system

Abstract

The system according to the embodiment aims to detect the possibility of special fraud and issue a warning without user operation. [Solution] The system according to the embodiment includes a monitoring unit, an analysis unit, a warning issuing unit, an information acquisition unit, and a learning unit. The monitoring unit monitors user conversations in real time. The analysis unit analyzes the content of the conversation monitored by the monitoring unit. The warning issuing unit issues a warning based on the results of the analysis by the analysis unit. The information acquisition unit acquires fraud information. The learning unit learns the fraud information acquired by the information acquisition unit.

Description

[Technical Field]

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

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

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

[0004] With conventional technology, there was a problem that the methods of special fraud were becoming more diverse, making it difficult for users to be careful.

[0005] The system according to the embodiment aims to detect the possibility of special fraud and issue a warning without user operation. [Means for solving the problem]

[0006] The system according to the embodiment includes a monitoring unit, an analysis unit, a warning issuing unit, an information acquisition unit, and a learning unit. The monitoring unit monitors user conversations in real time. The analysis unit analyzes the content of the conversation monitored by the monitoring unit. The warning issuing unit issues a warning based on the results of the analysis by the analysis unit. The information acquisition unit acquires fraud information. The learning unit learns the fraud information acquired by the information acquisition unit. [Effects of the Invention]

[0007] The system according to the embodiment can detect the possibility of special fraud and issue a warning without any user operation. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

[0028] (Example 1) The fraud prevention system according to an embodiment of the present invention automatically monitors user conversations, analyzes the conversation content, and issues a warning if a fraud is suspected. This fraud prevention system effectively prevents frauds by eliminating user interaction and making decisions based on the latest fraud information. For example, the fraud prevention system monitors user conversations in real time. The AI ​​then analyzes the conversation content and determines whether a fraud is suspected. If a fraud is suspected, a warning is issued to the user. This warning is displayed via voice or text, allowing the user to immediately become aware of the possibility of fraud. Furthermore, the system constantly references the latest fraud information, and the AI ​​makes decisions based on that information. For example, the system automatically obtains the latest fraud information provided by the police and related organizations, and the AI ​​learns from that information. This allows the system to constantly respond to the latest fraud techniques. This system prevents frauds without the user having to manually operate the system. For example, if elderly people are at high risk of falling victim to telephone fraud, this system can automatically monitor conversations and issue warnings if there is a possibility of fraud, preventing fraud for elderly people. This system can also be used not only in homes, but also in businesses and public facilities. For example, introducing this system in a company's customer support center can prevent fraud before it occurs. Furthermore, its use in public facilities is expected to reduce fraud damage throughout the entire community. In this way, the special fraud prevention system using generative AI can effectively prevent special fraud damage because it can be used without user operation and makes judgments based on the latest fraud information. As a result, the special fraud prevention system automatically monitors user conversations, the generative AI analyzes the content of the conversation, and issues warnings if there is a possibility of special fraud, preventing fraud damage.

[0029] The special fraud prevention system according to the embodiment includes a monitoring unit, an analysis unit, a warning issuing unit, an information acquisition unit, and a learning unit. The monitoring unit monitors user conversations in real time. User conversations include, but are not limited to, telephone calls, video calls, and face-to-face conversations. The monitoring unit converts the conversation content into text data using, for example, voice recognition technology. In the case of video calls, the monitoring unit can also analyze video data and extract the conversation content. In the case of face-to-face conversations, the monitoring unit can also collect and analyze audio data using a microphone. The analysis unit analyzes the conversation content monitored by the monitoring unit using a generation AI. The analysis is performed using, for example, natural language processing technology, but is not limited to, an example. For example, the generation AI analyzes the conversation content using a text generation AI (e.g., LLM) to determine whether there is a possibility of special fraud. The analysis unit can also analyze audio and video data using a multimodal generation AI. In addition, the analysis unit can extract and analyze important parts of the conversation content using the generation AI. The warning issuing unit issues a warning based on the results of the analysis by the analysis unit. The warning may be issued in the form of, for example, an audio alert, a text message, a visual warning, or the like, but is not limited to these examples. For example, the warning issuing unit may warn the user "Possible fraud" as an audio alert. The warning issuing unit may also send a warning message to the user's smartphone as a text message. Furthermore, the warning issuing unit may also display a warning message on the user's device screen as a visual warning. The information acquisition unit automatically acquires the latest fraud information provided by the police and related organizations. Information acquisition may be performed, for example, using an API, but is not limited to these examples. For example, the information acquisition unit may acquire the latest fraud information from a police database. The information acquisition unit may also scrape fraud information from the websites of related organizations. Furthermore, the information acquisition unit may collect fraud information from news articles and social media. The learning unit learns the fraud information acquired by the information acquisition unit, and the analysis unit uses the results.Learning is performed using, for example, a machine learning algorithm, but is not limited to such an example. For example, the learning unit uses supervised learning to learn patterns of fraudulent methods. The learning unit can also use unsupervised learning to discover new fraudulent methods. Furthermore, the learning unit can also use reinforcement learning to improve the accuracy of detecting fraudulent methods. As a result, the special fraud prevention system according to the embodiment can prevent fraud by monitoring user conversations in real time and issuing warnings based on the analysis results.

[0030] During monitoring, the monitoring unit can select the optimal monitoring method by referring to the user's past conversation history. For example, the monitoring unit may focus on monitoring specific keywords based on conversation patterns in which the user has previously shown signs of fraud. The monitoring unit can also detect, for example, a high risk of fraud during a specific time period from the user's past conversation history and intensify monitoring during that time period. The monitoring unit can also analyze the user's past conversation history and prioritize monitoring conversations with specific parties. This allows the optimal monitoring method to be selected by referring to the user's past conversation history. The past conversation history is stored, for example, in a text database or a voice database. The optimal monitoring method is selected using algorithms such as voice analysis, text analysis, and behavioral analysis. Some or all of the above-described processing in the monitoring unit may be performed using, for example, AI, or without AI. For example, the monitoring unit can input past conversation history data into a generation AI and have the generation AI select the optimal monitoring method.

[0031] During monitoring, the monitoring unit may perform filtering based on the user's lifestyle or areas of interest. For example, if the user is elderly, the monitoring unit may focus on monitoring specific keywords that are considered to be at high risk of fraud. Furthermore, for example, if the user is interested in finance-related topics, the monitoring unit may prioritize monitoring conversations related to financial fraud. Furthermore, the monitoring unit may strengthen monitoring during specific time periods or situations depending on the user's lifestyle. By performing filtering based on the user's lifestyle and areas of interest, conversations with a high risk of fraud can be prioritized for monitoring. The lifestyle is acquired based on information such as residence, occupation, and daily behavior patterns. The areas of interest are acquired based on information such as hobbies, topics of interest, and past search history. Some or all of the above-described processing in the monitoring unit may be performed using, or without, AI. For example, the monitoring unit may input data on the user's lifestyle and areas of interest into a generation AI and have the generation AI perform filtering.

[0032] During monitoring, the monitoring unit can prioritize monitoring highly relevant conversations based on the user's geographical location information. For example, if the user is in a specific area, the monitoring unit prioritizes monitoring conversations related to fraud methods that are likely to occur in that area. Furthermore, for example, if the user is traveling, the monitoring unit can prioritize monitoring conversations that pose a high risk of fraud at the user's travel destination. Furthermore, for example, if the user is at home, the monitoring unit can prioritize monitoring conversations related to fraud methods that are likely to occur at home. This allows highly relevant conversations to be prioritized by taking the user's geographical location information into consideration. Geographical location information is acquired based on, for example, GPS data or an IP address. Highly relevant conversations are selected using criteria such as the frequency of occurrence of specific keywords or past fraud patterns. Some or all of the above-described processing by the monitoring unit may be performed using, for example, AI, or without AI. For example, the monitoring unit can input the user's geographical location data into a generation AI and cause the generation AI to select highly relevant conversations.

[0033] During monitoring, the monitoring unit can analyze the user's social media activity and monitor related conversations. For example, if the user posts about fraud on social media, the monitoring unit can prioritize monitoring conversations related to that content. Furthermore, for example, if the user shares information about a specific fraud, the monitoring unit can also monitor conversations related to that information. Furthermore, for example, the monitoring unit can prioritize monitoring conversations with people who are identified as being at high risk for fraud based on the user's social media activity. This allows for prioritized monitoring of related conversations by analyzing the user's social media activity. Social media activity is analyzed based on, for example, the content of posts, comments, the number of likes, and the like. Relevant conversations are selected using criteria such as conversations about a specific topic or conversations with a specific user. Some or all of the above-described processing by the monitoring unit may be performed using, for example, AI, or may be performed without AI. For example, the monitoring unit can input the user's social media data into a generation AI and cause the generation AI to select relevant conversations.

[0034] During analysis, the analysis unit can adjust the level of detail of the analysis based on the importance of the conversation. For example, if an important keyword is included in a conversation, the analysis unit analyzes that part in detail. Furthermore, for example, if the importance of the conversation is low, the analysis unit can perform a normal analysis to save resources. Furthermore, for example, if the importance of the conversation is high, the analysis unit can perform a detailed analysis to avoid missing signs of fraud. Thus, by adjusting the level of detail of the analysis based on the importance of the conversation, important conversations can be analyzed in detail. The importance of the conversation is evaluated using criteria such as risk assessment and impact assessment. The level of detail of the analysis is adjusted using criteria such as the depth of the analysis and the type of algorithm used. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input conversation data to a generation AI and have the generation AI perform a detailed analysis based on the importance.

[0035] During analysis, the analysis unit can apply different analysis algorithms depending on the category of the conversation. For example, in the case of a conversation related to financial fraud, the analysis unit can apply an analysis algorithm specialized for financial fraud. Furthermore, in the case of a conversation related to "I'm your son" fraud, the analysis unit can also apply an analysis algorithm specialized for "I'm your son" fraud. Furthermore, in the case of a conversation related to online fraud, the analysis unit can also apply an analysis algorithm specialized for online fraud. By applying different analysis algorithms depending on the category of the conversation, signs of fraud can be detected more accurately. The categories of conversations are classified using criteria such as business conversations, personal conversations, and fraud-related conversations. The analysis algorithm is applied using criteria such as natural language processing algorithms and machine learning algorithms. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or without AI. For example, the analysis unit can input conversation data into a generation AI and have the generation AI perform analysis according to the category.

[0036] During analysis, the analysis unit can determine the analysis priority based on the time of submission of the conversation. For example, the analysis unit may prioritize analyzing recent conversations to address the latest fraudulent methods. The analysis unit can also analyze past conversations to check for signs of fraud. For example, the analysis unit can prioritize analyzing conversations submitted during a specific time period to evaluate the fraud risk during that time period. This allows the latest fraudulent methods to be addressed by determining the analysis priority based on the time of submission of the conversation. The time of submission of the conversation is obtained based on, for example, a timestamp or the date and time of submission. The analysis priority is determined using criteria such as risk assessment and importance assessment. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input conversation data into a generation AI and have the generation AI determine the priority based on the time of submission.

[0037] During analysis, the analysis unit can adjust the analysis order based on the relevance of the conversations. For example, the analysis unit prioritizes analysis of conversations that are deemed to have a high risk of fraud. The analysis unit can also prioritize analysis of important conversations, for example, by postponing analysis of less relevant conversations. The analysis unit can also evaluate the relevance of conversations and prioritize analysis of the most relevant conversations. This allows important conversations to be prioritized by adjusting the analysis order based on the relevance of the conversations. The relevance of conversations is evaluated using criteria such as common keywords or related topics. The analysis order is adjusted using criteria such as analyzing conversations with higher priority. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input conversation data to a generation AI and have the generation AI execute an analysis order based on the relevance.

[0038] The warning issuing unit can adjust the level of detail of the warning based on the importance of the conversation when issuing a warning. For example, the warning issuing unit issues a detailed warning for an important conversation. The warning issuing unit can also issue a brief warning for a conversation with low importance. The warning issuing unit can also issue a warning with an appropriate level of detail for a conversation with medium importance. Thus, by adjusting the level of detail of the warning based on the importance of the conversation, detailed warnings can be issued for important conversations. The importance of the conversation is evaluated using criteria such as risk assessment and impact assessment. The level of detail of the warning is adjusted using criteria such as a detailed explanation or a brief warning. Some or all of the above-described processing in the warning issuing unit may be performed using, for example, AI, or may be performed without using AI. For example, the warning issuing unit can input conversation data to a generation AI and cause the generation AI to issue detailed warnings based on the importance.

[0039] When issuing a warning, the warning issuing unit can apply different warning algorithms depending on the category of the conversation. For example, in the case of a conversation related to financial fraud, the warning issuing unit can apply a warning algorithm specialized for financial fraud. Furthermore, in the case of a conversation related to "I'm your son" fraud, the warning issuing unit can also apply a warning algorithm specialized for "I'm your son" fraud. Furthermore, in the case of a conversation related to online fraud, the warning issuing unit can also apply a warning algorithm specialized for online fraud. In this way, by applying different warning algorithms depending on the category of the conversation, appropriate warnings can be issued for conversations with a high risk of fraud. Conversation categories are classified using criteria such as business conversations, personal conversations, and fraud-related conversations. The warning algorithm is applied using criteria such as a risk assessment algorithm and an impact assessment algorithm. Some or all of the above-mentioned processing in the warning issuing unit may be performed using, for example, AI, or may be performed without using AI. For example, the warning issuing unit can input conversation data to a generation AI and cause the generation AI to execute a warning according to the category.

[0040] The alert issuing unit can determine the priority of alerts based on when the conversation was submitted. For example, the alert issuing unit can prioritize alerts for recent conversations. It can also issue alerts for past conversations to check for signs of fraud. Furthermore, the alert issuing unit can prioritize alerts for conversations submitted within a specific time period. This allows for a response to the latest fraud techniques by prioritizing alerts based on when the conversation was submitted. The conversation submission date is obtained based on, for example, a timestamp or submission date and time. The priority of alerts is determined using criteria such as risk assessment or importance assessment. Some or all of the above processing in the alert issuing unit may be performed using, for example, AI, or not. For example, the alert issuing unit can input conversation data into a generating AI and have the generating AI perform the priority determination based on the submission date.

[0041] The attention-issuing unit can adjust the order of attention based on the relevance of the conversations when issuing an attention. For example, the attention-issuing unit can prioritize issuing attention to conversations that are considered to have a high risk of fraud. The attention-issuing unit can also, for example, postpone issuing attention to less relevant conversations and issue attention first to important conversations. Furthermore, the attention-issuing unit can, for example, evaluate the relevance of conversations and prioritize issuing attention to the most relevant conversations. In this way, by adjusting the order of attention based on the relevance of conversations, it is possible to prioritize issuing attention to important conversations. The relevance of conversations is evaluated using criteria such as common keywords or related topics. The adjustment of the order of attention is performed using criteria such as issuing attention to high-priority conversations first. Some or all of the above processing in the attention-issuing unit may be performed using AI, for example, or not using AI. For example, the attention-issuing unit can input conversation data into a generating AI and have the generating AI execute the relevance-based attention order.

[0042] The information acquisition unit can select the optimal information acquisition method by referring to past fraud information when acquiring information. For example, the information acquisition unit can prioritize acquiring information containing specific keywords based on past fraud information. The information acquisition unit can also prioritize acquiring information related to fraud methods that are likely to occur during specific time periods from past fraud information. Furthermore, the information acquisition unit can analyze past fraud information and prioritize acquiring the most relevant information. This allows the optimal information acquisition method to be selected by referring to past fraud information. Past fraud information is stored, for example, in a database or archive. The selection of the optimal information acquisition method is performed using criteria such as web scraping or API usage. Some or all of the above processing in the information acquisition unit may be performed using AI, for example, or without AI. For example, the information acquisition unit can input past fraud information data into a generating AI and have the generating AI select the optimal information acquisition method.

[0043] The information acquisition unit can apply different information acquisition algorithms depending on the category of fraud information when acquiring information. For example, in the case of information related to financial fraud, the information acquisition unit can apply an information acquisition algorithm specialized for financial fraud. Also, in the case of information related to "ore-ore" fraud (a type of telephone fraud), the information acquisition unit can apply an information acquisition algorithm specialized for "ore-ore" fraud. Also, in the case of information related to online fraud, the information acquisition unit can apply an information acquisition algorithm specialized for online fraud. By applying different information acquisition algorithms depending on the category of fraud information, it is possible to prioritize the acquisition of information with a high risk of fraud. Fraud information categories are classified using criteria such as phishing scams, telephone scams, and email scams. The application of information acquisition algorithms is carried out using criteria such as natural language processing algorithms and machine learning algorithms. Some or all of the above processing in the information acquisition unit may be performed using AI, for example, or without AI. For example, the information acquisition unit can input fraud information data into a generating AI and have the generating AI perform information acquisition according to the category.

[0044] When acquiring information, the information acquisition unit can determine the priority of information acquisition based on the time of submission of the fraud information. The information acquisition unit, for example, prioritizes the acquisition of recent fraud information to respond to the latest fraud methods. The information acquisition unit can also, for example, acquire past fraud information and check whether there were any signs of fraud. The information acquisition unit can also, for example, prioritize the acquisition of fraud information submitted during a specific time period and evaluate the fraud risk during that time period. This allows the latest fraud methods to be responded to by determining the priority of information acquisition based on the time of submission of the fraud information. The time of submission of the fraud information is acquired based on, for example, a timestamp or the date and time of submission. The priority of information acquisition is determined using criteria such as risk assessment and importance assessment. Some or all of the above-mentioned processing in the information acquisition unit may be performed using, for example, AI, or may be performed without AI. For example, the information acquisition unit can input fraud information data to a generation AI and cause the generation AI to determine the priority based on the time of submission.

[0045] When acquiring information, the information acquisition unit can adjust the order of information acquisition based on the relevance of the fraud information. For example, the information acquisition unit prioritizes acquisition of information that is deemed to have a high risk of fraud. The information acquisition unit can also, for example, postpone acquisition of less relevant information and acquire important information first. The information acquisition unit can also, for example, evaluate the relevance of the fraud information and prioritize acquisition of the most relevant information. In this way, important information can be prioritized by adjusting the order of information acquisition based on the relevance of the fraud information. The relevance of the fraud information is evaluated using criteria such as common keywords or related topics. The order of information acquisition is adjusted using criteria such as a method of acquiring information first with a higher priority. Some or all of the above-described processing in the information acquisition unit may be performed using, for example, AI, or may be performed without AI. For example, the information acquisition unit can input fraud information data to a generation AI and cause the generation AI to execute an information acquisition order based on relevance.

[0046] The learning unit can adjust its learning algorithm by referring to past learning data during the learning process. For example, the learning unit can apply a learning algorithm specialized for a particular fraud scheme based on past learning data. The learning unit can also select the most effective learning algorithm from past learning data. Furthermore, the learning unit can analyze past learning data to improve the accuracy of the learning algorithm. This allows the optimal learning algorithm to be applied by referring to past learning data. Past learning data is stored, for example, in a database or archive. The application of the learning algorithm is performed using criteria such as supervised learning, unsupervised learning, or reinforcement learning. Some or all of the above processes in the learning unit may be performed using AI, for example, or without AI. For example, the learning unit can input past learning data into a generating AI and have the generating AI apply the optimal learning algorithm.

[0047] The learning unit can apply different learning algorithms during learning depending on the category of fraud information. For example, if the data is related to financial fraud, the learning unit will apply a learning algorithm specialized for financial fraud. Similarly, if the data is related to "ore-ore" fraud (a type of phone scam), the learning unit can apply a learning algorithm specialized for that type of fraud. Furthermore, if the data is related to online fraud, the learning unit can apply a learning algorithm specialized for online fraud. This allows the learning unit to prioritize learning information with a high risk of fraud by applying different learning algorithms depending on the category of fraud information. Fraud information categories are classified using criteria such as phishing scams, phone scams, and email scams. The application of learning algorithms is performed using criteria such as natural language processing algorithms and machine learning algorithms. Some or all of the above processing in the learning unit may be performed using AI, or not. For example, the learning unit can input fraud information data into a generating AI and have the generating AI perform learning according to the categories.

[0048] During learning, the learning unit can weight the learning data based on the time when the fraud information was submitted. For example, the learning unit may prioritize learning recent fraud information to respond to the latest fraud methods. The learning unit can also, for example, learn past fraud information and check whether there were any signs of fraud. The learning unit can also, for example, prioritize learning fraud information submitted during a specific time period to evaluate the fraud risk during that time period. This allows the latest fraud methods to be responded to by weighting the learning data based on the time when the fraud information was submitted. The time when the fraud information was submitted is obtained based on, for example, a timestamp or the date and time of submission. The weighting of the learning data is performed using criteria such as a method of assigning a higher weight to the most recent data. Some or all of the above-described processing in the learning unit may be performed using, for example, AI, or may be performed without AI. For example, the learning unit can input fraud information data into a generation AI and cause the generation AI to perform weighting based on the time of submission.

[0049] During learning, the learning unit can adjust the order of learning data based on the relevance of fraud information. For example, the learning unit prioritizes learning information that is deemed to have a high risk of fraud. The learning unit can also, for example, postpone learning information with low relevance and learn important information first. The learning unit can also, for example, evaluate the relevance of fraud information and prioritize learning the most relevant information. In this way, important information can be prioritized by adjusting the order of learning data based on the relevance of fraud information. The relevance of fraud information is evaluated using criteria such as common keywords or related topics. The order of learning data is adjusted using criteria such as a method of learning from high-priority data. Some or all of the above-described processing in the learning unit may be performed using, for example, AI, or may be performed without AI. For example, the learning unit can input fraud information data into a generation AI and have the generation AI execute the order of learning data based on the relevance.

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

[0051] The monitoring unit can estimate the user's health condition and adjust the intensity of monitoring based on the estimated health condition. For example, if the user is in poor health, the monitoring intensity can be increased to detect signs of fraud early. Also, if the user is healthy, the monitoring intensity can be maintained at a normal level to avoid excessive monitoring. Furthermore, if the user is feeling stressed, the monitoring intensity can be temporarily reduced to reduce the burden on the user. In this way, by adjusting the monitoring intensity according to the user's health condition, signs of fraud can be detected early.

[0052] The monitoring unit can select the optimal monitoring method by referring to the user's past behavioral patterns. For example, it can focus on monitoring specific behaviors based on behavioral patterns in which the user has shown signs of fraud in the past. It can also detect a high risk of fraud during specific time periods from the user's past behavioral patterns and intensify monitoring during those times. It can also analyze the user's past behavioral patterns and prioritize monitoring of interactions with specific parties. This makes it possible to select the optimal monitoring method by referring to the user's past behavioral patterns.

[0053] The monitoring unit can perform filtering based on the user's lifestyle or areas of interest. For example, if the user is elderly, it can focus on monitoring specific keywords that are considered to be at high risk of fraud. Also, if the user is interested in finance-related topics, it can prioritize monitoring of conversations related to financial fraud. Furthermore, it can strengthen monitoring during specific time periods or situations depending on the user's lifestyle. In this way, by filtering based on the user's lifestyle and areas of interest, it is possible to focus on monitoring conversations that pose a high risk of fraud.

[0054] The monitoring unit can prioritize monitoring conversations that are highly relevant based on the user's geographical location. For example, if a user is in a specific region, it can prioritize monitoring conversations related to fraud schemes that are likely to occur in that region. Similarly, if a user is traveling, it can prioritize monitoring conversations related to fraud risks in their travel destination. Furthermore, if a user is at home, it can prioritize monitoring conversations related to fraud schemes that are likely to occur within the home. In this way, by considering the user's geographical location, it is possible to prioritize monitoring conversations that are highly relevant.

[0055] The monitoring unit can analyze users' social media activity and monitor relevant conversations. For example, if a user posts about fraud on social media, it can prioritize monitoring conversations related to that content. It can also monitor conversations related to information shared by a user about a specific type of fraud. Furthermore, it can prioritize monitoring conversations with individuals deemed to pose a high risk of fraud based on the user's social media activity. In this way, by analyzing a user's social media activity, it can prioritize monitoring of relevant conversations.

[0056] The analysis unit can adjust the level of detail in its analysis based on the importance of the conversation. For example, if a conversation contains important keywords, it will analyze that section in detail. Conversely, if the conversation is of low importance, it can perform a standard analysis to conserve resources. Furthermore, if the conversation is of high importance, it can perform a detailed analysis to ensure that no signs of fraud are missed. In this way, by adjusting the level of detail in the analysis based on the importance of the conversation, important conversations can be analyzed in detail.

[0057] The analysis unit can apply different analysis algorithms depending on the category of the conversation. For example, in the case of a conversation related to financial fraud, an analysis algorithm specialized for financial fraud can be applied. In addition, in the case of a conversation related to "it's my son" fraud, an analysis algorithm specialized for "it's my son" fraud can be applied. Furthermore, in the case of a conversation related to online fraud, an analysis algorithm specialized for online fraud can be applied. In this way, by applying different analysis algorithms depending on the category of the conversation, it is possible to more accurately detect signs of fraud.

[0058] The processing flow of the first embodiment will be briefly explained below.

[0059] Step 1: The monitoring unit monitors the user's conversation in real time. User conversations include telephone calls, video calls, and face-to-face conversations. The monitoring unit converts the conversation content into text data using voice recognition technology. In the case of video calls, the monitoring unit can also analyze video data and extract the conversation content. In the case of face-to-face conversations, the monitoring unit can also collect and analyze audio data using a microphone. Step 2: The analysis unit analyzes the conversation content monitored by the monitoring unit. The analysis is carried out using natural language processing technology and generative AI (e.g., LLM) to determine whether there is a possibility of special fraud. The analysis unit can also analyze audio and video data, extracting and analyzing important parts of the conversation content. Step 3: The warning issuing unit issues a warning based on the results of the analysis by the analysis unit. The warning may be issued in the form of an audio alert, a text message, a visual warning, or the like. For example, a voice alert may warn, "Possible fraud," and a text message may be sent to the user's smartphone. A visual warning may also be displayed on the user's device screen. Step 4: The information acquisition unit automatically obtains the latest fraud information provided by the police and related agencies. Information acquisition is often done using APIs, but it is also possible to obtain the latest fraud information from police databases or scrape fraud information from the websites of related agencies. It is also possible to collect fraud information from news articles and social media. Step 5: The learning unit learns the fraud information acquired by the information acquisition unit, and the analysis unit uses the results. Learning is performed using machine learning algorithms, learning patterns of fraudulent methods using supervised learning, unsupervised learning, and reinforcement learning, discovering new fraudulent methods, and improving the accuracy of fraud detection.

[0060] (Example 2) The fraud prevention system according to an embodiment of the present invention automatically monitors user conversations, analyzes the conversation content, and issues a warning if a fraud is suspected. This fraud prevention system effectively prevents frauds by eliminating user interaction and making decisions based on the latest fraud information. For example, the fraud prevention system monitors user conversations in real time. The AI ​​then analyzes the conversation content and determines whether a fraud is suspected. If a fraud is suspected, a warning is issued to the user. This warning is displayed via voice or text, allowing the user to immediately become aware of the possibility of fraud. Furthermore, the system constantly references the latest fraud information, and the AI ​​makes decisions based on that information. For example, the system automatically obtains the latest fraud information provided by the police and related organizations, and the AI ​​learns from that information. This allows the system to constantly respond to the latest fraud techniques. This system prevents frauds without the user having to manually operate the system. For example, if elderly people are at high risk of falling victim to telephone fraud, this system can automatically monitor conversations and issue warnings if there is a possibility of fraud, preventing fraud for elderly people. This system can also be used not only in homes, but also in businesses and public facilities. For example, introducing this system in a company's customer support center can prevent fraud before it occurs. Furthermore, its use in public facilities is expected to reduce fraud damage throughout the entire community. In this way, the special fraud prevention system using generative AI can effectively prevent special fraud damage because it can be used without user operation and makes judgments based on the latest fraud information. As a result, the special fraud prevention system automatically monitors user conversations, the generative AI analyzes the content of the conversation, and issues warnings if there is a possibility of special fraud, preventing fraud damage.

[0061] The special fraud prevention system according to the embodiment includes a monitoring unit, an analysis unit, a warning issuing unit, an information acquisition unit, and a learning unit. The monitoring unit monitors user conversations in real time. User conversations include, but are not limited to, telephone calls, video calls, and face-to-face conversations. The monitoring unit converts the conversation content into text data using, for example, voice recognition technology. In the case of video calls, the monitoring unit can also analyze video data and extract the conversation content. In the case of face-to-face conversations, the monitoring unit can also collect and analyze audio data using a microphone. The analysis unit analyzes the conversation content monitored by the monitoring unit using a generation AI. The analysis is performed using, for example, natural language processing technology, but is not limited to, an example. For example, the generation AI analyzes the conversation content using a text generation AI (e.g., LLM) to determine whether there is a possibility of special fraud. The analysis unit can also analyze audio and video data using a multimodal generation AI. In addition, the analysis unit can extract and analyze important parts of the conversation content using the generation AI. The warning issuing unit issues a warning based on the results of the analysis by the analysis unit. The warning may be issued in the form of, for example, an audio alert, a text message, a visual warning, or the like, but is not limited to these examples. For example, the warning issuing unit may warn the user "Possible fraud" as an audio alert. The warning issuing unit may also send a warning message to the user's smartphone as a text message. Furthermore, the warning issuing unit may also display a warning message on the user's device screen as a visual warning. The information acquisition unit automatically acquires the latest fraud information provided by the police and related organizations. Information acquisition may be performed, for example, using an API, but is not limited to these examples. For example, the information acquisition unit may acquire the latest fraud information from a police database. The information acquisition unit may also scrape fraud information from the websites of related organizations. Furthermore, the information acquisition unit may collect fraud information from news articles and social media. The learning unit learns the fraud information acquired by the information acquisition unit, and the analysis unit uses the results.Learning is performed using, for example, a machine learning algorithm, but is not limited to such an example. For example, the learning unit uses supervised learning to learn patterns of fraudulent methods. The learning unit can also use unsupervised learning to discover new fraudulent methods. Furthermore, the learning unit can also use reinforcement learning to improve the accuracy of detecting fraudulent methods. As a result, the special fraud prevention system according to the embodiment can prevent fraud by monitoring user conversations in real time and issuing warnings based on the analysis results.

[0062] The monitoring unit can estimate the user's emotions and adjust the monitoring frequency based on the estimated user emotions. For example, if the user is feeling stressed, the monitoring unit can increase the monitoring frequency to detect signs of fraud early. Furthermore, for example, if the user is relaxed, the monitoring unit can maintain the monitoring frequency at a normal level to avoid excessive monitoring. Furthermore, for example, if the user is in a hurry, the monitoring unit can temporarily reduce the monitoring frequency to reduce the user's burden. This allows early detection of signs of fraud by adjusting the monitoring frequency according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the monitoring unit can be performed using, for example, an AI. For example, the monitoring unit can input the user's voice data into the generation AI and have the generation AI perform emotion estimation.

[0063] During monitoring, the monitoring unit can select the optimal monitoring method by referring to the user's past conversation history. For example, the monitoring unit may focus on monitoring specific keywords based on conversation patterns in which the user has previously shown signs of fraud. The monitoring unit can also detect, for example, a high risk of fraud during a specific time period from the user's past conversation history and intensify monitoring during that time period. The monitoring unit can also analyze the user's past conversation history and prioritize monitoring conversations with specific parties. This allows the optimal monitoring method to be selected by referring to the user's past conversation history. The past conversation history is stored, for example, in a text database or a voice database. The optimal monitoring method is selected using algorithms such as voice analysis, text analysis, and behavioral analysis. Some or all of the above-described processing in the monitoring unit may be performed using, for example, AI, or without AI. For example, the monitoring unit can input past conversation history data into a generation AI and have the generation AI select the optimal monitoring method.

[0064] During monitoring, the monitoring unit may perform filtering based on the user's lifestyle or areas of interest. For example, if the user is elderly, the monitoring unit may focus on monitoring specific keywords that are considered to be at high risk of fraud. Furthermore, for example, if the user is interested in finance-related topics, the monitoring unit may prioritize monitoring conversations related to financial fraud. Furthermore, the monitoring unit may strengthen monitoring during specific time periods or situations depending on the user's lifestyle. By performing filtering based on the user's lifestyle and areas of interest, conversations with a high risk of fraud can be prioritized for monitoring. The lifestyle is acquired based on information such as residence, occupation, and daily behavior patterns. The areas of interest are acquired based on information such as hobbies, topics of interest, and past search history. Some or all of the above-described processing in the monitoring unit may be performed using, or without, AI. For example, the monitoring unit may input data on the user's lifestyle and areas of interest into a generation AI and have the generation AI perform filtering.

[0065] The monitoring unit can estimate the user's emotions and determine the priority of conversations to monitor based on the estimated emotions. For example, if the user is feeling anxious, the monitoring unit will prioritize monitoring conversations with a high risk of fraud. Similarly, if the user is relaxed, the monitoring unit can prioritize monitoring normal conversations. Furthermore, if the user is excited, the monitoring unit can prioritize monitoring conversations containing specific keywords. This allows for prioritizing conversations based on the user's emotions, thereby enabling the monitoring of conversations with a high risk of fraud. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the monitoring unit may be performed using AI, or not. For example, the monitoring unit can input the user's voice data into a generative AI and have the generative AI perform emotion estimation.

[0066] During monitoring, the monitoring unit can prioritize monitoring highly relevant conversations based on the user's geographical location information. For example, if the user is in a specific area, the monitoring unit prioritizes monitoring conversations related to fraud methods that are likely to occur in that area. Furthermore, for example, if the user is traveling, the monitoring unit can prioritize monitoring conversations that pose a high risk of fraud at the user's travel destination. Furthermore, for example, if the user is at home, the monitoring unit can prioritize monitoring conversations related to fraud methods that are likely to occur at home. This allows highly relevant conversations to be prioritized by taking the user's geographical location information into consideration. Geographical location information is acquired based on, for example, GPS data or an IP address. Highly relevant conversations are selected using criteria such as the frequency of occurrence of specific keywords or past fraud patterns. Some or all of the above-described processing by the monitoring unit may be performed using, for example, AI, or without AI. For example, the monitoring unit can input the user's geographical location data into a generation AI and cause the generation AI to select highly relevant conversations.

[0067] During monitoring, the monitoring unit can analyze the user's social media activity and monitor related conversations. For example, if the user posts about fraud on social media, the monitoring unit can prioritize monitoring conversations related to that content. Furthermore, for example, if the user shares information about a specific fraud, the monitoring unit can also monitor conversations related to that information. Furthermore, for example, the monitoring unit can prioritize monitoring conversations with people who are identified as being at high risk for fraud based on the user's social media activity. This allows for prioritized monitoring of related conversations by analyzing the user's social media activity. Social media activity is analyzed based on, for example, the content of posts, comments, the number of likes, and the like. Relevant conversations are selected using criteria such as conversations about a specific topic or conversations with a specific user. Some or all of the above-described processing by the monitoring unit may be performed using, for example, AI, or may be performed without AI. For example, the monitoring unit can input the user's social media data into a generation AI and cause the generation AI to select relevant conversations.

[0068] The analysis unit can estimate the user's emotions and adjust the accuracy of the analysis based on the estimated user emotions. For example, if the user is feeling anxious, the analysis unit can increase the accuracy of the analysis and detect signs of fraud in more detail. Furthermore, for example, if the user is relaxed, the analysis unit can maintain normal analysis accuracy and avoid excessive analysis. Furthermore, for example, if the user is in a hurry, the analysis unit can temporarily reduce the analysis accuracy and perform a quick analysis. This allows for more detailed detection of signs of fraud by adjusting the analysis accuracy based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the analysis unit can be performed using, for example, an AI. For example, the analysis unit can input the user's voice data into the generation AI and have the generation AI perform emotion estimation.

[0069] During analysis, the analysis unit can adjust the level of detail of the analysis based on the importance of the conversation. For example, if an important keyword is included in a conversation, the analysis unit analyzes that part in detail. Furthermore, for example, if the importance of the conversation is low, the analysis unit can perform a normal analysis to save resources. Furthermore, for example, if the importance of the conversation is high, the analysis unit can perform a detailed analysis to avoid missing signs of fraud. Thus, by adjusting the level of detail of the analysis based on the importance of the conversation, important conversations can be analyzed in detail. The importance of the conversation is evaluated using criteria such as risk assessment and impact assessment. The level of detail of the analysis is adjusted using criteria such as the depth of the analysis and the type of algorithm used. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input conversation data to a generation AI and have the generation AI perform a detailed analysis based on the importance.

[0070] During analysis, the analysis unit can apply different analysis algorithms depending on the category of the conversation. For example, in the case of a conversation related to financial fraud, the analysis unit can apply an analysis algorithm specialized for financial fraud. Furthermore, in the case of a conversation related to "I'm your son" fraud, the analysis unit can also apply an analysis algorithm specialized for "I'm your son" fraud. Furthermore, in the case of a conversation related to online fraud, the analysis unit can also apply an analysis algorithm specialized for online fraud. By applying different analysis algorithms depending on the category of the conversation, signs of fraud can be detected more accurately. The categories of conversations are classified using criteria such as business conversations, personal conversations, and fraud-related conversations. The analysis algorithm is applied using criteria such as natural language processing algorithms and machine learning algorithms. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or without AI. For example, the analysis unit can input conversation data into a generation AI and have the generation AI perform analysis according to the category.

[0071] The analysis unit can estimate the user's emotions and determine the priority of analysis based on the estimated emotions. For example, if the user is feeling anxious, the analysis unit may prioritize analyzing conversations that carry a high risk of fraud. Similarly, if the user is relaxed, the analysis unit may prioritize analyzing normal conversations. Furthermore, if the user is excited, the analysis unit may prioritize analyzing conversations containing specific keywords. This allows for prioritizing analysis based on the user's emotions, thereby prioritizing the analysis of conversations with a high risk of fraud. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the analysis unit may be performed using AI, or not. For example, the analysis unit can input the user's voice data into a generative AI and have the generative AI perform emotion estimation.

[0072] During analysis, the analysis unit can determine the analysis priority based on the time of submission of the conversation. For example, the analysis unit may prioritize analyzing recent conversations to address the latest fraudulent methods. The analysis unit can also analyze past conversations to check for signs of fraud. For example, the analysis unit can prioritize analyzing conversations submitted during a specific time period to evaluate the fraud risk during that time period. This allows the latest fraudulent methods to be addressed by determining the analysis priority based on the time of submission of the conversation. The time of submission of the conversation is obtained based on, for example, a timestamp or the date and time of submission. The analysis priority is determined using criteria such as risk assessment and importance assessment. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input conversation data into a generation AI and have the generation AI determine the priority based on the time of submission.

[0073] During analysis, the analysis unit can adjust the analysis order based on the relevance of the conversations. For example, the analysis unit prioritizes analysis of conversations that are deemed to have a high risk of fraud. The analysis unit can also prioritize analysis of important conversations, for example, by postponing analysis of less relevant conversations. The analysis unit can also evaluate the relevance of conversations and prioritize analysis of the most relevant conversations. This allows important conversations to be prioritized by adjusting the analysis order based on the relevance of the conversations. The relevance of conversations is evaluated using criteria such as common keywords or related topics. The analysis order is adjusted using criteria such as analyzing conversations with higher priority. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input conversation data to a generation AI and have the generation AI execute an analysis order based on the relevance.

[0074] The warning issuing unit can estimate the user's emotion and adjust the manner in which the warning is expressed based on the estimated user's emotion. For example, if the user is nervous, the warning issuing unit can issue the warning in a calm manner. For example, if the user is relaxed, the warning issuing unit can issue the warning in a normal manner. For example, if the user is excited, the warning issuing unit can issue the warning in an emphasized manner. By adjusting the manner in which the warning is expressed based on the user's emotion, an appropriate warning can be issued to the user. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the warning issuing unit can be performed using, for example, an AI, or can be performed without using an AI. For example, the warning issuing unit can input the user's voice data to the generation AI and cause the generation AI to estimate the emotion.

[0075] The warning issuing unit can adjust the level of detail of the warning based on the importance of the conversation when issuing a warning. For example, the warning issuing unit issues a detailed warning for an important conversation. The warning issuing unit can also issue a brief warning for a conversation with low importance. The warning issuing unit can also issue a warning with an appropriate level of detail for a conversation with medium importance. Thus, by adjusting the level of detail of the warning based on the importance of the conversation, detailed warnings can be issued for important conversations. The importance of the conversation is evaluated using criteria such as risk assessment and impact assessment. The level of detail of the warning is adjusted using criteria such as a detailed explanation or a brief warning. Some or all of the above-described processing in the warning issuing unit may be performed using, for example, AI, or may be performed without using AI. For example, the warning issuing unit can input conversation data to a generation AI and cause the generation AI to issue detailed warnings based on the importance.

[0076] When issuing a warning, the warning issuing unit can apply different warning algorithms depending on the category of the conversation. For example, in the case of a conversation related to financial fraud, the warning issuing unit can apply a warning algorithm specialized for financial fraud. Furthermore, in the case of a conversation related to "I'm your son" fraud, the warning issuing unit can also apply a warning algorithm specialized for "I'm your son" fraud. Furthermore, in the case of a conversation related to online fraud, the warning issuing unit can also apply a warning algorithm specialized for online fraud. In this way, by applying different warning algorithms depending on the category of the conversation, appropriate warnings can be issued for conversations with a high risk of fraud. Conversation categories are classified using criteria such as business conversations, personal conversations, and fraud-related conversations. The warning algorithm is applied using criteria such as a risk assessment algorithm and an impact assessment algorithm. Some or all of the above-mentioned processing in the warning issuing unit may be performed using, for example, AI, or may be performed without using AI. For example, the warning issuing unit can input conversation data to a generation AI and cause the generation AI to execute a warning according to the category.

[0077] The warning issuing unit can estimate the user's emotion and adjust the length of the attention based on the estimated user's emotion. For example, if the user is nervous, the warning issuing unit can issue a short and to-the-point warning. For example, if the user is relaxed, the warning issuing unit can issue a detailed warning. For example, if the user is in a hurry, the warning issuing unit can issue a quick and concise warning. By adjusting the length of the attention based on the user's emotion, an appropriate length of attention can be issued to the user. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the warning issuing unit can be performed using, for example, an AI, or without an AI. For example, the warning issuing unit can input the user's voice data to the generation AI and cause the generation AI to estimate the emotion.

[0078] The alert issuing unit can determine the priority of alerts based on when the conversation was submitted. For example, the alert issuing unit can prioritize alerts for recent conversations. It can also issue alerts for past conversations to check for signs of fraud. Furthermore, the alert issuing unit can prioritize alerts for conversations submitted within a specific time period. This allows for a response to the latest fraud techniques by prioritizing alerts based on when the conversation was submitted. The conversation submission date is obtained based on, for example, a timestamp or submission date and time. The priority of alerts is determined using criteria such as risk assessment or importance assessment. Some or all of the above processing in the alert issuing unit may be performed using, for example, AI, or not. For example, the alert issuing unit can input conversation data into a generating AI and have the generating AI perform the priority determination based on the submission date.

[0079] The attention-issuing unit can adjust the order of attention based on the relevance of the conversations when issuing an attention. For example, the attention-issuing unit can prioritize issuing attention to conversations that are considered to have a high risk of fraud. The attention-issuing unit can also, for example, postpone issuing attention to less relevant conversations and issue attention first to important conversations. Furthermore, the attention-issuing unit can, for example, evaluate the relevance of conversations and prioritize issuing attention to the most relevant conversations. In this way, by adjusting the order of attention based on the relevance of conversations, it is possible to prioritize issuing attention to important conversations. The relevance of conversations is evaluated using criteria such as common keywords or related topics. The adjustment of the order of attention is performed using criteria such as issuing attention to high-priority conversations first. Some or all of the above processing in the attention-issuing unit may be performed using AI, for example, or not using AI. For example, the attention-issuing unit can input conversation data into a generating AI and have the generating AI execute the relevance-based attention order.

[0080] The information acquisition unit can estimate the user's emotions and adjust the frequency of information acquisition based on the estimated emotions. For example, if the user is feeling anxious, the information acquisition unit can increase the frequency of information acquisition to quickly obtain the latest fraud information. Also, if the user is relaxed, for example, the information acquisition unit can maintain the normal frequency of information acquisition and avoid excessive information acquisition. Furthermore, if the user is in a hurry, for example, the information acquisition unit can temporarily decrease the frequency of information acquisition to reduce the user's burden. In this way, by adjusting the frequency of information acquisition based on the user's emotions, the latest fraud information can be quickly obtained. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or a generative AI. The generative AI is a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above processing in the information acquisition unit may be performed using AI, for example, or without AI. For example, the information acquisition unit can input the user's voice data into the generative AI and have the generative AI perform emotion estimation.

[0081] The information acquisition unit can select the optimal information acquisition method by referring to past fraud information when acquiring information. For example, the information acquisition unit can prioritize acquiring information containing specific keywords based on past fraud information. The information acquisition unit can also prioritize acquiring information related to fraud methods that are likely to occur during specific time periods from past fraud information. Furthermore, the information acquisition unit can analyze past fraud information and prioritize acquiring the most relevant information. This allows the optimal information acquisition method to be selected by referring to past fraud information. Past fraud information is stored, for example, in a database or archive. The selection of the optimal information acquisition method is performed using criteria such as web scraping or API usage. Some or all of the above processing in the information acquisition unit may be performed using AI, for example, or without AI. For example, the information acquisition unit can input past fraud information data into a generating AI and have the generating AI select the optimal information acquisition method.

[0082] The information acquisition unit can apply different information acquisition algorithms depending on the category of fraud information when acquiring information. For example, in the case of information related to financial fraud, the information acquisition unit can apply an information acquisition algorithm specialized for financial fraud. Also, in the case of information related to "ore-ore" fraud (a type of telephone fraud), the information acquisition unit can apply an information acquisition algorithm specialized for "ore-ore" fraud. Also, in the case of information related to online fraud, the information acquisition unit can apply an information acquisition algorithm specialized for online fraud. By applying different information acquisition algorithms depending on the category of fraud information, it is possible to prioritize the acquisition of information with a high risk of fraud. Fraud information categories are classified using criteria such as phishing scams, telephone scams, and email scams. The application of information acquisition algorithms is carried out using criteria such as natural language processing algorithms and machine learning algorithms. Some or all of the above processing in the information acquisition unit may be performed using AI, for example, or without AI. For example, the information acquisition unit can input fraud information data into a generating AI and have the generating AI perform information acquisition according to the category.

[0083] The information acquisition unit can estimate the user's emotions and determine the priority of information acquisition based on the estimated user emotions. For example, if the user is feeling anxious, the information acquisition unit can prioritize information with a high risk of fraud. Furthermore, for example, if the user is relaxed, the information acquisition unit can prioritize information that is normal. Furthermore, for example, if the user is excited, the information acquisition unit can prioritize information that contains specific keywords. Thus, by determining the priority of information acquisition based on the user's emotions, it is possible to prioritize information with a high risk of fraud. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the information acquisition unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the information acquisition unit can input the user's voice data into the generation AI and cause the generation AI to estimate the emotion.

[0084] When acquiring information, the information acquisition unit can determine the priority of information acquisition based on the time of submission of the fraud information. The information acquisition unit, for example, prioritizes the acquisition of recent fraud information to respond to the latest fraud methods. The information acquisition unit can also, for example, acquire past fraud information and check whether there were any signs of fraud. The information acquisition unit can also, for example, prioritize the acquisition of fraud information submitted during a specific time period and evaluate the fraud risk during that time period. This allows the latest fraud methods to be responded to by determining the priority of information acquisition based on the time of submission of the fraud information. The time of submission of the fraud information is acquired based on, for example, a timestamp or the date and time of submission. The priority of information acquisition is determined using criteria such as risk assessment and importance assessment. Some or all of the above-mentioned processing in the information acquisition unit may be performed using, for example, AI, or may be performed without AI. For example, the information acquisition unit can input fraud information data to a generation AI and cause the generation AI to determine the priority based on the time of submission.

[0085] When acquiring information, the information acquisition unit can adjust the order of information acquisition based on the relevance of the fraud information. For example, the information acquisition unit prioritizes acquisition of information that is deemed to have a high risk of fraud. The information acquisition unit can also, for example, postpone acquisition of less relevant information and acquire important information first. The information acquisition unit can also, for example, evaluate the relevance of the fraud information and prioritize acquisition of the most relevant information. In this way, important information can be prioritized by adjusting the order of information acquisition based on the relevance of the fraud information. The relevance of the fraud information is evaluated using criteria such as common keywords or related topics. The order of information acquisition is adjusted using criteria such as a method of acquiring information first with a higher priority. Some or all of the above-described processing in the information acquisition unit may be performed using, for example, AI, or may be performed without AI. For example, the information acquisition unit can input fraud information data to a generation AI and cause the generation AI to execute an information acquisition order based on relevance.

[0086] The learning unit can estimate the user's emotions and select training data based on the estimated user emotions. For example, if the user is feeling anxious, the learning unit prioritizes learning data with a high risk of fraud. Furthermore, for example, if the user is relaxed, the learning unit can prioritize learning normal data. Furthermore, for example, if the user is excited, the learning unit can prioritize learning data containing specific keywords. Thus, by selecting training data based on the user's emotions, data with a high risk of fraud can be prioritized. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the learning unit may be performed using, for example, an AI, or may be performed without using an AI. For example, the learning unit can input the user's voice data into the generation AI and cause the generation AI to estimate emotions.

[0087] The learning unit can adjust its learning algorithm by referring to past learning data during the learning process. For example, the learning unit can apply a learning algorithm specialized for a particular fraud scheme based on past learning data. The learning unit can also select the most effective learning algorithm from past learning data. Furthermore, the learning unit can analyze past learning data to improve the accuracy of the learning algorithm. This allows the optimal learning algorithm to be applied by referring to past learning data. Past learning data is stored, for example, in a database or archive. The application of the learning algorithm is performed using criteria such as supervised learning, unsupervised learning, or reinforcement learning. Some or all of the above processes in the learning unit may be performed using AI, for example, or without AI. For example, the learning unit can input past learning data into a generating AI and have the generating AI apply the optimal learning algorithm.

[0088] The learning unit can apply different learning algorithms during learning depending on the category of fraud information. For example, if the data is related to financial fraud, the learning unit will apply a learning algorithm specialized for financial fraud. Similarly, if the data is related to "ore-ore" fraud (a type of phone scam), the learning unit can apply a learning algorithm specialized for that type of fraud. Furthermore, if the data is related to online fraud, the learning unit can apply a learning algorithm specialized for online fraud. This allows the learning unit to prioritize learning information with a high risk of fraud by applying different learning algorithms depending on the category of fraud information. Fraud information categories are classified using criteria such as phishing scams, phone scams, and email scams. The application of learning algorithms is performed using criteria such as natural language processing algorithms and machine learning algorithms. Some or all of the above processing in the learning unit may be performed using AI, or not. For example, the learning unit can input fraud information data into a generating AI and have the generating AI perform learning according to the categories.

[0089] The learning unit can estimate the user's emotions and adjust the learning frequency based on the estimated user emotions. For example, if the user feels anxious, the learning unit increases the learning frequency to quickly learn the latest scam information. Furthermore, for example, if the user feels relaxed, the learning unit can maintain a normal learning frequency to avoid excessive learning. Furthermore, for example, if the user is in a hurry, the learning unit can temporarily reduce the learning frequency to reduce the user's burden. This allows the latest scam information to be quickly learned by adjusting the learning frequency based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the learning unit may be performed using, for example, an AI. For example, the learning unit may input the user's voice data into the generation AI and cause the generation AI to estimate emotions.

[0090] During learning, the learning unit can weight the learning data based on the time when the fraud information was submitted. For example, the learning unit may prioritize learning recent fraud information to respond to the latest fraud methods. The learning unit can also, for example, learn past fraud information and check whether there were any signs of fraud. The learning unit can also, for example, prioritize learning fraud information submitted during a specific time period to evaluate the fraud risk during that time period. This allows the latest fraud methods to be responded to by weighting the learning data based on the time when the fraud information was submitted. The time when the fraud information was submitted is obtained based on, for example, a timestamp or the date and time of submission. The weighting of the learning data is performed using criteria such as a method of assigning a higher weight to the most recent data. Some or all of the above-described processing in the learning unit may be performed using, for example, AI, or may be performed without AI. For example, the learning unit can input fraud information data into a generation AI and cause the generation AI to perform weighting based on the time of submission.

[0091] During learning, the learning unit can adjust the order of learning data based on the relevance of fraud information. For example, the learning unit prioritizes learning information that is deemed to have a high risk of fraud. The learning unit can also, for example, postpone learning information with low relevance and learn important information first. The learning unit can also, for example, evaluate the relevance of fraud information and prioritize learning the most relevant information. In this way, important information can be prioritized by adjusting the order of learning data based on the relevance of fraud information. The relevance of fraud information is evaluated using criteria such as common keywords or related topics. The order of learning data is adjusted using criteria such as a method of learning from high-priority data. Some or all of the above-described processing in the learning unit may be performed using, for example, AI, or may be performed without AI. For example, the learning unit can input fraud information data into a generation AI and have the generation AI execute the order of learning data based on the relevance. === Hard Collateral 1-1 === Each of the multiple elements described above, including the monitoring unit, analysis unit, warning issuance unit, information acquisition unit, and learning unit, is implemented in at least one of the smart device 14 and the data processing unit 12. For example, the monitoring unit monitors the user's conversation in real time using the microphone 38B and camera 42 of the smart device 14. The analysis unit is implemented, for example, by the identification processing unit 290 of the data processing unit 12 and analyzes the conversation content using a generation AI. The warning issuance unit issues voice alerts or text messages using, for example, the output device 40 of the smart device 14. The information acquisition unit acquires the latest fraud information using, for example, the communication I / F 26 of the data processing unit 12. The learning unit is implemented, for example, by the identification processing unit 290 of the data processing unit 12 and learns the acquired fraud information. The monitoring unit can, for example, estimate the user's emotions and adjust the monitoring frequency based on the estimated emotions. === Hard Collateral 1-2 === Each of the multiple elements, including the monitoring unit, analysis unit, warning unit, information acquisition unit, and learning unit, described above, is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the monitoring unit monitors a user's conversation in real time using the microphone 238 or camera 42 of the smart glasses 214. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes the conversation content using a generation AI. The warning unit issues a voice alert or a text message using, for example, the speaker 240 of the smart glasses 214. The information acquisition unit acquires the latest fraud information using, for example, the communication I / F 26 of the data processing device 12. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and learns the acquired fraud information. The monitoring unit can, for example, estimate a user's emotions and adjust the monitoring frequency based on the estimated emotions. === Hard Collateral 1-3 === Each of the multiple elements, including the monitoring unit, analysis unit, warning unit, information acquisition unit, and learning unit, described above, is realized, for example, by at least one of the headset-type terminal 314 and the data processing device 12. For example, the monitoring unit monitors a user's conversation in real time using the microphone 238 or camera 42 of the headset-type terminal 314. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes the conversation content using a generation AI. The warning unit issues a voice alert or a text message using, for example, the speaker 240 of the headset-type terminal 314. The information acquisition unit acquires the latest fraud information using, for example, the communication I / F 26 of the data processing device 12. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and learns the acquired fraud information. The monitoring unit can, for example, estimate a user's emotion and adjust the monitoring frequency based on the estimated emotion. === Hard Collateral 1-4 === Each of the multiple elements, including the monitoring unit, analysis unit, warning unit, information acquisition unit, and learning unit, described above, is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the monitoring unit monitors a user's conversation in real time using the microphone 238 or camera 42 of the robot 414. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes the conversation content using a generation AI. The warning unit issues a voice alert or a text message using, for example, the speaker 240 of the robot 414. The information acquisition unit acquires the latest fraud information using, for example, the communication I / F 26 of the data processing device 12. The learning unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and learns the acquired fraud information. The monitoring unit can, for example, estimate a user's emotion and adjust the monitoring frequency based on the estimated emotion.

[0092] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0093] The monitoring unit can estimate the user's health condition and adjust the intensity of monitoring based on the estimated health condition. For example, if the user is in poor health, the monitoring intensity can be increased to detect signs of fraud early. Also, if the user is healthy, the monitoring intensity can be maintained at a normal level to avoid excessive monitoring. Furthermore, if the user is feeling stressed, the monitoring intensity can be temporarily reduced to reduce the burden on the user. In this way, by adjusting the monitoring intensity according to the user's health condition, signs of fraud can be detected early.

[0094] The monitoring unit can estimate the user's emotions and adjust the monitoring frequency based on the estimated user emotions. For example, if the user is feeling stressed, the monitoring frequency can be increased to detect signs of fraud early. Alternatively, if the user is relaxed, the monitoring frequency can be kept normal to avoid excessive monitoring. Alternatively, if the user is in a hurry, the monitoring frequency can be temporarily reduced to reduce the user's burden. In this way, by adjusting the monitoring frequency according to the user's emotions, signs of fraud can be detected early.

[0095] The monitoring unit can select the optimal monitoring method by referring to the user's past behavioral patterns. For example, it can focus on monitoring specific behaviors based on behavioral patterns in which the user has shown signs of fraud in the past. It can also detect a high risk of fraud during specific time periods from the user's past behavioral patterns and intensify monitoring during those times. It can also analyze the user's past behavioral patterns and prioritize monitoring of interactions with specific parties. This makes it possible to select the optimal monitoring method by referring to the user's past behavioral patterns.

[0096] The monitoring unit can perform filtering based on the user's lifestyle or areas of interest. For example, if the user is elderly, it can focus on monitoring specific keywords that are considered to be at high risk of fraud. Also, if the user is interested in finance-related topics, it can prioritize monitoring of conversations related to financial fraud. Furthermore, it can strengthen monitoring during specific time periods or situations depending on the user's lifestyle. In this way, by filtering based on the user's lifestyle and areas of interest, it is possible to focus on monitoring conversations that pose a high risk of fraud.

[0097] The monitoring unit can estimate the user's emotions and determine the priority of conversations to be monitored based on the estimated user's emotions. For example, if the user is feeling anxious, conversations with a high risk of fraud can be monitored with priority. Also, if the user is relaxed, normal conversations can be monitored with priority. Furthermore, if the user is excited, conversations containing specific keywords can be monitored with priority. In this way, by determining the priority of conversations to be monitored based on the user's emotions, conversations with a high risk of fraud can be monitored with priority.

[0098] The monitoring unit can prioritize monitoring conversations that are highly relevant based on the user's geographical location. For example, if a user is in a specific region, it can prioritize monitoring conversations related to fraud schemes that are likely to occur in that region. Similarly, if a user is traveling, it can prioritize monitoring conversations related to fraud risks in their travel destination. Furthermore, if a user is at home, it can prioritize monitoring conversations related to fraud schemes that are likely to occur within the home. In this way, by considering the user's geographical location, it is possible to prioritize monitoring conversations that are highly relevant.

[0099] The monitoring unit can analyze users' social media activity and monitor relevant conversations. For example, if a user posts about fraud on social media, it can prioritize monitoring conversations related to that content. It can also monitor conversations related to information shared by a user about a specific type of fraud. Furthermore, it can prioritize monitoring conversations with individuals deemed to pose a high risk of fraud based on the user's social media activity. In this way, by analyzing a user's social media activity, it can prioritize monitoring of relevant conversations.

[0100] The analysis unit can estimate the user's emotions and adjust the accuracy of the analysis based on those emotions. For example, if the user is feeling anxious, the accuracy of the analysis can be increased to detect signs of fraud in more detail. Conversely, if the user is relaxed, the normal accuracy of the analysis can be maintained, avoiding excessive analysis. Furthermore, if the user is in a hurry, the accuracy of the analysis can be temporarily reduced to perform a quick analysis. In this way, by adjusting the accuracy of the analysis based on the user's emotions, signs of fraud can be detected in more detail.

[0101] The analysis unit can adjust the level of detail in its analysis based on the importance of the conversation. For example, if a conversation contains important keywords, it will analyze that section in detail. Conversely, if the conversation is of low importance, it can perform a standard analysis to conserve resources. Furthermore, if the conversation is of high importance, it can perform a detailed analysis to ensure that no signs of fraud are missed. In this way, by adjusting the level of detail in the analysis based on the importance of the conversation, important conversations can be analyzed in detail.

[0102] The analysis unit can apply different analysis algorithms depending on the category of the conversation. For example, in the case of a conversation related to financial fraud, an analysis algorithm specialized for financial fraud can be applied. In addition, in the case of a conversation related to "it's my son" fraud, an analysis algorithm specialized for "it's my son" fraud can be applied. Furthermore, in the case of a conversation related to online fraud, an analysis algorithm specialized for online fraud can be applied. In this way, by applying different analysis algorithms depending on the category of the conversation, it is possible to more accurately detect signs of fraud.

[0103] The processing flow of the second embodiment will be briefly explained below.

[0104] Step 1: The monitoring unit monitors the user's conversation in real time. User conversations include telephone calls, video calls, and face-to-face conversations. The monitoring unit converts the conversation content into text data using voice recognition technology. In the case of video calls, the monitoring unit can also analyze video data and extract the conversation content. In the case of face-to-face conversations, the monitoring unit can also collect and analyze audio data using a microphone. Step 2: The analysis unit analyzes the conversation content monitored by the monitoring unit. The analysis is carried out using natural language processing technology and generative AI (e.g., LLM) to determine whether there is a possibility of special fraud. The analysis unit can also analyze audio and video data, extracting and analyzing important parts of the conversation content. Step 3: The warning issuing unit issues a warning based on the results of the analysis by the analysis unit. The warning may be issued in the form of an audio alert, a text message, a visual warning, or the like. For example, a voice alert may warn, "Possible fraud," and a text message may be sent to the user's smartphone. A visual warning may also be displayed on the user's device screen. Step 4: The information acquisition unit automatically obtains the latest fraud information provided by the police and related agencies. Information acquisition is often done using APIs, but it is also possible to obtain the latest fraud information from police databases or scrape fraud information from the websites of related agencies. It is also possible to collect fraud information from news articles and social media. Step 5: The learning unit learns the fraud information acquired by the information acquisition unit, and the analysis unit uses the results. Learning is performed using machine learning algorithms, learning patterns of fraudulent methods using supervised learning, unsupervised learning, and reinforcement learning, discovering new fraudulent methods, and improving the accuracy of fraud detection.

[0105] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0106] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of the generative AI include a neural network (NN) and a neural network (NN). The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, image data, etc. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and may perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-mentioned parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. The processing performed by an AI including the generative AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI including the generative AI.

[0107] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0108] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0109] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0110] 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.

[0111] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0112] 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.

[0113] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0114] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0115] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0116] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0117] 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.

[0118] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0119] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0120] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0121] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0122] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0123] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0124] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0125] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0126] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0127] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0128] 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.

[0129] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0130] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0131] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0132] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset 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.

[0133] 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.

[0134] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0135] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0136] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0137] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0138] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0139] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0140] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0141] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0142] 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.

[0143] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0144] 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.

[0145] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0146] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0147] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0148] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0149] 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.

[0150] 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.

[0151] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0152] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0153] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0154] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0155] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt including an instruction, as well as inference data such as audio data indicating speech, text data indicating text, and image data indicating an image (e.g., still image data or video data). The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in one or more data formats, such as audio data, text data, and image data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation models 58 include AIs other than the generation AI. Examples of AIs other than the generation AI include, but are not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and naive Bayes. These AIs can perform various types of processing, but are not limited to these examples. The AI ​​may also be an AI agent. When the processing of each of the above-described parts is performed by an AI, the processing may be performed in part or entirely by the AI, but is not limited to these examples. Processing performed by an AI, including the generation AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by an AI, including the generation AI.

[0156] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0157] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0158] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0159] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0160] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0161] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0162] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0163] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[0164] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[0165] 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.

[0166] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0167] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0168] 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.

[0169] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. A processor also includes a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0170] The hardware resource that executes the specific process may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0171] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0172] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0173] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0174] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0175] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0176] [Explanation of symbols]

[0177] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. a monitoring unit that monitors user conversations in real time; an analysis unit that analyzes the conversation content monitored by the monitoring unit; a warning issuing unit that issues a warning based on the result of the analysis by the analysis unit; an information acquisition unit that acquires fraud information; a learning unit that learns the fraud information acquired by the information acquisition unit; A system characterized by:

2. The monitoring unit Estimate the user's emotions and adjust the monitoring frequency based on the estimated user emotions.

2. The system of claim 1.

3. The monitoring unit When monitoring, refer to the user's past conversation history to determine the monitoring method.

2. The system of claim 1.

4. The monitoring unit When monitoring, filtering based on the user's life situation or areas of interest 2. The system of claim 1.

5. The monitoring unit Estimate the user's emotions and prioritize conversations to monitor based on the estimated user emotions.

2. The system of claim 1.

6. The monitoring unit When monitoring, prioritize relevant conversations based on the user's geographic location 2. The system of claim 1.

7. The monitoring unit During monitoring, analyze users' social media activity and monitor related conversations 2. The system of claim 1.

8. The analysis unit Estimate the user's emotions and adjust the accuracy of the analysis based on the estimated user emotions.

2. The system of claim 1.

Images