system

Abstract

The system according to this embodiment aims to protect children and the elderly with low IT literacy from online dangers. [Solution] The system according to the embodiment comprises an analysis unit, a prediction unit, a sharing unit, and a disconnection unit. The analysis unit analyzes the content of a conversation. The prediction unit predicts danger based on the content of the conversation analyzed by the analysis unit. The sharing unit shares the danger predicted by the prediction unit with family members who have been registered in advance. The disconnection unit forcibly disconnects the conversation based on the information shared by the sharing unit.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, there is a problem that children and the elderly with low IT literacy may be victims of fraud on the Internet, and there is a lack of effective means to prevent this.

[0005] The system according to the embodiment aims to protect children and the elderly with low IT literacy from risks on the Internet.

Means for Solving the Problems

[0006] The system according to this embodiment comprises an analysis unit, a prediction unit, a sharing unit, and a disconnection unit. The analysis unit analyzes the content of a conversation. The prediction unit predicts danger based on the content of the conversation analyzed by the analysis unit. The sharing unit shares the danger predicted by the prediction unit with family members who have been registered in advance. The disconnection unit forcibly disconnects the conversation based on the information shared by the sharing unit. [Effects of the Invention]

[0007] The system according to this embodiment can protect children and the elderly with low IT literacy from online dangers. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

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

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

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

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

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

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

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

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

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

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

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

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

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

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

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

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

[0028] (Example of form 1) The monitoring dog system according to an embodiment of the present invention is a system that detects and protects children and the elderly with low IT literacy from dangerous situations. This monitoring dog system allows the user to have the AI ​​analyze the content of messaging apps (e.g., LINE®) or phone calls, and the AI ​​predicts dangers such as wire fraud or involvement in criminal activity by minors, and shares this information with pre-registered family members. This mechanism allows family members to detect dangerous situations early and take appropriate action. For example, the user has the AI ​​analyze the content of messaging apps or phone calls. In this process, the AI ​​analyzes the conversation content in detail and identifies dangerous words and contexts. For example, if dangerous words such as "wire transfer" or "illegal part-time job" are included, the AI ​​will judge the conversation as dangerous. In addition, accounts with a history of being reported by other users and actions such as starting a video call immediately after adding an account are also factors that the AI ​​will judge as dangerous. Next, if the AI ​​determines that something is dangerous, it will share that information with pre-registered family members. Specifically, it will send screenshots or audio data of the conversation content to the family. In addition, the dangerous conversation content will be displayed on the messaging app's timeline. This allows family members to detect dangerous situations early and take appropriate action. Furthermore, if a family member determines that an exchange is inappropriate, they can forcibly terminate the conversation. This feature allows families to respond quickly and prevent harm before it occurs. This service is offered as a paid service for 100 yen per month. Users can gain a sense of security from having a "watchdog" (a Shiba Inu + police motif) watching over them. Also, because the monitoring is done by AI, the feeling of being spied on is diminished, reducing psychological burden. This system can create a society where children and the elderly can use IT with peace of mind. In addition, since criminals are immediately reported and punished, it can create an environment where crime is less likely to occur. In this way, the watchdog system can create a society where children and the elderly can use IT with peace of mind.

[0029] The monitoring dog system according to the embodiment comprises an analysis unit, a prediction unit, a sharing unit, and a disconnection unit. The analysis unit analyzes the content of conversations. The analysis unit analyzes the content of conversations using, for example, natural language processing technology. The analysis unit can also analyze the content of conversations using, for example, sentiment analysis technology. The analysis unit can analyze the content of conversations such as text chats, voice calls, and video calls. The prediction unit predicts dangers based on the conversation content analyzed by the analysis unit. The prediction unit predicts dangers such as wire fraud and involvement of minors in criminal activities. The prediction unit can determine dangers such as accounts with a history of being reported by other users or actions such as starting a video call immediately after adding an account. The prediction unit can identify dangerous words and contexts and predict dangers. The sharing unit shares the dangers predicted by the prediction unit with family members who have been registered in advance. The sharing unit can send, for example, screenshots or audio data of conversations to family members. The sharing unit can display dangerous conversation content on the timeline of a messaging app. The sharing unit can, for example, provide a "monitoring dog" themed after a Shiba Inu and the police. The disconnection unit forcibly disconnects the conversation based on the information shared by the sharing unit. The disconnection unit can forcibly disconnect the conversation by, for example, ending the call or temporarily suspending the account. In this way, the monitoring dog system according to the embodiment can protect users with low IT literacy by analyzing the content of the conversation, predicting danger and sharing it with the family, and forcibly disconnecting the conversation as necessary.

[0030] The analysis unit analyzes the content of conversations. For example, the analysis unit analyzes the content of conversations using natural language processing technology. Natural language processing technology is a technology for understanding text data and extracting meaning, and includes processes such as morphological analysis, syntactic analysis, and semantic analysis. The analysis unit can also analyze the content of conversations using sentiment analysis technology. Sentiment analysis technology is a technology that detects emotions from text and audio data and classifies them into sentiment categories such as positive, negative, and neutral. The analysis unit can analyze the content of conversations such as text chats, voice calls, and video calls. In the case of text chats, the analysis unit obtains chat logs and analyzes the content using natural language processing technology. In the case of voice calls, the analysis unit uses speech recognition technology to convert audio data into text, and then applies natural language processing technology. In the case of video calls, in addition to analyzing audio data, it is also possible to analyze facial expressions and gestures from video data. This allows the analysis unit to analyze various forms of conversation content from multiple angles and accurately grasp the user's intentions and emotions. Furthermore, the analysis unit can refer to past conversation data and user behavior history to perform more accurate analysis. For example, by analyzing the past conversations of a particular user and learning those patterns, the accuracy of analyzing the current conversation content can be improved. This allows the analysis unit to analyze conversation content in real time and provide a foundation for ensuring user safety.

[0031] The prediction unit predicts risks based on the conversation content analyzed by the analysis unit. For example, the prediction unit predicts risks such as wire fraud and involvement of minors in criminal activity. In the case of wire fraud, the prediction unit detects specific keywords and phrases and determines whether they are signs of fraud. For example, if phrases such as "Please transfer the money" or "Please tell me your bank account details" are included, the prediction unit will determine it to be dangerous. In the case of involvement of minors in criminal activity, the prediction unit considers the user's age and behavioral patterns to identify dangerous situations. For example, if a user frequently makes video calls late at night or interacts with accounts that have been reported by other users, the prediction unit may determine it to be dangerous. The prediction unit can determine, for example, that accounts with a history of being reported by other users or actions such as starting a video call immediately after adding an account are dangerous. This allows the prediction unit to comprehensively analyze the user's behavioral patterns and conversation content and detect potential risks early. Furthermore, the prediction unit can identify dangerous words and contexts and predict risks. For example, if phrases such as "Keep it a secret" or "Don't tell anyone" are included, it may indicate a dangerous situation. The prediction unit detects these phrases and predicts potential dangers. This allows the prediction unit to play a crucial role in ensuring user safety and preventing potential hazards before they occur.

[0032] The sharing function shares the dangers predicted by the prediction function with pre-registered family members. For example, the sharing function can send screenshots or audio data of conversations to family members. This allows family members to understand dangerous situations in real time and take appropriate action. The sharing function can, for example, display dangerous conversation content on the LINE timeline. This allows family and friends to share dangerous situations and cooperate in responding. The sharing function can, for example, provide a "watchdog" character based on a Shiba Inu and police. The "watchdog" is a character that visually represents dangerous situations and serves to warn the user. For example, the "watchdog" may appear on the screen and display icons or messages indicating dangerous conversation content. This allows users to intuitively recognize danger and take appropriate action. Furthermore, the sharing function can promote communication with family and friends and raise awareness of dangerous situations. For example, the sharing function can regularly notify family members about dangerous conversation content so that family members can constantly monitor the user's safety. In addition, the sharing function can quickly send notifications to family and friends when a dangerous situation occurs, encouraging early action. This means that shared areas play a crucial role in ensuring user safety and can strengthen connections with family and friends.

[0033] The disconnection unit forcibly terminates conversations based on information shared by the sharing unit. The disconnection unit can forcibly terminate conversations in ways such as ending a call or temporarily suspending an account. In the case of ending a call, the disconnection unit immediately terminates the call as soon as dangerous conversation content is detected. This quickly frees the user from a dangerous situation. In the case of temporarily suspending an account, the disconnection unit temporarily blocks contact with the dangerous account, ensuring the user's safety. For example, it temporarily blocks messages and calls from the dangerous account, preventing the user from making further contact. This allows the disconnection unit to provide a quick and effective response to ensure the user's safety. Furthermore, the disconnection unit can notify family and friends that the conversation has been forcibly terminated and share the situation. This allows family and friends to recognize that the user is facing a dangerous situation and take appropriate action. The disconnection unit can also warn the user to prevent dangerous situations from recurring. For example, if dangerous conversation content is detected, it can display a warning message to the user, indicating points to be careful about in the future. In this way, the disconnection unit plays a vital role in ensuring the user's safety and can quickly free them from dangerous situations.

[0034] The shared unit can send screenshots and audio data of conversations to family members. For example, the shared unit can take screenshots of conversations and send them to family members. The shared unit can also record audio data of conversations and send it to family members. For example, the shared unit can send screenshots and audio data of conversations to family members in real time. This allows family members to detect dangerous situations early and take appropriate action by sending screenshots and audio data of conversations.

[0035] The shared function can display dangerous conversation content on the messaging app's timeline. For example, the shared function can display dangerous conversation content on the messaging app's timeline. The shared function can also display dangerous conversation content on the messaging app's timeline in real time. The shared function can also notify users of dangerous conversation content on the messaging app's timeline. This allows family members to detect dangerous situations early and take appropriate action by displaying dangerous conversation content on the messaging app's timeline.

[0036] The shared section can provide a "guardian dog" themed after a Shiba Inu and the police. The shared section can, for example, provide a "guardian dog" themed after a Shiba Inu and the police as an application. The shared section can, for example, provide a "guardian dog" themed after a Shiba Inu and the police as a notification function. The shared section can, for example, provide a "guardian dog" themed after a Shiba Inu and the police as a web service. In this way, by providing a "guardian dog" themed after a Shiba Inu and the police, it is possible to give users a sense of security.

[0037] The analysis unit can analyze conversation content in detail and identify dangerous words and contexts. For example, the analysis unit can use natural language processing techniques to analyze conversation content in detail. The analysis unit can also use sentiment analysis techniques to analyze conversation content in detail. For example, the analysis unit can identify dangerous words and contexts based on specific keywords or contextual patterns. This allows for detailed analysis of conversation content and identification of dangerous words and contexts, enabling accurate detection of dangerous situations.

[0038] The prediction unit can determine that an account has a history of being reported by other users, or that an account immediately starts a video call after being added, is dangerous. For example, the prediction unit can determine that an account has a history of being reported by other users is dangerous. The prediction unit can also determine that an account immediately starts a video call after being added is dangerous. For example, the prediction unit can determine danger based on the type and frequency of reports. This allows for early detection of dangerous situations by determining that an account has a history of being reported by other users, or that an account immediately starts a video call after being added is dangerous.

[0039] The analysis unit can optimize its analysis algorithm by referring to past conversation history when analyzing conversation content. For example, if a user has a history of having dangerous conversations in the past, the AI ​​can learn that pattern and quickly detect similar conversations. For example, if a user has a history of having many safe conversations in the past, the AI ​​can learn that pattern and reduce false positives. For example, if a user tends to have dangerous conversations during certain time periods, the AI ​​can improve the accuracy of the analysis during those times. In this way, the accuracy of the analysis is improved by optimizing the analysis algorithm by referring to past conversation history.

[0040] The analysis unit can adjust the level of detail in its analysis of conversation content based on specific times of day or days of the week. For example, it can increase the level of detail during times when fraud is more likely to occur, such as at night or on weekends. It can also return to the normal level of detail during relatively safe times, such as during the daytime on weekdays. Furthermore, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail during those periods. By adjusting the level of detail based on specific times of day or days of the week, more effective analysis becomes possible.

[0041] The analysis unit can improve the accuracy of its analysis by considering the user's geographical location when analyzing conversation content. For example, if the user is in an area where fraud is prevalent, the analysis unit will increase its accuracy. For example, if the user is in a safe area, the analysis unit can perform the analysis with normal accuracy. For example, if the user is traveling, the analysis unit can adjust its accuracy by considering fraud information in the travel destination. By improving the accuracy of the analysis by considering the user's geographical location, more accurate analysis becomes possible.

[0042] The analysis unit can analyze the user's social media activity when analyzing conversation content and prioritize the analysis of relevant conversations. For example, if the user has made posts about fraud on social media, the analysis unit will prioritize the analysis of those posts. For example, if the user has interacted with a specific dangerous account on social media, the analysis unit can prioritize the analysis of those conversations. For example, if the user has participated in a specific event on social media, the analysis unit can prioritize the analysis of conversations related to that event. By analyzing the user's social media activity and prioritizing the analysis of relevant conversations, dangerous situations can be detected at an early stage.

[0043] The prediction unit can optimize its prediction algorithm by referring to past risk prediction data when predicting risks. For example, the prediction unit can learn patterns of fraud that have occurred in the past and quickly predict similar patterns. For example, the prediction unit can optimize algorithms to reduce false positives based on past risk prediction data. For example, the prediction unit can predict risks that are likely to occur during specific times of day or on specific days of the week based on past risk prediction data. As a result, the accuracy of predictions is improved by optimizing the prediction algorithm by referring to past risk prediction data.

[0044] The prediction unit can adjust the level of detail in its predictions based on specific behavioral patterns and time periods when predicting risk. For example, it can increase the level of detail in predictions during times when fraud is more likely to occur, such as at night or on weekends. It can also return to normal level of detail during relatively safe times, such as during the daytime on weekdays. For example, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail in predictions during those periods. By adjusting the level of detail in predictions based on specific behavioral patterns and time periods, more effective predictions become possible.

[0045] The prediction unit can improve the accuracy of its predictions by considering the user's geographical location when predicting danger. For example, if the user is in an area where fraud is common, the prediction unit will increase its accuracy. For example, if the user is in a safe area, the prediction unit can perform predictions with normal accuracy. For example, if the user is traveling, the prediction unit can adjust its accuracy by considering fraud information in the travel destination. By improving the accuracy of predictions by considering the user's geographical location, more accurate predictions become possible.

[0046] The prediction unit analyzes a user's social media activity when predicting risks and can prioritize predicting relevant risks. For example, if a user is posting about fraud on social media, the prediction unit will prioritize predicting the content of that post. For example, if a user is interacting with a specific dangerous account on social media, the prediction unit can prioritize predicting that behavior. For example, if a user is participating in a specific event on social media, the prediction unit can prioritize predicting the risks associated with that event. In this way, by analyzing a user's social media activity and prioritizing the prediction of relevant risks, dangerous situations can be detected early.

[0047] The sharing unit can select the optimal transmission method when sending shared information by referring to past sharing history. For example, the sharing unit can select the optimal transmission method based on sharing methods that were effective in the past. For example, the sharing unit can optimize methods to reduce erroneous transmissions based on past sharing history. For example, the sharing unit can select the optimal transmission method for a specific time of day or day of the week based on past sharing history. As a result, the accuracy of information sharing is improved by selecting the optimal transmission method by referring to past sharing history.

[0048] The sharing function can adjust the level of detail of information transmitted based on specific times of day or days of the week. For example, it can transmit detailed information when the situation is urgent, such as at night or on weekends. It can also transmit information at a normal level of detail, such as during weekday daytime hours. Furthermore, it can transmit detailed information when the situation is urgent, such as during specific holidays or event periods. By adjusting the level of detail based on specific times of day or days of the week, more effective information sharing becomes possible.

[0049] The sharing function can improve the accuracy of information transmission by considering the user's geographical location when sending shared information. For example, if the user is in an area with a high incidence of fraud, the sharing function will increase the accuracy of the transmission. For example, if the user is in a safe area, the sharing function can share information with normal transmission accuracy. For example, if the user is traveling, the sharing function can adjust the accuracy of the transmission by considering fraud information in the travel destination. This makes it possible to share information more accurately by improving the accuracy of transmission by considering the user's geographical location.

[0050] The sharing function analyzes a user's social media activity when sending shared information and can prioritize sharing relevant information. For example, if a user has made posts about fraud on social media, the sharing function will prioritize sharing that information. For example, if a user is interacting with a specific dangerous account on social media, the sharing function can prioritize sharing that information. For example, if a user is participating in a specific event on social media, the sharing function can prioritize sharing relevant information. This allows for early detection of dangerous situations by analyzing a user's social media activity and prioritizing the sharing of relevant information.

[0051] The disconnection unit can select the optimal disconnection method by referring to past disconnection history when a conversation is disconnected. For example, the disconnection unit can select the optimal disconnection method based on disconnection methods that have been effective in the past. For example, the disconnection unit can optimize methods to reduce false disconnections based on past disconnection history. For example, the disconnection unit can select the optimal disconnection method for a specific time of day or day of the week based on past disconnection history. In this way, false disconnections can be reduced by selecting the optimal disconnection method by referring to past disconnection history.

[0052] The disconnection mechanism can adjust the level of detail of a disconnection based on specific behavioral patterns or time periods when a conversation is disconnected. For example, the disconnection mechanism can increase the level of detail of a disconnection during times when fraud is more likely to occur, such as at night or on weekends. For example, the disconnection mechanism can return to normal level of detail during relatively safe times, such as during the daytime on weekdays. For example, if fraud tends to increase during certain holidays or event periods, the disconnection mechanism can increase the level of detail of a disconnection during those periods. This allows for more effective disconnections by adjusting the level of detail of a disconnection based on specific behavioral patterns or time periods.

[0053] The disconnection unit can improve the accuracy of disconnecting conversations by considering the user's geographical location. For example, if the user is in an area with a high incidence of fraud, the disconnection unit will increase the accuracy of the disconnection. If the user is in a safe area, for example, the disconnection unit can disconnect the conversation with normal accuracy. If the user is traveling, for example, the disconnection unit can adjust the accuracy of the disconnection by considering fraud information in the travel destination. This allows for more accurate disconnections by improving the accuracy of the disconnection by considering the user's geographical location.

[0054] The disconnection unit can analyze the user's social media activity when disconnecting a conversation and prioritize disconnecting relevant conversations. For example, if the user has made posts about fraud on social media, the disconnection unit will prioritize disconnecting conversations related to that. For example, if the user is interacting with a specific dangerous account on social media, the disconnection unit can prioritize disconnecting conversations related to that account. For example, if the user is participating in a specific event on social media, the disconnection unit can prioritize disconnecting conversations related to that event. This allows for early detection of dangerous situations by analyzing the user's social media activity and prioritizing the disconnection of relevant conversations.

[0055] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0056] The analysis unit can optimize its analysis algorithm by referring to past conversation history when analyzing conversation content. For example, if a user has a history of having dangerous conversations, the AI ​​can learn that pattern and quickly detect similar conversations. Also, if a user has a history of having many safe conversations, the AI ​​can learn that pattern and reduce false positives. Furthermore, if a user tends to have dangerous conversations during certain time periods, the AI ​​can improve the accuracy of the analysis during those times. In this way, the accuracy of the analysis is improved by optimizing the analysis algorithm by referring to past conversation history.

[0057] The analysis unit can adjust the level of detail in its analysis of conversation content based on specific times of day or days of the week. For example, it can increase the level of detail during times when fraud is more likely to occur, such as at night or on weekends. Conversely, it can return to normal level of detail during relatively safe times, such as during the daytime on weekdays. Furthermore, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail during those periods. By adjusting the level of detail based on specific times of day or days of the week, more effective analysis becomes possible.

[0058] The analysis unit can improve the accuracy of its analysis by considering the user's geographical location. For example, if the user is in an area where fraud is common, the accuracy of the analysis can be increased. Conversely, if the user is in a safe area, the analysis can be performed with normal accuracy. Furthermore, if the user is traveling, the accuracy of the analysis can be adjusted to take into account fraud information in the travel destination. By improving the accuracy of the analysis by considering the user's geographical location, more accurate analysis becomes possible.

[0059] The analysis unit can analyze a user's social media activity and prioritize the analysis of relevant conversations. For example, if a user posts about fraud on social media, that content can be prioritized for analysis. Similarly, if a user interacts with a specific dangerous account on social media, that conversation can be prioritized for analysis. Furthermore, if a user participates in a specific event on social media, related conversations can be prioritized for analysis. By analyzing a user's social media activity and prioritizing the analysis of relevant conversations, dangerous situations can be detected early.

[0060] The prediction unit can optimize its prediction algorithm by referring to past risk prediction data when predicting risks. For example, it can learn patterns of fraud that have occurred in the past and quickly predict similar patterns. It can also optimize algorithms to reduce false positives based on past risk prediction data. Furthermore, it can predict risks that are likely to occur during specific time periods or days of the week based on past risk prediction data. In this way, the accuracy of predictions is improved by optimizing the prediction algorithm by referring to past risk prediction data.

[0061] The prediction unit can adjust the level of detail in its predictions based on specific behavioral patterns and time periods when predicting risk. For example, it can increase the level of detail during times when fraud is more likely to occur, such as at night or on weekends. Conversely, it can return to normal level of detail during relatively safe times, such as during the daytime on weekdays. Furthermore, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail during those periods. By adjusting the level of detail based on specific behavioral patterns and time periods, more effective predictions become possible.

[0062] The prediction unit can improve prediction accuracy by considering the user's geographical location. For example, if the user is in an area with a high incidence of fraud, the prediction accuracy can be increased. Conversely, if the user is in a safe area, predictions can be made with normal accuracy. Furthermore, if the user is traveling, the prediction accuracy can be adjusted by considering fraud information in the travel destination. In this way, by improving prediction accuracy by considering the user's geographical location, more accurate predictions become possible.

[0063] The prediction unit can analyze a user's social media activity and prioritize predicting related risks. For example, if a user posts about fraud on social media, it can prioritize predicting the content of such posts. Similarly, if a user interacts with a specific dangerous account on social media, it can prioritize predicting that behavior. Furthermore, if a user participates in a specific event on social media, it can prioritize predicting the related risks. This allows for early detection of dangerous situations by analyzing a user's social media activity and prioritizing the prediction of related risks.

[0064] The following briefly describes the processing flow for example form 1.

[0065] Step 1: The analysis unit analyzes the conversation content. The analysis unit can analyze the content of conversations such as text chats, voice calls, and video calls using natural language processing and sentiment analysis technologies. Step 2: The prediction unit predicts risks based on the conversation content analyzed by the analysis unit. The prediction unit can predict risks such as wire fraud and involvement of minors in criminal activities, and can determine that accounts with a history of being reported by other users or actions such as starting a video call immediately after adding an account are risky. It can also identify dangerous words and contexts to predict risks. Step 3: The sharing unit shares the dangers predicted by the prediction unit with pre-registered family members. The sharing unit can send screenshots or audio data of conversations to family members, or display dangerous conversations on the messaging app's timeline. It can also provide a "monitoring dog" modeled after a Shiba Inu and the police. Step 4: The disconnection unit forcibly disconnects the conversation based on the information shared by the sharing unit. The disconnection unit can forcibly disconnect the conversation by methods such as ending the call or temporarily suspending the account.

[0066] (Example of form 2) The monitoring dog system according to an embodiment of the present invention is a system that detects and protects children and the elderly with low IT literacy from dangerous situations. This monitoring dog system allows the user to have the AI ​​analyze the content of messaging apps (e.g., LINE) or phone calls, and the AI ​​predicts dangers such as wire fraud or involvement of minors in criminal activity, and shares this information with pre-registered family members. This mechanism allows family members to detect dangerous situations early and take appropriate action. For example, the user has the AI ​​analyze the content of messaging apps or phone calls. In this process, the AI ​​analyzes the conversation content in detail and identifies dangerous words and contexts. For example, if dangerous words such as "wire transfer" or "illegal part-time job" are included, the AI ​​will judge the conversation as dangerous. In addition, accounts with a history of being reported by other users and actions such as starting a video call immediately after adding an account are also factors that the AI ​​will judge as dangerous. Next, if the AI ​​determines that something is dangerous, it will share that information with pre-registered family members. Specifically, it will send screenshots or audio data of the conversation content to the family. In addition, the dangerous conversation content will be displayed on the messaging app's timeline. This allows family members to detect dangerous situations early and take appropriate action. Furthermore, if a family member determines that an exchange is inappropriate, they can forcibly terminate the conversation. This feature allows families to respond quickly and prevent harm before it occurs. This service is offered as a paid service for 100 yen per month. Users can gain a sense of security from having a "watchdog" (a Shiba Inu + police motif) watching over them. Also, because the monitoring is done by AI, the feeling of being spied on is diminished, reducing psychological burden. This system can create a society where children and the elderly can use IT with peace of mind. In addition, since criminals are immediately reported and punished, it can create an environment where crime is less likely to occur. In this way, the watchdog system can create a society where children and the elderly can use IT with peace of mind.

[0067] The monitoring dog system according to the embodiment comprises an analysis unit, a prediction unit, a sharing unit, and a disconnection unit. The analysis unit analyzes the content of conversations. The analysis unit analyzes the content of conversations using, for example, natural language processing technology. The analysis unit can also analyze the content of conversations using, for example, sentiment analysis technology. The analysis unit can analyze the content of conversations such as text chats, voice calls, and video calls. The prediction unit predicts dangers based on the conversation content analyzed by the analysis unit. The prediction unit predicts dangers such as wire fraud and involvement of minors in criminal activities. The prediction unit can determine dangers such as accounts with a history of being reported by other users or actions such as starting a video call immediately after adding an account. The prediction unit can identify dangerous words and contexts and predict dangers. The sharing unit shares the dangers predicted by the prediction unit with family members who have been registered in advance. The sharing unit can send, for example, screenshots or audio data of conversations to family members. The sharing unit can display dangerous conversation content on the timeline of a messaging app. The sharing unit can, for example, provide a "monitoring dog" themed after a Shiba Inu and the police. The disconnection unit forcibly disconnects the conversation based on the information shared by the sharing unit. The disconnection unit can forcibly disconnect the conversation by, for example, ending the call or temporarily suspending the account. In this way, the monitoring dog system according to the embodiment can protect users with low IT literacy by analyzing the content of the conversation, predicting danger and sharing it with the family, and forcibly disconnecting the conversation as necessary.

[0068] The analysis unit analyzes the content of conversations. For example, the analysis unit analyzes the content of conversations using natural language processing technology. Natural language processing technology is a technology for understanding text data and extracting meaning, and includes processes such as morphological analysis, syntactic analysis, and semantic analysis. The analysis unit can also analyze the content of conversations using sentiment analysis technology. Sentiment analysis technology is a technology that detects emotions from text and audio data and classifies them into sentiment categories such as positive, negative, and neutral. The analysis unit can analyze the content of conversations such as text chats, voice calls, and video calls. In the case of text chats, the analysis unit obtains chat logs and analyzes the content using natural language processing technology. In the case of voice calls, the analysis unit uses speech recognition technology to convert audio data into text, and then applies natural language processing technology. In the case of video calls, in addition to analyzing audio data, it is also possible to analyze facial expressions and gestures from video data. This allows the analysis unit to analyze various forms of conversation content from multiple angles and accurately grasp the user's intentions and emotions. Furthermore, the analysis unit can refer to past conversation data and user behavior history to perform more accurate analysis. For example, by analyzing the past conversations of a particular user and learning those patterns, the accuracy of analyzing the current conversation content can be improved. This allows the analysis unit to analyze conversation content in real time and provide a foundation for ensuring user safety.

[0069] The prediction unit predicts risks based on the conversation content analyzed by the analysis unit. For example, the prediction unit predicts risks such as wire fraud and involvement of minors in criminal activity. In the case of wire fraud, the prediction unit detects specific keywords and phrases and determines whether they are signs of fraud. For example, if phrases such as "Please transfer the money" or "Please tell me your bank account details" are included, the prediction unit will determine it to be dangerous. In the case of involvement of minors in criminal activity, the prediction unit considers the user's age and behavioral patterns to identify dangerous situations. For example, if a user frequently makes video calls late at night or interacts with accounts that have been reported by other users, the prediction unit may determine it to be dangerous. The prediction unit can determine, for example, that accounts with a history of being reported by other users or actions such as starting a video call immediately after adding an account are dangerous. This allows the prediction unit to comprehensively analyze the user's behavioral patterns and conversation content and detect potential risks early. Furthermore, the prediction unit can identify dangerous words and contexts and predict risks. For example, if phrases such as "Keep it a secret" or "Don't tell anyone" are included, it may indicate a dangerous situation. The prediction unit detects these phrases and predicts potential dangers. This allows the prediction unit to play a crucial role in ensuring user safety and preventing potential hazards before they occur.

[0070] The sharing function shares the dangers predicted by the prediction function with pre-registered family members. For example, the sharing function can send screenshots or audio data of conversations to family members. This allows family members to understand dangerous situations in real time and take appropriate action. The sharing function can also display dangerous conversations on the timeline of a messaging app. This allows family and friends to share dangerous situations and cooperate in responding to them. The sharing function can also provide a "guardian dog" character, for example, based on a Shiba Inu and police. The "guardian dog" is a character that visually represents dangerous situations and serves to warn the user. For example, the "guardian dog" can be displayed on the screen, showing icons or messages indicating dangerous conversations. This allows users to intuitively recognize danger and take appropriate action. Furthermore, the sharing function can promote communication with family and friends and raise awareness of dangerous situations. For example, the sharing function can regularly notify family members about dangerous conversations so that they can constantly monitor the user's safety. In addition, the sharing function can quickly send notifications to family and friends when a dangerous situation occurs, encouraging early action. This means that shared areas play a crucial role in ensuring user safety and can strengthen connections with family and friends.

[0071] The disconnection unit forcibly terminates conversations based on information shared by the sharing unit. The disconnection unit can forcibly terminate conversations in ways such as ending a call or temporarily suspending an account. In the case of ending a call, the disconnection unit immediately terminates the call as soon as dangerous conversation content is detected. This quickly frees the user from a dangerous situation. In the case of temporarily suspending an account, the disconnection unit temporarily blocks contact with the dangerous account, ensuring the user's safety. For example, it temporarily blocks messages and calls from the dangerous account, preventing the user from making further contact. This allows the disconnection unit to provide a quick and effective response to ensure the user's safety. Furthermore, the disconnection unit can notify family and friends that the conversation has been forcibly terminated and share the situation. This allows family and friends to recognize that the user is facing a dangerous situation and take appropriate action. The disconnection unit can also warn the user to prevent dangerous situations from recurring. For example, if dangerous conversation content is detected, it can display a warning message to the user, indicating points to be careful about in the future. In this way, the disconnection unit plays a vital role in ensuring the user's safety and can quickly free them from dangerous situations.

[0072] The shared unit can send screenshots and audio data of conversations to family members. For example, the shared unit can take screenshots of conversations and send them to family members. The shared unit can also record audio data of conversations and send it to family members. For example, the shared unit can send screenshots and audio data of conversations to family members in real time. This allows family members to detect dangerous situations early and take appropriate action by sending screenshots and audio data of conversations.

[0073] The shared function can display dangerous conversation content on the messaging app's timeline. For example, the shared function can display dangerous conversation content on the messaging app's timeline. The shared function can also display dangerous conversation content on the messaging app's timeline in real time. The shared function can also notify users of dangerous conversation content on the messaging app's timeline. This allows family members to detect dangerous situations early and take appropriate action by displaying dangerous conversation content on the messaging app's timeline.

[0074] The shared section can provide a "guardian dog" themed after a Shiba Inu and the police. The shared section can, for example, provide a "guardian dog" themed after a Shiba Inu and the police as an application. The shared section can, for example, provide a "guardian dog" themed after a Shiba Inu and the police as a notification function. The shared section can, for example, provide a "guardian dog" themed after a Shiba Inu and the police as a web service. In this way, by providing a "guardian dog" themed after a Shiba Inu and the police, it is possible to give users a sense of security.

[0075] The analysis unit can analyze conversation content in detail and identify dangerous words and contexts. For example, the analysis unit can use natural language processing techniques to analyze conversation content in detail. The analysis unit can also use sentiment analysis techniques to analyze conversation content in detail. For example, the analysis unit can identify dangerous words and contexts based on specific keywords or contextual patterns. This allows for detailed analysis of conversation content and identification of dangerous words and contexts, enabling accurate detection of dangerous situations.

[0076] The prediction unit can determine that an account has a history of being reported by other users, or that an account immediately starts a video call after being added, is dangerous. For example, the prediction unit can determine that an account has a history of being reported by other users is dangerous. The prediction unit can also determine that an account immediately starts a video call after being added is dangerous. For example, the prediction unit can determine danger based on the type and frequency of reports. This allows for early detection of dangerous situations by determining that an account has a history of being reported by other users, or that an account immediately starts a video call after being added is dangerous.

[0077] The analysis unit can estimate the user's emotions and adjust the accuracy of the conversation analysis based on the estimated emotions. For example, if the user is feeling anxious, the AI ​​will analyze the conversation in more detail to avoid missing any dangerous elements. For example, if the user is relaxed, the AI ​​will analyze the conversation with normal accuracy to avoid excessive warnings. For example, if the user is excited, the AI ​​will quickly analyze the conversation to immediately detect danger. This allows for more accurate analysis by adjusting the accuracy of the conversation analysis based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0078] The analysis unit can optimize its analysis algorithm by referring to past conversation history when analyzing conversation content. For example, if a user has a history of having dangerous conversations in the past, the AI ​​can learn that pattern and quickly detect similar conversations. For example, if a user has a history of having many safe conversations in the past, the AI ​​can learn that pattern and reduce false positives. For example, if a user tends to have dangerous conversations during certain time periods, the AI ​​can improve the accuracy of the analysis during those times. In this way, the accuracy of the analysis is improved by optimizing the analysis algorithm by referring to past conversation history.

[0079] The analysis unit can adjust the level of detail in its analysis of conversation content based on specific times of day or days of the week. For example, it can increase the level of detail during times when fraud is more likely to occur, such as at night or on weekends. It can also return to the normal level of detail during relatively safe times, such as during the daytime on weekdays. Furthermore, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail during those periods. By adjusting the level of detail based on specific times of day or days of the week, more effective analysis becomes possible.

[0080] The analysis unit can estimate the user's emotions and determine the priority of analysis results based on the estimated emotions. For example, if the user is feeling anxious, the analysis unit will prioritize displaying analysis results with a high risk level. For example, if the user is relaxed, the analysis unit can display analysis results with the normal priority level. For example, if the user is excited, the analysis unit can immediately notify the user of analysis results with a high risk level. In this way, by prioritizing analysis results based on the user's emotions, important information can be displayed preferentially. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0081] The analysis unit can improve the accuracy of its analysis by considering the user's geographical location when analyzing conversation content. For example, if the user is in an area where fraud is prevalent, the analysis unit will increase its accuracy. For example, if the user is in a safe area, the analysis unit can perform the analysis with normal accuracy. For example, if the user is traveling, the analysis unit can adjust its accuracy by considering fraud information in the travel destination. By improving the accuracy of the analysis by considering the user's geographical location, more accurate analysis becomes possible.

[0082] The analysis unit can analyze the user's social media activity when analyzing conversation content and prioritize the analysis of relevant conversations. For example, if the user has made posts about fraud on social media, the analysis unit will prioritize the analysis of those posts. For example, if the user has interacted with a specific dangerous account on social media, the analysis unit can prioritize the analysis of those conversations. For example, if the user has participated in a specific event on social media, the analysis unit can prioritize the analysis of conversations related to that event. By analyzing the user's social media activity and prioritizing the analysis of relevant conversations, dangerous situations can be detected at an early stage.

[0083] The prediction unit can estimate the user's emotions and adjust the risk prediction criteria based on the estimated emotions. For example, if the user is feeling anxious, the prediction unit will tighten the risk prediction criteria and detect more dangers. For example, if the user is relaxed, the prediction unit can predict danger using normal criteria. For example, if the user is excited, the prediction unit can instantly predict danger and respond quickly. This allows for more accurate risk prediction by adjusting the risk prediction criteria based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0084] The prediction unit can optimize its prediction algorithm by referring to past risk prediction data when predicting risks. For example, the prediction unit can learn patterns of fraud that have occurred in the past and quickly predict similar patterns. For example, the prediction unit can optimize algorithms to reduce false positives based on past risk prediction data. For example, the prediction unit can predict risks that are likely to occur during specific times of day or on specific days of the week based on past risk prediction data. As a result, the accuracy of predictions is improved by optimizing the prediction algorithm by referring to past risk prediction data.

[0085] The prediction unit can adjust the level of detail in its predictions based on specific behavioral patterns and time periods when predicting risk. For example, it can increase the level of detail in predictions during times when fraud is more likely to occur, such as at night or on weekends. It can also return to normal level of detail during relatively safe times, such as during the daytime on weekdays. For example, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail in predictions during those periods. By adjusting the level of detail in predictions based on specific behavioral patterns and time periods, more effective predictions become possible.

[0086] The prediction unit can estimate the user's emotions and determine the priority of prediction results based on the estimated emotions. For example, if the user is feeling anxious, the prediction unit will prioritize displaying prediction results with a high risk level. For example, if the user is relaxed, the prediction unit can display prediction results with the normal priority level. For example, if the user is excited, the prediction unit can immediately notify the user of prediction results with a high risk level. In this way, by prioritizing prediction results based on the user's emotions, important information can be displayed preferentially. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0087] The prediction unit can improve the accuracy of its predictions by considering the user's geographical location when predicting danger. For example, if the user is in an area where fraud is common, the prediction unit will increase its accuracy. For example, if the user is in a safe area, the prediction unit can perform predictions with normal accuracy. For example, if the user is traveling, the prediction unit can adjust its accuracy by considering fraud information in the travel destination. By improving the accuracy of predictions by considering the user's geographical location, more accurate predictions become possible.

[0088] The prediction unit analyzes a user's social media activity when predicting risks and can prioritize predicting relevant risks. For example, if a user is posting about fraud on social media, the prediction unit will prioritize predicting the content of that post. For example, if a user is interacting with a specific dangerous account on social media, the prediction unit can prioritize predicting that behavior. For example, if a user is participating in a specific event on social media, the prediction unit can prioritize predicting the risks associated with that event. In this way, by analyzing a user's social media activity and prioritizing the prediction of relevant risks, dangerous situations can be detected early.

[0089] The sharing function can estimate the user's emotions and adjust the way shared information is presented based on the estimated emotions. For example, if the user is feeling anxious, the sharing function can provide a sharing method that includes detailed information. For example, if the user is relaxed, the sharing function can provide a normal sharing method. For example, if the user is excited, the sharing function can provide a way to share information quickly. This allows for more appropriate information sharing by adjusting the way shared information is presented based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0090] The sharing unit can select the optimal transmission method when sending shared information by referring to past sharing history. For example, the sharing unit can select the optimal transmission method based on sharing methods that were effective in the past. For example, the sharing unit can optimize methods to reduce erroneous transmissions based on past sharing history. For example, the sharing unit can select the optimal transmission method for a specific time of day or day of the week based on past sharing history. As a result, the accuracy of information sharing is improved by selecting the optimal transmission method by referring to past sharing history.

[0091] The sharing function can adjust the level of detail of information transmitted based on specific times of day or days of the week. For example, it can transmit detailed information when the situation is urgent, such as at night or on weekends. It can also transmit information at a normal level of detail, such as during weekday daytime hours. Furthermore, it can transmit detailed information when the situation is urgent, such as during specific holidays or event periods. By adjusting the level of detail based on specific times of day or days of the week, more effective information sharing becomes possible.

[0092] The sharing function can estimate the user's emotions and prioritize shared information based on those emotions. For example, if the user is feeling anxious, the sharing function will prioritize sharing information with a high degree of risk. If the user is relaxed, the sharing function can share information with a normal priority. If the user is excited, the sharing function can immediately share information with a high degree of risk. This allows important information to be shared preferentially by prioritizing shared information based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0093] The sharing function can improve the accuracy of information transmission by considering the user's geographical location when sending shared information. For example, if the user is in an area with a high incidence of fraud, the sharing function will increase the accuracy of the transmission. For example, if the user is in a safe area, the sharing function can share information with normal transmission accuracy. For example, if the user is traveling, the sharing function can adjust the accuracy of the transmission by considering fraud information in the travel destination. This makes it possible to share information more accurately by improving the accuracy of transmission by considering the user's geographical location.

[0094] The sharing function analyzes a user's social media activity when sending shared information and can prioritize sharing relevant information. For example, if a user has made posts about fraud on social media, the sharing function will prioritize sharing that information. For example, if a user is interacting with a specific dangerous account on social media, the sharing function can prioritize sharing that information. For example, if a user is participating in a specific event on social media, the sharing function can prioritize sharing relevant information. This allows for early detection of dangerous situations by analyzing a user's social media activity and prioritizing the sharing of relevant information.

[0095] The disconnection unit can estimate the user's emotions and adjust the timing of the conversation disconnection based on the estimated emotions. For example, if the user is feeling anxious, the disconnection unit may disconnect the conversation earlier. For example, if the user is relaxed, the disconnection unit may disconnect the conversation at a normal time. For example, if the user is excited, the disconnection unit may disconnect the conversation immediately. This allows for more appropriate timing of the conversation disconnection by adjusting the timing based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0096] The disconnection unit can select the optimal disconnection method by referring to past disconnection history when a conversation is disconnected. For example, the disconnection unit can select the optimal disconnection method based on disconnection methods that have been effective in the past. For example, the disconnection unit can optimize methods to reduce false disconnections based on past disconnection history. For example, the disconnection unit can select the optimal disconnection method for a specific time of day or day of the week based on past disconnection history. In this way, false disconnections can be reduced by selecting the optimal disconnection method by referring to past disconnection history.

[0097] The disconnection mechanism can adjust the level of detail of a disconnection based on specific behavioral patterns or time periods when a conversation is disconnected. For example, the disconnection mechanism can increase the level of detail of a disconnection during times when fraud is more likely to occur, such as at night or on weekends. For example, the disconnection mechanism can return to normal level of detail during relatively safe times, such as during the daytime on weekdays. For example, if fraud tends to increase during certain holidays or event periods, the disconnection mechanism can increase the level of detail of a disconnection during those periods. This allows for more effective disconnections by adjusting the level of detail of a disconnection based on specific behavioral patterns or time periods.

[0098] The disconnection unit can estimate the user's emotions and determine the priority of disconnections based on the estimated emotions. For example, if the user is feeling anxious, the disconnection unit will prioritize disconnecting high-risk conversations. For example, if the user is relaxed, the disconnection unit can disconnect conversations with normal priority. For example, if the user is excited, the disconnection unit can immediately disconnect high-risk conversations. In this way, important conversations can be prioritized by determining the priority of disconnections based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI.

[0099] The disconnection unit can improve the accuracy of disconnecting conversations by considering the user's geographical location. For example, if the user is in an area with a high incidence of fraud, the disconnection unit will increase the accuracy of the disconnection. If the user is in a safe area, for example, the disconnection unit can disconnect the conversation with normal accuracy. If the user is traveling, for example, the disconnection unit can adjust the accuracy of the disconnection by considering fraud information in the travel destination. This allows for more accurate disconnections by improving the accuracy of the disconnection by considering the user's geographical location.

[0100] The disconnection unit can analyze the user's social media activity when disconnecting a conversation and prioritize disconnecting relevant conversations. For example, if the user has made posts about fraud on social media, the disconnection unit will prioritize disconnecting conversations related to that. For example, if the user is interacting with a specific dangerous account on social media, the disconnection unit can prioritize disconnecting conversations related to that account. For example, if the user is participating in a specific event on social media, the disconnection unit can prioritize disconnecting conversations related to that event. This allows for early detection of dangerous situations by analyzing the user's social media activity and prioritizing the disconnection of relevant conversations.

[0101] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0102] The analysis unit can estimate the user's emotions and adjust the accuracy of the conversation analysis based on those emotions. For example, if the user is feeling anxious, the AI ​​can analyze the conversation in more detail to avoid missing any dangerous elements. If the user is relaxed, the AI ​​can analyze the conversation with normal accuracy, avoiding excessive warnings. Furthermore, if the user is excited, the AI ​​can quickly analyze the conversation and immediately detect danger. By adjusting the accuracy of the conversation analysis based on the user's emotions, more accurate analysis becomes possible.

[0103] The analysis unit can optimize its analysis algorithm by referring to past conversation history when analyzing conversation content. For example, if a user has a history of having dangerous conversations, the AI ​​can learn that pattern and quickly detect similar conversations. Also, if a user has a history of having many safe conversations, the AI ​​can learn that pattern and reduce false positives. Furthermore, if a user tends to have dangerous conversations during certain time periods, the AI ​​can improve the accuracy of the analysis during those times. In this way, the accuracy of the analysis is improved by optimizing the analysis algorithm by referring to past conversation history.

[0104] The analysis unit can adjust the level of detail in its analysis of conversation content based on specific times of day or days of the week. For example, it can increase the level of detail during times when fraud is more likely to occur, such as at night or on weekends. Conversely, it can return to normal level of detail during relatively safe times, such as during the daytime on weekdays. Furthermore, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail during those periods. By adjusting the level of detail based on specific times of day or days of the week, more effective analysis becomes possible.

[0105] The analysis unit can improve the accuracy of its analysis by considering the user's geographical location. For example, if the user is in an area where fraud is common, the accuracy of the analysis can be increased. Conversely, if the user is in a safe area, the analysis can be performed with normal accuracy. Furthermore, if the user is traveling, the accuracy of the analysis can be adjusted to take into account fraud information in the travel destination. By improving the accuracy of the analysis by considering the user's geographical location, more accurate analysis becomes possible.

[0106] The analysis unit can analyze a user's social media activity and prioritize the analysis of relevant conversations. For example, if a user posts about fraud on social media, that content can be prioritized for analysis. Similarly, if a user interacts with a specific dangerous account on social media, that conversation can be prioritized for analysis. Furthermore, if a user participates in a specific event on social media, related conversations can be prioritized for analysis. By analyzing a user's social media activity and prioritizing the analysis of relevant conversations, dangerous situations can be detected early.

[0107] The prediction unit can estimate the user's emotions and adjust the risk prediction criteria based on those emotions. For example, if the user is feeling anxious, the risk prediction criteria can be tightened to detect more dangers. If the user is relaxed, the system can predict danger using normal criteria. Furthermore, if the user is excited, the system can instantly predict danger and respond quickly. By adjusting the risk prediction criteria based on the user's emotions, more accurate risk prediction becomes possible.

[0108] The prediction unit can optimize its prediction algorithm by referring to past risk prediction data when predicting risks. For example, it can learn patterns of fraud that have occurred in the past and quickly predict similar patterns. It can also optimize algorithms to reduce false positives based on past risk prediction data. Furthermore, it can predict risks that are likely to occur during specific time periods or days of the week based on past risk prediction data. In this way, the accuracy of predictions is improved by optimizing the prediction algorithm by referring to past risk prediction data.

[0109] The prediction unit can adjust the level of detail in its predictions based on specific behavioral patterns and time periods when predicting risk. For example, it can increase the level of detail during times when fraud is more likely to occur, such as at night or on weekends. Conversely, it can return to normal level of detail during relatively safe times, such as during the daytime on weekdays. Furthermore, if fraud tends to increase during certain holidays or event periods, it can increase the level of detail during those periods. By adjusting the level of detail based on specific behavioral patterns and time periods, more effective predictions become possible.

[0110] The prediction unit can improve prediction accuracy by considering the user's geographical location. For example, if the user is in an area with a high incidence of fraud, the prediction accuracy can be increased. Conversely, if the user is in a safe area, predictions can be made with normal accuracy. Furthermore, if the user is traveling, the prediction accuracy can be adjusted by considering fraud information in the travel destination. In this way, by improving prediction accuracy by considering the user's geographical location, more accurate predictions become possible.

[0111] The prediction unit can analyze a user's social media activity and prioritize predicting related risks. For example, if a user posts about fraud on social media, it can prioritize predicting the content of such posts. Similarly, if a user interacts with a specific dangerous account on social media, it can prioritize predicting that behavior. Furthermore, if a user participates in a specific event on social media, it can prioritize predicting the related risks. This allows for early detection of dangerous situations by analyzing a user's social media activity and prioritizing the prediction of related risks.

[0112] The following briefly describes the processing flow for example form 2.

[0113] Step 1: The analysis unit analyzes the conversation content. The analysis unit can analyze the content of conversations such as text chats, voice calls, and video calls using natural language processing and sentiment analysis technologies. Step 2: The prediction unit predicts risks based on the conversation content analyzed by the analysis unit. The prediction unit can predict risks such as wire fraud and involvement of minors in criminal activities, and can determine that accounts with a history of being reported by other users or actions such as starting a video call immediately after adding an account are risky. It can also identify dangerous words and contexts to predict risks. Step 3: The sharing unit shares the dangers predicted by the prediction unit with pre-registered family members. The sharing unit can send screenshots or audio data of conversations to family members, or display dangerous conversations on the messaging app's timeline. It can also provide a "monitoring dog" modeled after a Shiba Inu and the police. Step 4: The disconnection unit forcibly disconnects the conversation based on the information shared by the sharing unit. The disconnection unit can forcibly disconnect the conversation by methods such as ending the call or temporarily suspending the account.

[0114] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0115] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0116] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0117] Each of the multiple elements described above, including the analysis unit, prediction unit, sharing unit, and disconnection unit, is implemented in at least one of the smart device 14 and the data processing unit 12. For example, the analysis unit is implemented by the processor 46 of the smart device 14 and analyzes the content of the conversation using natural language processing technology and sentiment analysis technology. The prediction unit is implemented by the specific processing unit 290 of the data processing unit 12 and predicts danger based on the analyzed conversation content. The sharing unit is implemented by the control unit 46A of the smart device 14 and shares the predicted danger with family members. The disconnection unit is implemented by the specific processing unit 290 of the data processing unit 12 and forcibly disconnects the conversation as needed. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0118] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0119] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0120] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0121] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0122] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0123] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0124] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0125] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0126] The processor 28 reads a specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0127] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0128] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0129] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0130] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0131] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0132] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0133] Each of the multiple elements described above, including the analysis unit, prediction unit, sharing unit, and disconnection unit, is implemented in at least one of the smart glasses 214 and the data processing unit 12. For example, the analysis unit is implemented by the processor 46 of the smart glasses 214 and analyzes the content of a conversation using natural language processing technology and sentiment analysis technology. The prediction unit is implemented by the identification processing unit 290 of the data processing unit 12 and predicts danger based on the analyzed conversation content. The sharing unit is implemented by the control unit 46A of the smart glasses 214 and shares the predicted danger with family members. The disconnection unit is implemented by the identification processing unit 290 of the data processing unit 12 and forcibly disconnects the conversation as needed. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0134] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0135] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0136] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0137] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0138] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0139] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0140] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0141] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0142] The processor 28 reads a specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0143] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0144] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0145] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0146] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0147] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0148] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0149] Each of the multiple elements described above, including the analysis unit, prediction unit, sharing unit, and disconnection unit, is implemented in at least one of the headset terminal 314 and the data processing unit 12. For example, the analysis unit is implemented by the processor 46 of the headset terminal 314 and analyzes the conversation content using natural language processing technology and sentiment analysis technology. The prediction unit is implemented by the specific processing unit 290 of the data processing unit 12 and predicts danger based on the analyzed conversation content. The sharing unit is implemented by the control unit 46A of the headset terminal 314 and shares the predicted danger with family members. The disconnection unit is implemented by the specific processing unit 290 of the data processing unit 12 and forcibly disconnects the conversation as needed. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0150] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0151] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0152] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0153] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0154] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0155] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0156] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0157] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0158] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0159] The processor 28 reads a specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0160] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0161] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0162] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0163] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0164] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0165] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0166] Each of the multiple elements described above, including the analysis unit, prediction unit, sharing unit, and disconnection unit, is implemented in at least one of the robot 414 and the data processing unit 12. For example, the analysis unit is implemented by the processor 46 of the robot 414 and analyzes the content of the conversation using natural language processing technology and sentiment analysis technology. The prediction unit is implemented by the specific processing unit 290 of the data processing unit 12 and predicts danger based on the analyzed conversation content. The sharing unit is implemented by the control unit 46A of the robot 414 and shares the predicted danger with the family. The disconnection unit is implemented by the specific processing unit 290 of the data processing unit 12 and forcibly disconnects the conversation as needed. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0167] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0168] Figure 9 shows the emotion map 400, in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0169] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0170] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0171] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0172] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0173] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0174] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0175] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0176] 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.

[0177] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0178] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0179] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0180] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0181] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0182] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0183] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0184] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0185] (Note 1) An analysis unit that analyzes the content of the conversation, A prediction unit that predicts danger based on the conversation content analyzed by the aforementioned analysis unit, A sharing unit that shares the risks predicted by the prediction unit with family members who have been registered in advance, A cutting unit that forcibly disconnects the conversation based on the information shared by the aforementioned sharing unit, Equipped with A system characterized by the following features. (Note 2) The aforementioned shared portion is, Send screenshots or audio recordings of the conversation to your family. The system described in Appendix 1, characterized by the features described herein. (Note 3) The aforementioned shared portion is, Displaying dangerous conversation content on the messaging app's timeline The system described in Appendix 1, characterized by the features described herein. (Note 4) The aforementioned shared portion is, Offering a "guard dog" that combines the motifs of a Shiba Inu and the police. The system described in Appendix 1, characterized by the features described herein. (Note 5) The aforementioned analysis unit, The conversation content is analyzed in detail to identify dangerous words and contexts. The system described in Appendix 1, characterized by the features described herein. (Note 6) The prediction unit, Accounts that have a history of being reported by other users, or those that immediately initiate video calls after adding an account, are deemed risky. The system described in Appendix 1, characterized by the features described herein. (Note 7) The aforementioned analysis unit, It estimates the user's emotions and adjusts the accuracy of the conversation analysis based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 8) The aforementioned analysis unit, When analyzing conversation content, the analysis algorithm is optimized by referring to past conversation history. The system described in Appendix 1, characterized by the features described herein. (Note 9) The aforementioned analysis unit, When analyzing conversation content, the level of detail of the analysis is adjusted based on specific time periods and days of the week. The system described in Appendix 1, characterized by the features described herein. (Note 10) The aforementioned analysis unit, It estimates the user's emotions and prioritizes the analysis results based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned analysis unit, When analyzing conversation content, the system takes into account the user's geographical location to improve the accuracy of the analysis. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned analysis unit, When analyzing conversation content, the system analyzes the user's social media activity and prioritizes analyzing relevant conversation content. The system described in Appendix 1, characterized by the features described herein. (Note 13) The prediction unit, The system estimates user sentiment and adjusts risk prediction criteria based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 14) The prediction unit, When predicting hazards, the prediction algorithm is optimized by referring to past hazard prediction data. The system described in Appendix 1, characterized by the features described herein. (Note 15) The prediction unit, When predicting risks, adjust the level of detail in the prediction based on specific behavioral patterns and time periods. The system described in Appendix 1, characterized by the features described herein. (Note 16) The prediction unit, It estimates the user's emotions and prioritizes the prediction results based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 17) The prediction unit, When predicting risks, the system improves prediction accuracy by taking into account the user's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 18) The prediction unit, When predicting risks, the system analyzes users' social media activity and prioritizes predicting relevant risks. The system described in Appendix 1, characterized by the features described herein. (Note 19) The aforementioned shared portion is, It estimates the user's emotions and adjusts how shared information is presented based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 20) The aforementioned shared portion is, When sending shared information, the system will refer to past sharing history to select the most suitable sending method. The system described in Appendix 1, characterized by the features described herein. (Note 21) The aforementioned shared portion is, When sending shared information, adjust the level of detail based on specific time slots or days of the week. The system described in Appendix 1, characterized by the features described herein. (Note 22) The aforementioned shared portion is, It estimates user sentiment and prioritizes shared information based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 23) The aforementioned shared portion is, When sending shared information, we improve the accuracy of transmission by taking into account the user's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 24) The aforementioned shared portion is, When sending shared information, the system analyzes the user's social media activity and prioritizes sharing relevant information. The system described in Appendix 1, characterized by the features described herein. (Note 25) The aforementioned cut portion is It estimates the user's emotions and adjusts the timing of conversation termination based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 26) The aforementioned cut portion is When a conversation is disconnected, the system refers to past disconnection history to select the most suitable disconnection method. The system described in Appendix 1, characterized by the features described herein. (Note 27) The aforementioned cut portion is When a conversation is disconnected, adjust the level of detail of the disconnection based on specific behavioral patterns or time periods. The system described in Appendix 1, characterized by the features described herein. (Note 28) The aforementioned cut portion is It estimates the user's emotions and determines the priority of disconnections based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 29) The aforementioned cut portion is When disconnecting a conversation, the system will improve the accuracy of the disconnection by taking into account the user's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 30) The aforementioned cut portion is When disconnecting a conversation, the system analyzes the user's social media activity and prioritizes disconnecting related conversations. The system described in Appendix 1, characterized by the features described herein. [Explanation of symbols]

[0186] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. An analysis unit that analyzes the content of the conversation, A prediction unit that predicts danger based on the conversation content analyzed by the aforementioned analysis unit, A sharing unit that shares the risks predicted by the prediction unit with family members who have been registered in advance, A cutting unit that forcibly disconnects the conversation based on the information shared by the aforementioned sharing unit, Equipped with A system characterized by the following features.

2. The aforementioned shared portion is, Send screenshots or audio recordings of the conversation to your family. The system according to feature 1.

3. The aforementioned shared portion is, Displaying dangerous conversation content on the messaging app's timeline The system according to feature 1.

4. The aforementioned shared portion is, We offer guard dogs that are a combination of Shiba Inu and police motifs. The system according to feature 1.

5. The aforementioned analysis unit, The conversation content is analyzed in detail to identify dangerous words and contexts. The system according to feature 1.

6. The prediction unit, Accounts that have a history of being reported by other users, or those that immediately initiate video calls after adding an account, are deemed risky. The system according to feature 1.

7. The aforementioned analysis unit, It estimates the user's emotions and adjusts the accuracy of the conversation analysis based on the estimated user emotions. The system according to feature 1.

8. The aforementioned analysis unit, When analyzing conversation content, the analysis algorithm is optimized by referring to past conversation history. The system according to feature 1.

Images