluation criterion. For example, the scoring unit will give a higher credibility score to reviewers who post on a variety of topics. It can also give a higher credibility score to reviewers who offer different perspectives and opinions. This allows the scoring unit to evaluate the diversity of reviewers' posts, enabling a more balanced review evaluation.

[0091] The filtering function can adjust the display order of reviews by considering the user's browsing history. For example, the filtering function analyzes the trends of reviews the user has viewed in the past and prioritizes the display of similar reviews. It can also prioritize the display of posts from reviewers that the user has previously given high ratings to. In this way, the filtering function, by considering the user's browsing history, can display reviews that are optimal for the user.

[0092] The comparison section can evaluate review content by considering the reviewer's posting frequency and activity history. For example, the comparison section will give a higher rating to reviews from reviewers with a high posting frequency. It can also give a higher rating to reviews from reviewers with a long activity history. This allows the comparison section to identify highly reliable reviews by considering the reviewer's posting frequency and activity history.

[0093] The scoring unit can calculate a credibility score by analyzing a reviewer's social media activity. For example, if a reviewer's social media activity is active and highly reliable, the scoring unit will assign a higher credibility score to that reviewer. The scoring unit can also calculate a credibility score based on the reviewer's number of followers and engagement rate on social media. This allows the scoring unit to improve the accuracy of its credibility score by analyzing the reviewer's social media activity.

[0094] The comparison unit can estimate the user's emotions and adjust the evaluation criteria for the review content based on those estimated emotions. For example, the comparison unit can classify the user's emotions as positive, negative, or neutral and adjust the evaluation criteria accordingly. If the user has positive emotions, the unit will emphasize the specificity of the review to mitigate emotional bias. If the user has negative emotions, the unit can exclude emotional expressions from the evaluation criteria and emphasize factual content. In this way, the comparison unit can adjust the evaluation criteria for the review content based on the user's emotions, enabling more appropriate review evaluations.

[0095] The score calculation unit can estimate the user's emotions and adjust the method for calculating the credit score based on those emotions. For example, the score calculation unit can classify the user's emotions as positive, negative, or neutral and adjust the method for calculating the credit score according to the emotion. If the user has positive emotions, the credit score of positive reviews will be increased. If the user has negative emotions, the credit score of negative reviews will be increased. In this way, the score calculation unit can adjust the method for calculating the credit score based on the user's emotions, enabling it to calculate a more appropriate credit score.

[0096] The filtering unit can estimate the user's emotions and adjust the filtering criteria based on those emotions. For example, the filtering unit can classify the user's emotions as positive, negative, or neutral and adjust the filtering criteria accordingly. If the user has positive emotions, positive reviews will be displayed preferentially. If the user has negative emotions, negative reviews will be displayed preferentially. In this way, the filtering unit can adjust the filtering criteria based on the user's emotions, enabling the display of reviews that are optimal for the user.

[0097] The comparison unit can estimate the user's emotions and adjust the order in which review ratings are displayed based on those estimated emotions. For example, the comparison unit can classify user emotions as positive, negative, or neutral and adjust the display order of ratings according to the emotion. If the user has positive emotions, positive reviews will be displayed preferentially. If the user has negative emotions, negative reviews will be displayed preferentially. In this way, the comparison unit can adjust the order in which review ratings are displayed based on the user's emotions, enabling the display of reviews that are optimal for the user.

[0098] The score calculation unit can estimate the user's emotions and adjust the display method of the credit score based on the estimated user emotions. For example, the score calculation unit can classify the user's emotions as positive, negative, or neutral and adjust the display method of the credit score according to the emotions. If the user has positive emotions, the credit score of positive reviews can be displayed more prominently. If the user has negative emotions, the credit score of negative reviews can be displayed more prominently. In this way, the score calculation unit can adjust the display method of the credit score based on the user's emotions, enabling the display to be optimal for the user.

[0099] The following briefly describes the processing flow for example form 2.

[0100] Step 1: The comparison unit automatically compares review comments. The comparison unit uses, for example, natural language processing technology to evaluate the consistency, specificity, and emotional bias of the review content. For example, the comparison unit detects reviews with the same content or nearly identical content and assigns a low rating to them. The comparison unit also assigns a low rating to reviews that contain extreme emotional expressions. Step 2: The score calculation unit calculates a credibility score based on the reviews evaluated by the comparison unit. The score calculation unit calculates the credibility score in real time, for example, based on the reviewer's history and the characteristics of the text. For example, the score calculation unit highly rates reviews with specific examples and abundant information, and lowers the rate of reviews that contain extreme expressions or repetitive content. Step 3: The filtering unit filters the reviews based on the credibility score calculated by the score calculation unit. The filtering unit, for example, displays only reviews whose credibility score exceeds a certain standard, and restricts the display of unreliable reviews. This allows the AI ​​agent according to the embodiment to improve the reliability of reviews, eliminate misinformation, and ensure fairness.

[0101] 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.

[0102] Data generation model 58 is a form 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> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. 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 (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0103] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0104] Each of the multiple elements, including the comparison unit, score calculation unit, and filtering unit described above, is implemented in at least one of the smart device 14 and the data processing device 12. For example, the comparison unit is implemented by the control unit 46A of the smart device 14 and automatically compares review comments. The score calculation unit is implemented by the identification processing unit 290 of the data processing device 12 and calculates a credit score. The filtering unit is implemented by the control unit 46A of the smart device 14 and filters reviews based on the credit score. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0105] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0106] 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.

[0107] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. 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 and / or LAN.

[0108] 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.

[0109] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, 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.

[0110] 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, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0111] 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.

[0112] 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 by the processor 28. The storage 32 stores the specific processing program 56.

[0113] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0114] 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. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0115] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0116] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0117] 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.

[0118] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0119] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0120] Each of the multiple elements, including the comparison unit, score calculation unit, and filtering unit described above, is implemented in at least one of the smart glasses 214 and the data processing device 12. For example, the comparison unit is implemented by the control unit 46A of the smart glasses 214 and automatically compares review comments. The score calculation unit is implemented by the identification processing unit 290 of the data processing device 12 and calculates a credit score. The filtering unit is implemented by the control unit 46A of the smart glasses 214 and filters reviews based on the credit score. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0121] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0122] 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.

[0123] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. 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 and / or LAN.

[0124] 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.

[0125] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, 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.

[0126] 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, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0127] 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.

[0128] 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.

[0129] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0130] 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. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0131] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0132] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0133] 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.

[0134] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0135] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0136] Each of the multiple elements, including the comparison unit, score calculation unit, and filtering unit described above, is implemented in at least one of the headset terminal 314 and the data processing device 12. For example, the comparison unit is implemented by the control unit 46A of the headset terminal 314 and automatically compares review comments. The score calculation unit is implemented by the identification processing unit 290 of the data processing device 12 and calculates a credit score. The filtering unit is implemented by the control unit 46A of the headset terminal 314 and filters reviews based on the credit score. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0137] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0138] 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.

[0139] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. 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 and / or LAN.

[0140] 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.

[0141] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, 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.

[0142] 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 image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0143] 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.

[0144] 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. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0145] 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.

[0146] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0147] 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. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0148] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0149] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0150] 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.

[0151] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0152] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0153] Each of the multiple elements, including the comparison unit, score calculation unit, and filtering unit described above, is implemented in at least one of the robot 414 and the data processing unit 12. For example, the comparison unit is implemented by the control unit 46A of the robot 414 and automatically compares review comments. The score calculation unit is implemented by the identification processing unit 290 of the data processing unit 12 and calculates a credit score. The filtering unit is implemented by the control unit 46A of the robot 414 and filters reviews based on the credit score. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0154] 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.

[0155] Figure 9 shows the 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.

[0156] 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.

[0157] 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.

[0158] 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, and motorcycles, 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 based, for example, 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.

[0159] 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."

[0160] 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.

[0161] 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 method for the specific process may be used, which includes computer 22 and multiple other computers.

[0162] 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.

[0163] 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.

[0164] 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.

[0165] 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.

[0166] 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.

[0167] 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.

[0168] 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.

[0169] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0170] 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 other things 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.

[0171] 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.

[0172] (Note 1) A comparison section that automatically compares review comments, A score calculation unit calculates a credit score based on the review evaluated by the comparison unit, A filtering unit that filters reviews based on the creditworthiness score calculated by the score calculation unit, A system equipped with these features. (Note 2) The comparison unit is, Natural language processing techniques are used to evaluate the consistency, specificity, and emotional bias of review content. The system described in Appendix 1, characterized by the features described herein. (Note 3) The score calculation unit, A credibility score is calculated in real time based on the reviewer's history and the characteristics of the text. The system described in Appendix 1, characterized by the features described herein. (Note 4) The filtering unit is Only reviews with a credit score exceeding a certain threshold will be displayed. The system described in Appendix 1, characterized by the features described herein. (Note 5) The comparison unit is, Detect reviews that are identical or nearly identical copies. The system described in Appendix 1, characterized by the features described herein. (Note 6) The score calculation unit, Reviews with specific examples and abundant information will receive high ratings, while reviews containing extreme language or repetitive content will receive low ratings. The system described in Appendix 1, characterized by the features described herein. (Note 7) The comparison unit is, The system estimates user sentiment and adjusts the evaluation criteria for review content based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 8) The comparison unit is, When evaluating the consistency and specificity of reviews, referencing the reviewer's past posting history improves the accuracy of the evaluation. The system described in Appendix 1, characterized by the features described herein. (Note 9) The comparison unit is, When evaluating review content, the frequency of specific keywords and phrases is taken into consideration. The system described in Appendix 1, characterized by the features described herein. (Note 10) The comparison unit is, It estimates user sentiment and adjusts the order in which review ratings are displayed based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 11) The comparison unit is, When evaluating review content, the reviewer's geographical location information will be taken into consideration. The system described in Appendix 1, characterized by the features described herein. (Note 12) The comparison unit is, When evaluating review content, we analyze the reviewer's social media activity to improve the accuracy of the evaluation. The system described in Appendix 1, characterized by the features described herein. (Note 13) The score calculation unit, The system estimates the user's emotions and adjusts the method for calculating the credit score based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 14) The score calculation unit, When calculating the credibility score, the quality and consistency of the reviewer's past posts are referenced to optimize the score. The system described in Appendix 1, characterized by the features described herein. (Note 15) The score calculation unit, When calculating the credibility score, the reviewer's posting frequency and activity history are taken into consideration and adjusted accordingly. The system described in Appendix 1, characterized by the features described herein. (Note 16) The score calculation unit, The system estimates the user's emotions and adjusts how the credit score is displayed based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 17) The score calculation unit, When calculating the credit score, the reviewer's geographical location information is taken into consideration. The system described in Appendix 1, characterized by the features described herein. (Note 18) The score calculation unit, When calculating the credit score, we analyze the reviewer's social media activity to improve the accuracy of the score. The system described in Appendix 1, characterized by the features described herein. (Note 19) The filtering unit is It estimates the user's sentiment and adjusts the filtering criteria based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 20) The filtering unit is During filtering, the displayed content is updated in real time to reflect changes in the credit score. The system described in Appendix 1, characterized by the features described herein. (Note 21) The filtering unit is Provides a customization feature to prioritize the display of posts from specific reviewers during filtering. The system described in Appendix 1, characterized by the features described herein. (Note 22) The filtering unit is It estimates the user's emotions and adjusts how the filtering results are displayed based on the estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 23) The filtering unit is When filtering, the displayed content is adjusted to take into account the reviewer's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 24) The filtering unit is During filtering, the system analyzes reviewers' social media activity to improve the accuracy of displayed content. The system described in Appendix 1, characterized by the features described herein. (Note 25) The filtering unit is When filtering, the displayed content is adjusted considering the reviewer's posting frequency. The system described in Appendix 1, characterized by the features described herein. [Explanation of symbols]

[0173] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. A comparison section that automatically compares review comments, A score calculation unit calculates a credit score based on the review evaluated by the comparison unit, A filtering unit that filters reviews based on the creditworthiness score calculated by the score calculation unit, A system equipped with these features.

2. The comparison unit is, Natural language processing techniques are used to evaluate the consistency, specificity, and emotional bias of review content. The system according to feature 1.

3. The score calculation unit, A credibility score is calculated in real time based on the reviewer's history and the characteristics of the text. The system according to feature 1.

4. The filtering unit is Only reviews with a credit score exceeding a certain threshold will be displayed. The system according to feature 1.

5. The comparison unit is, Detect reviews that are identical or nearly identical copies. The system according to feature 1.

6. The score calculation unit, Reviews with specific examples and abundant information will receive high ratings, while reviews containing extreme language or repetitive content will receive low ratings. The system according to feature 1.

7. The comparison unit is, The system estimates user sentiment and adjusts the evaluation criteria for review content based on the estimated user sentiment. The system according to feature 1.

8. The comparison unit is, When evaluating the consistency and specificity of reviews, referencing the reviewer's past posting history improves the accuracy of the evaluation. The system according to feature 1.

Images