System

Abstract

To provide a system according to an embodiment, capable of effectively performing a fact check.SOLUTION: A system according to an embodiment, comprises a specification part, a generation part, and a confirmation part. The specification part specifies an objective portion of a fact check. The generation part generates an inquiry against a place specified by the specification part. The confirmation part confirms fact relevance on the basis of the inquiry generated by the generation part.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

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

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] With conventional technology, fact-checking required a large amount of manpower and was difficult to carry out efficiently.

[0005] The system according to the embodiment aims to efficiently carry out fact-checking. [Means for solving the problem]

[0006] The system according to the embodiment includes an identification unit, a generation unit, and a confirmation unit. The identification unit identifies a target portion for fact checking. The generation unit generates a question for the portion identified by the identification unit. The confirmation unit confirms the factual relationship based on the question generated by the generation unit. Effect of the Invention

[0007] The system according to the embodiment can efficiently perform fact checking. [Brief description of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

[0028] (Example 1) The fact-checking system according to the embodiment of the present invention is a system that uses an LLM (large-scale language model) to make fact-checking more efficient. In this fact-checking system, first, the LLM analyzes a newly written text and identifies a section that needs fact-checking. Next, the LLM generates a question for that section. The generated question uses a browsing function to search for reliable sources on the Internet and confirm the factual relationships. Finally, based on the confirmed factual relationships, it is determined whether the text needs to be corrected. This mechanism significantly improves the efficiency of fact-checking and reduces the burden of manual work. For example, by having the LLM automatically generate questions and confirm the factual relationships using the browsing function, fact-checking can be performed more quickly and accurately than conventional manual fact-checking. In addition, since the questions generated by the LLM are intended to confirm specific factual relationships, it is possible to reduce unnecessary searches and collect information efficiently. Furthermore, when the LLM uses the browsing function to confirm factual relationships, it is designed to preferentially search reliable sources. This makes it possible to prevent fact-checking based on erroneous information and provide accurate information. For example, websites of government agencies and academic institutions, reliable news media, etc. are used as information sources. In this way, a fact-checking system using LLM can efficiently and accurately verify facts, and is expected to improve the reliability of articles in the media. As a result, the fact-checking system can significantly improve the efficiency of fact-checking and reduce the burden of manual work.

[0029] The fact-check system according to the embodiment includes an identifying unit, a generating unit, and a verifying unit. The identifying unit identifies a target portion of the fact-check. For example, the identifying unit uses a natural language processing algorithm to analyze a text and identify a portion that requires a fact-check. The identifying unit can also use a machine learning model to analyze a past fact-check history and identify the target portion based on a specific pattern. For example, the identifying unit preferentially identifies a portion that had many errors in the past. The identifying unit can also analyze the context of the text and identify the target portion based on a specific keyword or phrase. For example, the identifying unit identifies the target portion based on a specific keyword (e.g., statistical data, citation) in the text. The generating unit generates a question for the portion identified by the identifying unit. The generating unit generates a question for a specific data or event using, for example, a generating AI. The generating AI generates a question based on a specific factual relationship using a text generating AI (e.g., LLM). The generating unit can also generate an optimal question by referring to past question data. For example, the generating unit generates an optimal question based on question data used in the past. The verifying unit verifies the factual relationship based on the question generated by the generating unit. The confirmation unit, for example, uses a browsing function to search for reliable information sources on the Internet and confirms the factual relationships. The confirmation unit is designed to preferentially search for reliable information sources. For example, the confirmation unit uses websites of government agencies and academic institutions, reliable news media, etc. as information sources. As a result, the fact-checking system according to the embodiment can significantly improve the efficiency of fact-checking and reduce the burden of manual work. Some or all of the above-mentioned processing in the confirmation unit may be performed, for example, using AI or may be performed without using AI. For example, the confirmation unit can confirm the factual relationships using an AI model that uses the questions generated by the generation unit as input and confirms the factual relationships.

[0030] The identification unit identifies the target part of the fact check. For example, the identification unit uses a natural language processing algorithm to analyze the text and identify the part that needs to be fact checked. Specifically, the natural language processing algorithm analyzes the grammatical structure and meaning of the text and extracts important information and suspicious information based on the context. For example, the identification unit detects statistical data, quotations, assertions, etc. in the text and determines that they are likely to be the subject of the fact check. The identification unit can also use a machine learning model to analyze past fact check history and identify the target part based on a specific pattern. For example, the identification unit preferentially identifies parts that have had many errors in the past or information related to a specific theme. This allows the identification unit to efficiently find the part that needs to be fact checked. Furthermore, the identification unit can also analyze the context of the text and identify the target part based on a specific keyword or phrase. For example, the identification unit detects keywords such as "statistical data" and "quote" in the text and identifies information related to them as the target of the fact check. This allows the identification unit to efficiently extract important information from a wide range of text and improve the accuracy of the fact check.

[0031] The generation unit generates questions for the part identified by the identification unit. The generation unit generates questions for specific data or events using, for example, a generation AI. The generation AI generates questions based on specific factual relationships using a text generation AI (for example, LLM). Specifically, the generation AI generates appropriate questions based on information related to the part identified by the identification unit. For example, the generation AI generates questions such as "What is the source of this data?" or "How was this statistical data collected?" for statistical data. The generation unit can also generate optimal questions by referring to past question data. For example, the generation unit generates optimal questions for similar situations based on question data used in the past. This allows the generation unit to efficiently and effectively generate questions and improve the accuracy of fact checking. Furthermore, the generation unit can evaluate the quality of the generated questions and make corrections or improvements as necessary. For example, if the question generated by the generation AI is inappropriate, the generation unit corrects the content of the question and generates a more appropriate question. This allows the generation unit to always provide high-quality questions and improve the accuracy and efficiency of fact checking.

[0032] The verification unit verifies the factual relationships based on the questions generated by the generation unit. For example, the verification unit uses a browsing function to search for reliable information sources on the Internet and verify the factual relationships. Specifically, the verification unit uses a search engine to search for related information using the generated questions as input. The verification unit is designed to preferentially search for reliable information sources. For example, the verification unit uses websites of government agencies and academic institutions, reliable news media, etc. as information sources. This allows the verification unit to collect accurate and reliable information and verify the factual relationships. Furthermore, the verification unit analyzes the collected information and provides answers to the questions generated by the generation unit. For example, the verification unit verifies the source and collection method of statistical data based on the collected information and provides accurate answers. The verification unit can also verify the factual relationships using AI. For example, the verification unit can use an AI model that uses the questions generated by the generation unit as input to verify the factual relationships. This allows the verification unit to efficiently and accurately verify the factual relationships and improve the accuracy and efficiency of fact checking. Furthermore, the verification unit can record the verification results and use them as a reference for future fact checking. This allows the verification department to continually improve the accuracy and efficiency of fact-checking.

[0033] The identification unit can identify the target part of the fact check using a specific algorithm or method. The identification unit, for example, uses a natural language processing algorithm to analyze the text and identify the part that needs to be fact checked. The natural language processing algorithm analyzes the context of the text and identifies the target part based on a specific keyword or phrase. For example, the identification unit can identify the target part based on a specific keyword (e.g., statistical data, citation) in the text. The identification unit can also use a machine learning model to analyze the past fact check history and identify the target part based on a specific pattern. For example, the identification unit preferentially identifies parts that have had many errors in the past. In this way, by using a specific algorithm or method, the target part of the fact check can be efficiently identified. A part or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit can identify the target part using an AI model that analyzes the text using a natural language processing algorithm and identifies the part that needs to be fact checked.

[0034] The generation unit can generate questions for specific data or events. The generation unit generates questions for specific data or events using, for example, a generation AI. The generation AI generates questions based on specific factual relationships using a text generation AI (for example, LLM). For example, the generation unit generates questions based on specific data (e.g., statistical data, historical events). The generation unit can also generate optimal questions by referring to past question data. For example, the generation unit generates optimal questions based on question data used in the past. This allows efficient fact checking by generating questions for specific data or events. Some or all of the above-mentioned processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit can generate questions using an AI model that generates questions based on specific data or events.

[0035] The verification unit can search highly reliable sources with priority. The verification unit, for example, uses a browsing function to search highly reliable sources on the Internet and verify the factual relationships. Highly reliable sources include, for example, government and academic institutions' websites, and reliable news media. The verification unit is designed to search highly reliable sources with priority. For example, the verification unit evaluates the official website of a government institution as a highly reliable source and searches it with priority. The verification unit can also evaluate papers of academic institutions as highly reliable sources and search it with priority. This allows accurate fact checking by searching highly reliable sources with priority. Some or all of the above-mentioned processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can verify the factual relationships using an AI model that searches for highly reliable sources.

[0036] The identification unit can analyze the past fact-check history and identify the target part based on a specific pattern. The identification unit, for example, analyzes the past fact-check history and identifies the target part based on a specific pattern. The past fact-check history includes, for example, past check results and used data. The identification unit preferentially identifies parts that had many errors in the past. The identification unit can also identify parts related to a specific theme or topic from the past fact-check history. For example, the identification unit analyzes the tendency of a specific writer based on the past fact-check history and identifies the target part. In this way, the target part can be efficiently identified by analyzing the past fact-check history. A part or all of the above-mentioned processing in the identification unit may be performed, for example, using AI, or may be performed without using AI. For example, the identification unit can identify the target part using an AI model that analyzes the past fact-check history and identifies the target part based on a specific pattern.

[0037] The identification unit can analyze the context of the text and identify the target location based on a specific keyword or phrase. The identification unit can analyze the context of the text using, for example, a natural language processing algorithm and identify the target location based on a specific keyword or phrase. The natural language processing algorithm analyzes the context of the text and identifies the target location based on a specific keyword or phrase. For example, the identification unit can identify the target location based on a specific keyword (e.g., statistical data, citation) in the text. The identification unit can also identify a phrase that is likely to include a factual relationship based on the context of the text. For example, the identification unit can identify a paragraph that includes an important factual relationship based on the context of the text. In this way, the target location can be efficiently identified by analyzing the context of the text. A part or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit can identify the target location using an AI model that analyzes the context of the text using a natural language processing algorithm and identifies the target location based on a specific keyword or phrase.

[0038] The identification unit can customize a specific algorithm based on the genre or category of the text. The identification unit customizes the algorithm of the identification unit based on, for example, the genre or category of the text. The genre or category of the text includes, for example, news articles, academic papers, entertainment articles, and the like. In the case of news articles, the identification unit uses an algorithm that emphasizes speed. In addition, in the case of academic papers, the identification unit can also use an algorithm that performs detailed verification. Furthermore, in the case of entertainment articles, the identification unit can also use an algorithm that preferentially identifies parts that attract the interest of readers. In this way, by customizing the algorithm based on the genre or category of the text, fact checking can be performed efficiently. Part or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit can customize the algorithm of the identification unit using an AI model that customizes the algorithm based on the genre or category of the text.

[0039] The identification unit can double-check the identified part in cooperation with another AI system. For example, the identification unit transmits the part identified by the identification unit to another AI system and performs a double check. The identification unit can also evaluate the reliability of the identified part in cooperation with the other AI system. For example, the identification unit transmits the identified part to another natural language processing system and performs a double check. The identification unit can also determine the optimal identified part in cooperation with the other AI system. As a result, the accuracy of the fact check is improved by double-checking in cooperation with the other AI system. A part or all of the above-mentioned processing in the identification unit may be performed, for example, using AI or may be performed without using AI. For example, the identification unit can double-check the identified part using an AI model that double-checks the identified part in cooperation with the other AI system.

[0040] The generation unit can generate an optimal question by referring to past question data when generating a question. The generation unit generates an optimal question by referring to, for example, past question data. The past question data includes, for example, a history of past questions and answer results to questions. The generation unit generates an optimal question based on question data used in the past. The generation unit can also generate a question related to a specific theme from the past question data. For example, the generation unit analyzes past question data and generates an effective question. As a result, an optimal question is generated by referring to the past question data. A part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can generate a question using an AI model that generates an optimal question by referring to past question data.

[0041] The generation unit can adjust the level of detail of the question based on specific factual relationships when generating a question. The generation unit adjusts the level of detail of the question based on specific factual relationships, for example. The specific factual relationships include, for example, details of the incident and statistical data. The generation unit generates questions related to important factual relationships in detail. Also, questions related to general factual relationships can be generated concisely. Furthermore, the level of detail of the question can be adjusted based on specific factual relationships. As a result, an efficient question is generated by adjusting the level of detail of the question based on the specific factual relationships. A part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can adjust the level of detail of the question using an AI model that adjusts the level of detail of the question based on the specific factual relationships.

[0042] The generation unit can customize the format of the question based on the genre or category of the text when generating the question. The generation unit customizes the format of the question based on, for example, the genre or category of the text. The genre or category of the text includes, for example, news articles, academic papers, entertainment articles, and the like. In the case of news articles, the generation unit generates questions that emphasize timeliness. In addition, in the case of academic papers, the generation unit can also generate questions that perform detailed verification. Furthermore, in the case of entertainment articles, the generation unit can also generate questions that attract the interest of readers. As a result, efficient questions are generated by customizing the format of the question based on the genre or category of the text. Part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can customize the format of the question using an AI model that customizes the format of the question based on the genre or category of the text.

[0043] The generation unit can improve the accuracy of the question in cooperation with another AI system when generating a question. The generation unit can improve the accuracy of the question in cooperation with, for example, another AI system. The generation unit can evaluate the accuracy of the generated question in cooperation with, for example, another natural language processing system. The generation unit can also improve the accuracy of the question based on feedback from the other AI system. Furthermore, the generation unit can also generate an optimal question in cooperation with the other AI system. As a result, the accuracy of the question is improved by cooperation with the other AI system. A part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can improve the accuracy of the question by using an AI model that improves the accuracy of the question in cooperation with the other AI system.

[0044] The confirmation unit can optimize the confirmation algorithm by referring to past confirmation data at the time of confirmation. The confirmation unit optimizes the confirmation algorithm by referring to, for example, past confirmation data. The past confirmation data includes, for example, past confirmation results and used data. The confirmation unit selects an optimal confirmation algorithm based on the past confirmation data. The confirmation unit can also optimize a confirmation algorithm related to a specific theme from the past confirmation data. Furthermore, the confirmation unit can analyze the past confirmation data and optimize an effective confirmation algorithm. As a result, the confirmation algorithm is optimized by referring to the past confirmation data. A part or all of the above-mentioned processing in the confirmation unit may be performed, for example, using AI, or may be performed without using AI. For example, the confirmation unit can optimize the confirmation algorithm by using an AI model that optimizes the confirmation algorithm by referring to the past confirmation data.

[0045] The verification unit can evaluate the reliability of a specific information source at the time of verification and reflect it in the verification result. The verification unit, for example, evaluates the reliability of a specific information source and reflects it in the verification result. The reliability of a specific information source includes, for example, an official website of a government agency, a paper of an academic institution, and a reliable news media. The verification unit evaluates the official website of a government agency as a reliable information source. The verification unit can also evaluate a paper of an academic institution as a reliable information source. Furthermore, the verification unit can evaluate a reliable news media as an information source. As a result, an accurate verification result can be obtained by evaluating the reliability of a specific information source. A part or all of the above-mentioned processing in the verification unit may be performed, for example, using AI, or may be performed without using AI. For example, the verification unit can evaluate the reliability of a specific information source and reflect the verification result using an AI model that reflects the reliability of the specific information source in the verification result.

[0046] The verification unit can customize the verification algorithm based on the genre and category of the text at the time of verification. The verification unit customizes the verification algorithm based on, for example, the genre and category of the text. The genre and category of the text include, for example, news articles, academic papers, entertainment articles, and the like. In the case of news articles, the verification unit uses a verification algorithm that emphasizes timeliness. In addition, in the case of academic papers, the verification unit can also use a verification algorithm that performs detailed verification. Furthermore, in the case of entertainment articles, the verification unit can also use a verification algorithm that attracts the interest of readers. In this way, efficient verification can be performed by customizing the verification algorithm based on the genre and category of the text. Part or all of the above-mentioned processing in the verification unit may be performed, for example, using AI, or may be performed without using AI. For example, the verification unit can customize the verification algorithm using an AI model that customizes the verification algorithm based on the genre and category of the text.

[0047] The confirmation unit can improve the accuracy of the confirmation result in cooperation with other AI systems at the time of confirmation. The confirmation unit can improve the accuracy of the confirmation result in cooperation with other AI systems, for example. The confirmation unit can evaluate the accuracy of the confirmation result in cooperation with other AI systems. For example, the confirmation unit can evaluate the accuracy of the confirmation result in cooperation with other natural language processing systems. The confirmation unit can also improve the accuracy of the confirmation result based on feedback from the other AI systems. Furthermore, the confirmation unit can also provide an optimal confirmation result in cooperation with other AI systems. As a result, the accuracy of the confirmation result is improved by cooperation with other AI systems. A part or all of the above-mentioned processing in the confirmation unit may be performed, for example, using AI, or may be performed without using AI. For example, the confirmation unit can improve the accuracy of the confirmation result by using an AI model that improves the accuracy of the confirmation result in cooperation with other AI systems.

[0048] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0049] The identification unit can also analyze the user's past search history and identify target passages for fact checking based on a specific pattern. For example, it preferentially identifies keywords or topics that the user has frequently searched for in the past. The identification unit can also identify passages related to a specific theme or topic from the user's past search history. Furthermore, the identification unit can analyze the tendencies of a specific writer based on the user's past search history and identify target passages. This makes it possible to efficiently identify target passages by analyzing the user's past search history.

[0050] The generation unit can also analyze the user's past question history and generate optimal questions based on a specific pattern. For example, the generation unit preferentially generates keywords or topics that the user has frequently asked about in the past. The generation unit can also generate questions related to a specific theme or topic from the user's past question history. Furthermore, the generation unit can analyze a specific question format based on the user's past question history and generate optimal questions. In this way, optimal questions are efficiently generated by analyzing the user's past question history.

[0051] The verification unit can also analyze the user's past verification history and optimize the verification algorithm based on a specific pattern. For example, it prioritizes verification of keywords or topics that the user frequently verified in the past. The verification unit can also optimize a verification algorithm related to a specific theme or topic from the user's past verification history. Furthermore, the verification unit can analyze a specific verification format based on the user's past verification history and select an optimal verification algorithm. In this way, the verification algorithm is efficiently optimized by analyzing the user's past verification history.

[0052] The identification unit can also analyze the user's past search history and identify target passages for fact checking based on a specific pattern. For example, it preferentially identifies keywords or topics that the user has frequently searched for in the past. The identification unit can also identify passages related to a specific theme or topic from the user's past search history. Furthermore, the identification unit can analyze the tendencies of a specific writer based on the user's past search history and identify target passages. This makes it possible to efficiently identify target passages by analyzing the user's past search history.

[0053] The process flow of the first embodiment will be briefly described below.

[0054] Step 1: The identification unit identifies target passages for fact-checking. The identification unit uses natural language processing algorithms to analyze the text and identify passages that need to be fact-checked. It can also use machine learning models to analyze past fact-check history and identify target passages based on specific patterns. It also analyzes the context of the text and identifies target passages based on specific keywords and phrases. Step 2: The generator generates questions for the parts identified by the identifier. The generator uses a generation AI to generate questions for specific data or events. The generator uses a text generation AI (e.g., LLM) to generate questions based on specific factual relationships. It can also generate optimal questions by referring to past question data. Step 3: The verification unit verifies the facts based on the questions generated by the generation unit. The verification unit uses a browsing function to search for reliable information sources on the Internet and verify the facts. The verification unit uses websites of government agencies and academic institutions, reliable news media, etc. as information sources. The processing in the verification unit may be performed using AI or without using AI.

[0055] (Example 2) The fact-checking system according to the embodiment of the present invention is a system that uses an LLM (large-scale language model) to make fact-checking more efficient. In this fact-checking system, first, the LLM analyzes a newly written text and identifies a section that needs fact-checking. Next, the LLM generates a question for that section. The generated question uses a browsing function to search for reliable sources on the Internet and confirm the factual relationships. Finally, based on the confirmed factual relationships, it is determined whether the text needs to be corrected. This mechanism significantly improves the efficiency of fact-checking and reduces the burden of manual work. For example, by having the LLM automatically generate questions and confirm the factual relationships using the browsing function, fact-checking can be performed more quickly and accurately than conventional manual fact-checking. In addition, since the questions generated by the LLM are intended to confirm specific factual relationships, it is possible to reduce unnecessary searches and collect information efficiently. Furthermore, when the LLM uses the browsing function to confirm factual relationships, it is designed to preferentially search reliable sources. This makes it possible to prevent fact-checking based on erroneous information and provide accurate information. For example, websites of government agencies and academic institutions, reliable news media, etc. are used as information sources. In this way, a fact-checking system using LLM can efficiently and accurately verify facts, and is expected to improve the reliability of articles in the media. As a result, the fact-checking system can significantly improve the efficiency of fact-checking and reduce the burden of manual work.

[0056] The fact-check system according to the embodiment includes an identifying unit, a generating unit, and a verifying unit. The identifying unit identifies a target portion of the fact-check. For example, the identifying unit uses a natural language processing algorithm to analyze a text and identify a portion that requires a fact-check. The identifying unit can also use a machine learning model to analyze a past fact-check history and identify the target portion based on a specific pattern. For example, the identifying unit preferentially identifies a portion that had many errors in the past. The identifying unit can also analyze the context of the text and identify the target portion based on a specific keyword or phrase. For example, the identifying unit identifies the target portion based on a specific keyword (e.g., statistical data, citation) in the text. The generating unit generates a question for the portion identified by the identifying unit. The generating unit generates a question for a specific data or event using, for example, a generating AI. The generating AI generates a question based on a specific factual relationship using a text generating AI (e.g., LLM). The generating unit can also generate an optimal question by referring to past question data. For example, the generating unit generates an optimal question based on question data used in the past. The verifying unit verifies the factual relationship based on the question generated by the generating unit. The confirmation unit, for example, uses a browsing function to search for reliable information sources on the Internet and confirms the factual relationships. The confirmation unit is designed to preferentially search for reliable information sources. For example, the confirmation unit uses websites of government agencies and academic institutions, reliable news media, etc. as information sources. As a result, the fact-checking system according to the embodiment can significantly improve the efficiency of fact-checking and reduce the burden of manual work. Some or all of the above-mentioned processing in the confirmation unit may be performed, for example, using AI or may be performed without using AI. For example, the confirmation unit can confirm the factual relationships using an AI model that uses the questions generated by the generation unit as input and confirms the factual relationships.

[0057] The identification unit identifies the target part of the fact check. For example, the identification unit uses a natural language processing algorithm to analyze the text and identify the part that needs to be fact checked. Specifically, the natural language processing algorithm analyzes the grammatical structure and meaning of the text and extracts important information and suspicious information based on the context. For example, the identification unit detects statistical data, quotations, assertions, etc. in the text and determines that they are likely to be the subject of the fact check. The identification unit can also use a machine learning model to analyze past fact check history and identify the target part based on a specific pattern. For example, the identification unit preferentially identifies parts that have had many errors in the past or information related to a specific theme. This allows the identification unit to efficiently find the part that needs to be fact checked. Furthermore, the identification unit can also analyze the context of the text and identify the target part based on a specific keyword or phrase. For example, the identification unit detects keywords such as "statistical data" and "quote" in the text and identifies information related to them as the target of the fact check. This allows the identification unit to efficiently extract important information from a wide range of text and improve the accuracy of the fact check.

[0058] The generation unit generates questions for the part identified by the identification unit. The generation unit generates questions for specific data or events using, for example, a generation AI. The generation AI generates questions based on specific factual relationships using a text generation AI (for example, LLM). Specifically, the generation AI generates appropriate questions based on information related to the part identified by the identification unit. For example, the generation AI generates questions such as "What is the source of this data?" or "How was this statistical data collected?" for statistical data. The generation unit can also generate optimal questions by referring to past question data. For example, the generation unit generates optimal questions for similar situations based on question data used in the past. This allows the generation unit to efficiently and effectively generate questions and improve the accuracy of fact checking. Furthermore, the generation unit can evaluate the quality of the generated questions and make corrections or improvements as necessary. For example, if the question generated by the generation AI is inappropriate, the generation unit corrects the content of the question and generates a more appropriate question. This allows the generation unit to always provide high-quality questions and improve the accuracy and efficiency of fact checking.

[0059] The verification unit verifies the factual relationships based on the questions generated by the generation unit. For example, the verification unit uses a browsing function to search for reliable information sources on the Internet and verify the factual relationships. Specifically, the verification unit uses a search engine to search for related information using the generated questions as input. The verification unit is designed to preferentially search for reliable information sources. For example, the verification unit uses websites of government agencies and academic institutions, reliable news media, etc. as information sources. This allows the verification unit to collect accurate and reliable information and verify the factual relationships. Furthermore, the verification unit analyzes the collected information and provides answers to the questions generated by the generation unit. For example, the verification unit verifies the source and collection method of statistical data based on the collected information and provides accurate answers. The verification unit can also verify the factual relationships using AI. For example, the verification unit can use an AI model that uses the questions generated by the generation unit as input to verify the factual relationships. This allows the verification unit to efficiently and accurately verify the factual relationships and improve the accuracy and efficiency of fact checking. Furthermore, the verification unit can record the verification results and use them as a reference for future fact checking. This allows the verification department to continually improve the accuracy and efficiency of fact-checking.

[0060] The identification unit can identify the target part of the fact check using a specific algorithm or method. The identification unit, for example, uses a natural language processing algorithm to analyze the text and identify the part that needs to be fact checked. The natural language processing algorithm analyzes the context of the text and identifies the target part based on a specific keyword or phrase. For example, the identification unit can identify the target part based on a specific keyword (e.g., statistical data, citation) in the text. The identification unit can also use a machine learning model to analyze the past fact check history and identify the target part based on a specific pattern. For example, the identification unit preferentially identifies parts that have had many errors in the past. In this way, by using a specific algorithm or method, the target part of the fact check can be efficiently identified. A part or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit can identify the target part using an AI model that analyzes the text using a natural language processing algorithm and identifies the part that needs to be fact checked.

[0061] The generation unit can generate questions for specific data or events. The generation unit generates questions for specific data or events using, for example, a generation AI. The generation AI generates questions based on specific factual relationships using a text generation AI (for example, LLM). For example, the generation unit generates questions based on specific data (e.g., statistical data, historical events). The generation unit can also generate optimal questions by referring to past question data. For example, the generation unit generates optimal questions based on question data used in the past. This allows efficient fact checking by generating questions for specific data or events. Some or all of the above-mentioned processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit can generate questions using an AI model that generates questions based on specific data or events.

[0062] The verification unit can search highly reliable sources with priority. The verification unit, for example, uses a browsing function to search highly reliable sources on the Internet and verify the factual relationships. Highly reliable sources include, for example, government and academic institutions' websites, and reliable news media. The verification unit is designed to search highly reliable sources with priority. For example, the verification unit evaluates the official website of a government institution as a highly reliable source and searches it with priority. The verification unit can also evaluate papers of academic institutions as highly reliable sources and search it with priority. This allows accurate fact checking by searching highly reliable sources with priority. Some or all of the above-mentioned processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can verify the factual relationships using an AI model that searches for highly reliable sources.

[0063] The identification unit can estimate the user's emotion and determine the priority of the target parts of the fact check based on the estimated user's emotion. The identification unit estimates the user's emotion using, for example, facial expression recognition technology. For example, the identification unit captures the user's facial expression with a camera and estimates the emotion using an emotion estimation algorithm. The identification unit can also estimate the user's emotion using text analysis technology. For example, the identification unit analyzes text written by the user and estimates the emotion. The identification unit determines the priority of the target parts of the fact check based on the estimated user's emotion. For example, when the user feels anxious, the identification unit preferentially identifies the parts related to important factual relationships. In addition, when the user is relaxed, the identification unit can also identify the target parts taking into consideration the overall balance. Furthermore, when the user is in a hurry, the identification unit can also preferentially identify the parts that can be confirmed in a short time. As a result, by determining the priority based on the user's emotion, more effective fact checking can be performed. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit may identify target passages using an AI model that estimates a user's emotions and determines the priority order of target passages for fact checking based on the estimated emotions.

[0064] The identification unit can analyze the past fact-check history and identify the target part based on a specific pattern. The identification unit, for example, analyzes the past fact-check history and identifies the target part based on a specific pattern. The past fact-check history includes, for example, past check results and used data. The identification unit preferentially identifies parts that had many errors in the past. The identification unit can also identify parts related to a specific theme or topic from the past fact-check history. For example, the identification unit analyzes the tendency of a specific writer based on the past fact-check history and identifies the target part. In this way, the target part can be efficiently identified by analyzing the past fact-check history. A part or all of the above-mentioned processing in the identification unit may be performed, for example, using AI, or may be performed without using AI. For example, the identification unit can identify the target part using an AI model that analyzes the past fact-check history and identifies the target part based on a specific pattern.

[0065] The identification unit can analyze the context of the text and identify the target location based on a specific keyword or phrase. The identification unit can analyze the context of the text using, for example, a natural language processing algorithm and identify the target location based on a specific keyword or phrase. The natural language processing algorithm analyzes the context of the text and identifies the target location based on a specific keyword or phrase. For example, the identification unit can identify the target location based on a specific keyword (e.g., statistical data, citation) in the text. The identification unit can also identify a phrase that is likely to include a factual relationship based on the context of the text. For example, the identification unit can identify a paragraph that includes an important factual relationship based on the context of the text. In this way, the target location can be efficiently identified by analyzing the context of the text. A part or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit can identify the target location using an AI model that analyzes the context of the text using a natural language processing algorithm and identifies the target location based on a specific keyword or phrase.

[0066] The identification unit can estimate the user's emotion and adjust the operation of the identification unit based on the estimated user's emotion. The identification unit estimates the user's emotion using, for example, facial expression recognition technology. For example, the identification unit captures the user's facial expression with a camera and estimates the emotion using an emotion estimation algorithm. The identification unit can also estimate the user's emotion using text analysis technology. For example, the identification unit analyzes text written by the user and estimates the emotion. The identification unit adjusts the operation of the identification unit based on the estimated user's emotion. For example, when the user is stressed, the identification unit performs the operation quickly. Also, when the user is relaxed, the identification unit can perform the operation in detail. Furthermore, when the user is in a hurry, the identification unit can simplify the operation. As a result, by adjusting the operation of the identification unit based on the user's emotion, more effective fact checking can be performed. The estimation of emotion is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the determination unit may be performed using, for example, AI, or may be performed without using AI. For example, the determination unit can adjust the operation of the determination unit using an AI model that estimates the user's emotion and adjusts the operation of the determination unit based on the estimated emotion.

[0067] The identification unit can customize a specific algorithm based on the genre or category of the text. The identification unit customizes the algorithm of the identification unit based on, for example, the genre or category of the text. The genre or category of the text includes, for example, news articles, academic papers, entertainment articles, and the like. In the case of news articles, the identification unit uses an algorithm that emphasizes speed. In addition, in the case of academic papers, the identification unit can also use an algorithm that performs detailed verification. Furthermore, in the case of entertainment articles, the identification unit can also use an algorithm that preferentially identifies parts that attract the interest of readers. In this way, by customizing the algorithm based on the genre or category of the text, fact checking can be performed efficiently. Part or all of the above-mentioned processing in the identification unit may be performed using, for example, AI, or may be performed without using AI. For example, the identification unit can customize the algorithm of the identification unit using an AI model that customizes the algorithm based on the genre or category of the text.

[0068] The identification unit can double-check the identified part in cooperation with another AI system. For example, the identification unit transmits the part identified by the identification unit to another AI system and performs a double check. The identification unit can also evaluate the reliability of the identified part in cooperation with the other AI system. For example, the identification unit transmits the identified part to another natural language processing system and performs a double check. The identification unit can also determine the optimal identified part in cooperation with the other AI system. As a result, the accuracy of the fact check is improved by double-checking in cooperation with the other AI system. A part or all of the above-mentioned processing in the identification unit may be performed, for example, using AI or may be performed without using AI. For example, the identification unit can double-check the identified part using an AI model that double-checks the identified part in cooperation with the other AI system.

[0069] The generation unit can estimate the user's emotion and adjust the way of expressing the question based on the estimated user's emotion. The generation unit estimates the user's emotion using, for example, facial expression recognition technology. For example, the generation unit captures the user's facial expression with a camera and estimates the emotion using an emotion estimation algorithm. The generation unit can also estimate the user's emotion using text analysis technology. For example, the generation unit analyzes text written by the user and estimates the emotion. The generation unit adjusts the way of expressing the question based on the estimated user's emotion. For example, if the user is nervous, a concise and clear question is generated. Also, if the user is relaxed, a detailed question can be generated. Furthermore, if the user is in a hurry, a question that can be answered quickly can be generated. As a result, a more effective question is generated by adjusting the way of expressing the question based on the user's emotion. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processes in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit can adjust the way the question is phrased using an AI model that estimates the user's emotions and adjusts the way the question is phrased based on the estimated emotions.

[0070] The generation unit can generate an optimal question by referring to past question data when generating a question. The generation unit generates an optimal question by referring to, for example, past question data. The past question data includes, for example, a history of past questions and answer results to questions. The generation unit generates an optimal question based on question data used in the past. The generation unit can also generate a question related to a specific theme from the past question data. For example, the generation unit analyzes past question data and generates an effective question. As a result, an optimal question is generated by referring to the past question data. A part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can generate a question using an AI model that generates an optimal question by referring to past question data.

[0071] The generation unit can adjust the level of detail of the question based on specific factual relationships when generating a question. The generation unit adjusts the level of detail of the question based on specific factual relationships, for example. The specific factual relationships include, for example, details of the incident and statistical data. The generation unit generates questions related to important factual relationships in detail. Also, questions related to general factual relationships can be generated concisely. Furthermore, the level of detail of the question can be adjusted based on specific factual relationships. As a result, an efficient question is generated by adjusting the level of detail of the question based on the specific factual relationships. A part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can adjust the level of detail of the question using an AI model that adjusts the level of detail of the question based on the specific factual relationships.

[0072] The generation unit can estimate the user's emotion and determine the priority of questions based on the estimated user's emotion. The generation unit estimates the user's emotion using, for example, facial expression recognition technology. For example, the generation unit captures the user's facial expression with a camera and estimates the emotion using an emotion estimation algorithm. The generation unit can also estimate the user's emotion using text analysis technology. For example, the generation unit analyzes text written by the user and estimates the emotion. The generation unit determines the priority of questions based on the estimated user's emotion. For example, when the user feels anxious, important questions are preferentially generated. Also, when the user is relaxed, questions can be generated taking into account the overall balance. Furthermore, when the user is in a hurry, questions that can be answered quickly can be preferentially generated. As a result, more effective questions are generated by determining the priority of questions based on the user's emotion. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the generation unit may be performed using, for example, AI, or may be performed without using AI. For example, the generation unit can determine the priority of questions using an AI model that estimates the user's emotions and determines the priority of questions based on the estimated emotions.

[0073] The generation unit can customize the format of the question based on the genre or category of the text when generating the question. The generation unit customizes the format of the question based on, for example, the genre or category of the text. The genre or category of the text includes, for example, news articles, academic papers, entertainment articles, and the like. In the case of news articles, the generation unit generates questions that emphasize timeliness. In addition, in the case of academic papers, the generation unit can also generate questions that perform detailed verification. Furthermore, in the case of entertainment articles, the generation unit can also generate questions that attract the interest of readers. As a result, efficient questions are generated by customizing the format of the question based on the genre or category of the text. Part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can customize the format of the question using an AI model that customizes the format of the question based on the genre or category of the text.

[0074] The generation unit can improve the accuracy of the question in cooperation with another AI system when generating a question. The generation unit can improve the accuracy of the question in cooperation with, for example, another AI system. The generation unit can evaluate the accuracy of the generated question in cooperation with, for example, another natural language processing system. The generation unit can also improve the accuracy of the question based on feedback from the other AI system. Furthermore, the generation unit can also generate an optimal question in cooperation with the other AI system. As a result, the accuracy of the question is improved by cooperation with the other AI system. A part or all of the above-mentioned processing in the generation unit may be performed, for example, using AI, or may be performed without using AI. For example, the generation unit can improve the accuracy of the question by using an AI model that improves the accuracy of the question in cooperation with the other AI system.

[0075] The confirmation unit can estimate the user's emotion and adjust the display method of the confirmation result based on the estimated user's emotion. The confirmation unit estimates the user's emotion using, for example, facial expression recognition technology. For example, the confirmation unit captures the user's facial expression with a camera and estimates the emotion using an emotion estimation algorithm. The confirmation unit can also estimate the user's emotion using text analysis technology. For example, the confirmation unit analyzes text written by the user and estimates the emotion. The confirmation unit adjusts the display method of the confirmation result based on the estimated user's emotion. For example, when the user is nervous, a simple and highly visible display method is provided. Also, when the user is relaxed, a display method including detailed information can be provided. Furthermore, when the user is in a hurry, a display method that focuses on the main points can be provided. As a result, by adjusting the display method of the confirmation result based on the user's emotion, a more effective display can be performed. The emotion estimation is realized using, for example, an emotion engine or a generation AI, or the like, using an emotion estimation function. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processes in the confirmation unit may be performed using AI or may be performed without using AI. For example, the confirmation unit can estimate the user's emotions and adjust the display method of the confirmation result based on the estimated emotions using an AI model.

[0076] The confirmation unit can optimize the confirmation algorithm by referring to past confirmation data at the time of confirmation. The confirmation unit optimizes the confirmation algorithm by referring to, for example, past confirmation data. The past confirmation data includes, for example, past confirmation results and used data. The confirmation unit selects an optimal confirmation algorithm based on the past confirmation data. The confirmation unit can also optimize a confirmation algorithm related to a specific theme from the past confirmation data. Furthermore, the confirmation unit can analyze the past confirmation data and optimize an effective confirmation algorithm. As a result, the confirmation algorithm is optimized by referring to the past confirmation data. A part or all of the above-mentioned processing in the confirmation unit may be performed, for example, using AI, or may be performed without using AI. For example, the confirmation unit can optimize the confirmation algorithm by using an AI model that optimizes the confirmation algorithm by referring to the past confirmation data.

[0077] The verification unit can evaluate the reliability of a specific information source at the time of verification and reflect it in the verification result. The verification unit, for example, evaluates the reliability of a specific information source and reflects it in the verification result. The reliability of a specific information source includes, for example, an official website of a government agency, a paper of an academic institution, and a reliable news media. The verification unit evaluates the official website of a government agency as a reliable information source. The verification unit can also evaluate a paper of an academic institution as a reliable information source. Furthermore, the verification unit can evaluate a reliable news media as an information source. As a result, an accurate verification result can be obtained by evaluating the reliability of a specific information source. A part or all of the above-mentioned processing in the verification unit may be performed, for example, using AI, or may be performed without using AI. For example, the verification unit can evaluate the reliability of a specific information source and reflect the verification result using an AI model that reflects the reliability of the specific information source in the verification result.

[0078] The confirmation unit can estimate the user's emotion and determine the priority of the confirmation results based on the estimated user's emotion. The confirmation unit estimates the user's emotion using, for example, facial expression recognition technology. For example, the confirmation unit captures the user's facial expression with a camera and estimates the emotion using an emotion estimation algorithm. The confirmation unit can also estimate the user's emotion using text analysis technology. For example, the confirmation unit analyzes text written by the user and estimates the emotion. The confirmation unit determines the priority of the confirmation results based on the estimated user's emotion. For example, when the user feels anxious, important confirmation results are preferentially displayed. Also, when the user is relaxed, confirmation results can be displayed taking into consideration the overall balance. Furthermore, when the user is in a hurry, results that can be quickly confirmed can be preferentially displayed. As a result, by determining the priority of the confirmation results based on the user's emotion, more effective confirmation can be performed. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, but is not limited to, a text generation AI (for example, LLM) or a multimodal generation AI. Some or all of the above-mentioned processes in the confirmation unit may be performed using, for example, AI, or may be performed without using AI. For example, the confirmation unit can determine the priority of the confirmation results using an AI model that estimates the user's emotions and determines the priority of the confirmation results based on the estimated emotions.

[0079] The verification unit can customize the verification algorithm based on the genre and category of the text at the time of verification. The verification unit customizes the verification algorithm based on, for example, the genre and category of the text. The genre and category of the text include, for example, news articles, academic papers, entertainment articles, and the like. In the case of news articles, the verification unit uses a verification algorithm that emphasizes timeliness. In addition, in the case of academic papers, the verification unit can also use a verification algorithm that performs detailed verification. Furthermore, in the case of entertainment articles, the verification unit can also use a verification algorithm that attracts the interest of readers. In this way, efficient verification can be performed by customizing the verification algorithm based on the genre and category of the text. Part or all of the above-mentioned processing in the verification unit may be performed, for example, using AI, or may be performed without using AI. For example, the verification unit can customize the verification algorithm using an AI model that customizes the verification algorithm based on the genre and category of the text.

[0080] The confirmation unit can improve the accuracy of the confirmation result in cooperation with other AI systems at the time of confirmation. The confirmation unit can improve the accuracy of the confirmation result in cooperation with other AI systems, for example. The confirmation unit can evaluate the accuracy of the confirmation result in cooperation with other AI systems. For example, the confirmation unit can evaluate the accuracy of the confirmation result in cooperation with other natural language processing systems. The confirmation unit can also improve the accuracy of the confirmation result based on feedback from the other AI systems. Furthermore, the confirmation unit can also provide an optimal confirmation result in cooperation with other AI systems. As a result, the accuracy of the confirmation result is improved by cooperation with other AI systems. A part or all of the above-mentioned processing in the confirmation unit may be performed, for example, using AI, or may be performed without using AI. For example, the confirmation unit can improve the accuracy of the confirmation result by using an AI model that improves the accuracy of the confirmation result in cooperation with other AI systems.

[0081] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0082] The identification unit can also analyze the user's past search history and identify target passages for fact checking based on a specific pattern. For example, it preferentially identifies keywords or topics that the user has frequently searched for in the past. The identification unit can also identify passages related to a specific theme or topic from the user's past search history. Furthermore, the identification unit can analyze the tendencies of a specific writer based on the user's past search history and identify target passages. This makes it possible to efficiently identify target passages by analyzing the user's past search history.

[0083] The generation unit can also analyze the user's past question history and generate optimal questions based on a specific pattern. For example, the generation unit preferentially generates keywords or topics that the user has frequently asked about in the past. The generation unit can also generate questions related to a specific theme or topic from the user's past question history. Furthermore, the generation unit can analyze a specific question format based on the user's past question history and generate optimal questions. In this way, optimal questions are efficiently generated by analyzing the user's past question history.

[0084] The verification unit can also analyze the user's past verification history and optimize the verification algorithm based on a specific pattern. For example, it prioritizes verification of keywords or topics that the user frequently verified in the past. The verification unit can also optimize a verification algorithm related to a specific theme or topic from the user's past verification history. Furthermore, the verification unit can analyze a specific verification format based on the user's past verification history and select an optimal verification algorithm. In this way, the verification algorithm is efficiently optimized by analyzing the user's past verification history.

[0085] The identification unit can estimate the user's emotions and adjust the operation of the identification unit based on the estimated user's emotions. For example, if the user is stressed, the identification unit can perform the operation quickly. Also, if the user is relaxed, the identification unit can perform the operation in detail. Furthermore, if the user is in a hurry, the identification unit can simplify the operation. In this way, by adjusting the operation of the identification unit based on the user's emotions, more effective fact checking can be performed.

[0086] The generation unit can estimate the user's emotion and adjust the way in which the question is expressed based on the estimated user's emotion. For example, if the user is nervous, a concise and clear question can be generated. Also, if the user is relaxed, a detailed question can be generated. Furthermore, if the user is in a hurry, a question that can be answered quickly can be generated. In this way, by adjusting the way in which the question is expressed based on the user's emotion, more effective questions can be generated.

[0087] The confirmation unit can estimate the user's emotions and adjust the display method of the confirmation results based on the estimated user's emotions. For example, if the user is nervous, a simple and highly visible display method can be provided. If the user is relaxed, a display method including detailed information can be provided. Furthermore, if the user is in a hurry, a display method that focuses on the main points can be provided. In this way, a more effective display can be achieved by adjusting the display method of the confirmation results based on the user's emotions.

[0088] The identification unit can estimate the user's emotions and determine the priority of target parts for fact checking based on the estimated user's emotions. For example, if the user feels anxious, it preferentially identifies parts related to important factual relationships. Also, if the user is relaxed, it is possible to identify target parts taking into consideration the overall balance. Furthermore, if the user is in a hurry, it is possible to preferentially identify parts that can be checked in a short time. In this way, by determining the priority based on the user's emotions, more effective fact checking can be performed.

[0089] The generation unit can estimate the user's emotions and determine the priority order of questions based on the estimated user's emotions. For example, if the user feels anxious, important questions are generated with priority. Also, if the user feels relaxed, questions can be generated with consideration given to overall balance. Furthermore, if the user is in a hurry, questions that can be answered quickly can be generated with priority. In this way, more effective questions can be generated by determining the priority order of questions based on the user's emotions.

[0090] The confirmation unit can estimate the user's emotions and determine the priority order of confirmation results based on the estimated user's emotions. For example, if the user is feeling anxious, important confirmation results can be displayed with priority. Also, if the user is relaxed, confirmation results can be displayed with consideration given to the overall balance. Furthermore, if the user is in a hurry, results that can be confirmed quickly can be displayed with priority. In this way, by determining the priority order of confirmation results based on the user's emotions, more effective confirmation can be performed.

[0091] The identification unit can also analyze the user's past search history and identify target passages for fact checking based on a specific pattern. For example, it preferentially identifies keywords or topics that the user has frequently searched for in the past. The identification unit can also identify passages related to a specific theme or topic from the user's past search history. Furthermore, the identification unit can analyze the tendencies of a specific writer based on the user's past search history and identify target passages. This makes it possible to efficiently identify target passages by analyzing the user's past search history.

[0092] The process flow of the second embodiment will be briefly described below.

[0093] Step 1: The identification unit identifies target passages for fact-checking. The identification unit uses natural language processing algorithms to analyze the text and identify passages that need to be fact-checked. It can also use machine learning models to analyze past fact-check history and identify target passages based on specific patterns. It also analyzes the context of the text and identifies target passages based on specific keywords and phrases. Step 2: The generator generates questions for the parts identified by the identifier. The generator uses a generation AI to generate questions for specific data or events. The generator uses a text generation AI (e.g., LLM) to generate questions based on specific factual relationships. It can also generate optimal questions by referring to past question data. Step 3: The verification unit verifies the facts based on the questions generated by the generation unit. The verification unit uses a browsing function to search for reliable information sources on the Internet and verify the facts. The verification unit uses websites of government agencies and academic institutions, reliable news media, etc. as information sources. The processing in the verification unit may be performed using AI or without using AI.

[0094] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0095] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) are input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may also be an AI agent. In addition, when the processes of each of the above-mentioned parts are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples.In addition, processing performed by AI including the generation AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI including the generation AI.

[0096] Furthermore, the above-described processing by the data processing system 10 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or an external device.

[0097] Each of the multiple elements including the above-mentioned identification unit, generation unit, and confirmation unit is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the identification unit is realized by the processor 46 of the smart device 14, and analyzes the text to identify a portion that requires fact checking. The generation unit is realized, for example, by the identification processing unit 290 of the data processing device 12, and generates a question for the identified portion. The confirmation unit is realized, for example, by the control unit 46A of the smart device 14, and searches for a reliable information source on the Internet based on the generated question to confirm the factual relationship. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0098] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0099] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0100] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0101] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0102] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0103] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0104] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0105] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0106] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0107] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0108] In the smart glasses 214, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart glasses 214 have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0109] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0110] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0111] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0112] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device.

[0113] Each of the multiple elements including the above-mentioned identification unit, generation unit, and confirmation unit is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the identification unit is realized by the processor 46 of the smart glasses 214, and analyzes the text to identify a portion that requires fact checking. The generation unit is realized, for example, by the identification processing unit 290 of the data processing device 12, and generates a question for the identified portion. The confirmation unit is realized, for example, by the control unit 46A of the smart glasses 214, and searches for a reliable information source on the Internet based on the generated question to confirm the factual relationship. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0114] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0115] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0116] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0117] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0118] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0119] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0120] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0121] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0122] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0123] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0124] In the headset type terminal 314, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50, and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0125] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0126] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0127] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0128] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device.

[0129] Each of the multiple elements including the above-mentioned identifying unit, generating unit, and verifying unit is realized, for example, by at least one of the headset type terminal 314 and the data processing device 12. For example, the identifying unit is realized by the processor 46 of the headset type terminal 314, and analyzes the text to identify a portion that requires fact checking. The generating unit is realized, for example, by the identifying processing unit 290 of the data processing device 12, and generates a question for the identified portion. The verifying unit is realized, for example, by the control unit 46A of the headset type terminal 314, and searches for a reliable information source on the Internet based on the generated question to verify the factual relationship. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0130] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0131] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0132] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0133] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0134] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0135] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of a typical healthy person).

[0136] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0137] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and legs. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, and legs. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0138] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0139] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0140] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0141] In the robot 414, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads out the specific program 60 from the storage 50, and executes the specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. The robot 414 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0142] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0143] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0144] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0145] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0146] Each of the multiple elements including the above-mentioned identification unit, generation unit, and confirmation unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the identification unit is realized by the processor 46 of the robot 414, and analyzes the text to identify a portion that requires fact checking. The generation unit is realized, for example, by the identification processing unit 290 of the data processing device 12, and generates a question for the identified portion. The confirmation unit is realized, for example, by the control unit 46A of the robot 414, and searches for a reliable information source on the Internet based on the generated question to confirm the factual relationship. The correspondence between each unit and the device or control unit is not limited to the above-mentioned example, and various changes are possible.

[0147] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0148] FIG. 9 is a diagram showing an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0149] These emotions are distributed in the 3 o'clock direction of emotion map 400 and usually fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0150] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out you go on emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0151] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0152] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0153] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in Fig. 10. Fig. 10 shows an example in which multiple emotions, "relief," "calm," and "encouraging," have similar emotion values.

[0154] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers, including computer 22.

[0155] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable computer-readable non-transitory storage medium such as a Universal Serial Bus (USB) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0156] In addition, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 upon request from the data processing device 12.

[0157] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0158] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing a specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[0159] The hardware resource that executes the specific process may be one of these various processors, or may be a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0160] As an example of a configuration using one processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a configuration using a processor that realizes the functions of the entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0161] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[0162] In the above example, the first to fourth embodiments are described separately, but some or all of these embodiments may be combined. The smart device 14, the smart glasses 214, the headset terminal 314, and the robot 414 are only examples, and they may be combined with each other, or may be other devices. In the above example, the first and second embodiments are described separately, but they may be combined.

[0163] The above description and illustrations are detailed descriptions of the parts related to the technology of the present disclosure, and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, function, action, and effect is an example of the configuration, function, action, and effect of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require particular explanation in enabling the implementation of the technology of the present disclosure.

[0164] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0165] (Appendix 1) An identification unit that identifies the target part of the fact check; a generation unit that generates a question for the portion identified by the identification unit; a confirmation unit that confirms factual relationships based on the questions generated by the generation unit; Equipped A system characterized by: (Appendix 2) The identification unit is Use a specific algorithm or methodology to identify what to fact-check 2. The system of claim 1, (Appendix 3) The generation unit is Generate questions for specific data or events 2. The system of claim 1, (Appendix 4) The confirmation unit is Prioritize searching reliable sources 2. The system of claim 1, (Appendix 5) The identification unit is Infer user sentiment and prioritize fact-checking targets based on the inferred user sentiment 2. The system of claim 1, (Appendix 6) The identification unit is Analyze past fact-check history and target areas based on specific patterns 2. The system of claim 1, (Appendix 7) The identification unit is Analyzes the context of text and identifies target passages based on specific keywords or phrases 2. The system of claim 1, (Appendix 8) The identification unit is To estimate a user's emotion and adjust an operation of a specific part based on the estimated user's emotion 2. The system of claim 1, (Appendix 9) The identification unit is Customize specific algorithms based on the genre or category of text 2. The system of claim 1, (Appendix 10) The identification unit is Double-check identified areas in collaboration with other AI systems 2. The system of claim 1, (Appendix 11) The generation unit is Infer user sentiment and adjust question wording based on the inferred sentiment 2. The system of claim 1, (Appendix 12) The generation unit is When generating questions, the most appropriate questions are generated by referencing past question data. 2. The system of claim 1, (Appendix 13) The generation unit is When generating questions, adjust the level of detail based on specific facts 2. The system of claim 1, (Appendix 14) The generation unit is Estimate user sentiment and prioritize questions based on the estimated user sentiment 2. The system of claim 1, (Appendix 15) The generation unit is Customize question formatting based on text genre or category during question generation 2. The system of claim 1, (Appendix 16) The generation unit is When generating questions, the system works with other AI systems to improve the accuracy of the questions. 2. The system of claim 1, (Appendix 17) The confirmation unit is Inferring user emotions and adjusting the way confirmation results are displayed based on the estimated user emotions 2. The system of claim 1, (Appendix 18) The confirmation unit is During verification, the verification algorithm is optimized by referencing past verification data. 2. The system of claim 1, (Appendix 19) The confirmation unit is During verification, the reliability of specific sources is evaluated and reflected in the verification results. 2. The system of claim 1, (Appendix 20) The confirmation unit is Estimate the user's emotions and prioritize the confirmation results based on the estimated user emotions. 2. The system of claim 1, (Appendix 21) The confirmation unit is During review, customize the review algorithm based on the genre or category of the text 2. The system of claim 1, (Appendix 22) The confirmation unit is During verification, the system will work with other AI systems to improve the accuracy of the verification results. 2. The system of claim 1, [Explanation of symbols]

[0166] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. An identification unit that identifies the target part of the fact check; a generation unit that generates a question for the portion identified by the identification unit; a confirmation unit that confirms factual relationships based on the questions generated by the generation unit; Equipped A system characterized by:

2. The confirmation unit is Prioritize searching reliable sources 2. The system of claim 1 .

3. The identification unit is Infer user sentiment and prioritize fact-checking targets based on the inferred user sentiment 2. The system of claim 1 .

4. The identification unit is Analyze past fact-check history and target areas based on specific patterns 2. The system of claim 1 .

5. The identification unit is Analyzes the context of text and identifies target passages based on specific keywords or phrases 2. The system of claim 1 .

6. The identification unit is The emotion of the user is estimated, and the operation of the specifying unit is adjusted based on the estimated emotion of the user.

2. The system of claim 1 .

Images