System

Abstract

A system is provided.SOLUTION: The system includes a means for receiving a report through a communication means, an analysis means including a generation AI for analyzing contents of the received report, a means for scoring urgency of the report based on an analysis result, a means for instructing a necessary response based on the score, and a means for feeding back a response result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Child consultation centers suffer from a chronic shortage of staff, while the number of cases of abuse is increasing year by year. Under these circumstances, it is difficult to respond to reports of abuse quickly and accurately, which may increase the risk of abuse being overlooked. In addition, the scrutiny of reports and the collection of detailed information are done manually, which is inefficient and often results in delayed responses. This delays the response of child consultation centers and puts the safety of children at risk, which is an issue. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. An analysis means is provided that uses a generation AI to receive reports via a communication means and analyze their contents. The analysis means analyzes the received report content using natural language processing technology and extracts important keywords and phrases. Furthermore, if the report content is unclear or additional information is required, the generation AI engages in a chat-style dialogue with the user to collect detailed information. Next, a means is provided that scores the urgency of the report based on the analysis results of the generation AI. A means is provided that instructs the necessary response based on the scoring results, allowing child consultation centers to respond quickly according to priority. Furthermore, a means is provided that provides feedback on the response results, and this feedback is used as learning data for the generation AI to improve the accuracy of the system. In this way, the present invention provides a system that reduces the burden on child consultation centers and strengthens efforts to protect children from abuse.

[0006] "Communication means" is a general term for technologies including interfaces and protocols for sending and receiving messages.

[0007] A "report" is information reported about inappropriate behavior such as abuse.

[0008] "Receiving" is the act of acquiring the report content sent by the user.

[0009] "Generative AI" is an artificial intelligence technology that uses natural language processing technology to analyze the content of reports and generate and process information.

[0010] The "analysis means" is a component that has the function of analyzing the received report content using generation AI and extracting important information.

[0011] "Urgency" is an indicator that indicates the importance of the report content and the priority of the response.

[0012] "Scoring" is the process of numerically evaluating the urgency of the analyzed report content.

[0013] "Dialogue" refers to communication in which the generative AI asks additional questions of the user to gather more detailed information.

[0014] "Detailed information" refers to supplementary data necessary to understand the details of the report more specifically, such as the circumstances and frequency of the abuse.

[0015] "Instruction" refers to the act of providing guidance on necessary response actions based on the analysis results and scoring.

[0016] "Feedback" is the act of inputting data to improve system performance based on response results and additional information.

[0017] "Learning data" refers to data including past response results and report contents that is used to improve the accuracy of the generating AI.

[0018] A "child consultation center" is a public institution that deals with issues such as child abuse and neglect. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7]FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The present invention relates to a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the reports. Hereinafter, an embodiment of the present invention will be described.

[0041] System Configuration

[0042] The system mainly consists of the following components:

[0043] 1. A means of communication for users to report (e.g., LINE app).

[0044] 2. A server that receives and analyzes the reports.

[0045] 3. Generative AI that scores the importance of reports and provides specific instructions on how to respond.

[0046] 4. A device that sends notifications based on the scoring results.

[0047] Report acceptance

[0048] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[0049] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[0050] Scrutiny of the report contents

[0051] The server passes the received report content to the generation AI. This generation AI analyzes the report content using natural language processing technology. As a result of the analysis, important keywords and phrases are extracted. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0052] Next, if the report is unclear or if additional information is needed, the AI ​​generates follow-up questions to the user. These questions are asked to understand the situation in more detail. For example, it generates specific questions such as, "Tell me more about this. How often does the person cry?"

[0053] The server sends the generated question to the user as a LINE message.

[0054] If the user responds with additional information, for example, "I cry late every night and my parents keep yelling at me," the device will again send this message to the server.

[0055] Priority Scoring

[0056] The AI ​​generator then scores the urgency of the call based on all the details. The urgency is determined by evaluating the importance of each keyword and phrase and making a comprehensive judgment. For example, information such as "crying every night" and "parents yelling" generates an urgency score of 9 / 10.

[0057] The server stores the score obtained from this generative AI in a database and adds it to a list of reports that child consultation center personnel can use to respond quickly and based on priority.

[0058] Generate response instructions

[0059] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[0060] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[0061] Get feedback

[0062] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[0063] The server uses this feedback as training data for the generative AI to continuously improve the accuracy of the system. In this way, the system will enable the efficient operation of child consultation centers and strengthen efforts to protect children from abuse.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[0067] Step 2:

[0068] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[0069] Step 3:

[0070] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[0071] Step 4:

[0072] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[0073] Step 5:

[0074] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[0075] Step 6:

[0076] If the AI ​​generator is unclear about the content of the report, it generates additional questions for the user. Example question: "Please tell me more about it. How often do you cry?"

[0077] Step 7:

[0078] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[0079] Step 8:

[0080] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[0081] Step 9:

[0082] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[0083] Step 10:

[0084] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[0085] Step 11:

[0086] The AI ​​generator performs a final analysis of the report and generates an urgency score, such as an "urgency score of 9 / 10."

[0087] Step 12:

[0088] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[0089] Step 13:

[0090] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[0091] Step 14:

[0092] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[0093] Step 15:

[0094] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[0095] Step 16:

[0096] The server receives the feedback data and stores it as training data for the generative AI, which improves the system's analysis accuracy.

[0097] Example 1

[0098] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0099] In modern society, reports of abuse need to be responded to quickly and accurately, but existing systems often have difficulty properly determining the priority of the response. As a result, if the report is vague or lacks detailed information, the assessment of the urgency and response may be delayed. This creates the problem of not being able to provide prompt support to victims of abuse.

[0100] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0101] In this invention, the server includes a means for receiving reports via a communication means, an analysis means including a generation AI for analyzing the received report content, and a means for scoring the urgency of the report based on the analysis results. This makes it possible to automatically collect detailed information about the report even when the report content is vague, and quickly determine the priority of an appropriate response.

[0102] "Communication means" refers to a means for receiving notifications from users and transmitting them to other components in the system.

[0103] "Generative AI" is an artificial intelligence technology that analyzes the content of received reports, extracts important information, and scores the urgency of the report.

[0104] "Analysis means" refers to a means that includes a generation AI and has the function of analyzing the content of the received report based on natural language processing technology.

[0105] The "scoring method" is a method for evaluating the urgency of a report and assigning a score based on the report content analyzed by the generation AI.

[0106] The "response instruction means" is a means for instructing the necessary response based on the scored report content.

[0107] A "feedback method" is a means for collecting response results and using them as learning data for the generative AI.

[0108] The "means for generating additional questions" is a means by which the generation AI automatically generates additional questions when the content of the report is unclear and sends them to the user via a communication means.

[0109] The "database storage means" is a means of storing the report content and urgency score analyzed by the generation AI in a database and managing the list of reports.

[0110] This invention describes a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the report. This system includes a communication means, a generative AI, an analysis means, a scoring means, a response instruction means, a feedback means, a follow-up question generation means, and a database storage means.

[0111] System Configuration

[0112] The system consists of the following main components:

[0113] 1. Communication methods (e.g., messaging applications)

[0114] 2. Server with analysis tools

[0115] 3. Generative AI with scoring methods

[0116] 4. Terminal equipped with response instruction means

[0117] 5. Server with Feedback Mechanism

[0118] Report acceptance and processing flow

[0119] A user reports abuse using a messaging application (e.g., LINE). For example, the user might type, "I'm worried because my neighbor's child is crying late at night," and send it. The device receives this report message, formats it as text data, and sends it to the server.

[0120] The server passes the received text data to a generative AI model (e.g., GPT-4 (registered trademark)). The generative AI uses natural language processing technology to analyze the content of the call and extract important keywords and phrases such as "late at night," "crying," and "anxiety."

[0121] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates a question such as, "Please tell me more about it. How often does the child cry?" and the server sends this to the user via communication means. Once the user answers with more detailed information, additional information such as, "The child cries late into the night every night and their parents continue to yell at them" is sent to the server via the device.

[0122] Priority scoring and notification

[0123] The AI ​​generates a score for the urgency of the report based on all the details, including additional information. For example, based on the information "crying every night" and "parents yelling," the AI ​​might assign an urgency score of 9 / 10. The server stores this score in a database and updates the list of reports, which are sorted by urgency.

[0124] The server will then notify the child consultation center staff that reports with a high urgency score should be handled with the highest priority. The device will then display a message such as "This is a top priority" to prompt the child consultation center staff to take action.

[0125] Get feedback

[0126] After completing a case, the child consultation center staff member will provide feedback on the situation to the server. For example, they might enter feedback such as, "I visited the site and confirmed the safety of the child." This feedback information is received by the server and stored in a database as learning data for the generation AI. This feedback continuously improves the accuracy of the system.

[0127] Examples and prompts

[0128] For example, if a user reports on the LINE app that "a child in the neighborhood is crying late at night," the device will send this to the server. The server will then ask the generation AI to analyze the report and extract the keywords "late at night" and "crying." The generation AI will then generate a follow-up question, "How often does the child cry?", and send it to the user. The user will respond with "The child cries late every night, and their parents are constantly yelling at them," and the device will again send this to the server. The generation AI will then rate the urgency score as 9 / 10, and the server will store this score. The server will then issue a notification based on this information, allowing a child consultation center staff member to respond promptly.

[0129] Example prompt sentence:

[0130] User: "I'm worried because my neighbor's child is crying late at night."

[0131] Generator: "Thank you for reporting this. How often do you cry?"

[0132] In this way, the system can quickly and accurately analyze the contents of the report and take appropriate action.

[0133] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0134] Step 1:

[0135] The user opens the LINE app and enters an abuse report. For example, they can enter "I'm worried because a child in my neighborhood is crying late at night" and send it. The entry is completed by pressing the send button on the LINE app.

[0136] Step 2:

[0137] The device receives a report message from the LINE app. The received message is converted into text data and sent to the server. This process is performed automatically, and the report data is transferred to the server in real time. The input data is the user's message, and the output data is transferred to the server as text data.

[0138] Step 3:

[0139] The server passes the received text data to a generative AI model. The generative AI model (e.g., GPT-4) uses natural language processing technology to analyze the report content. The input data is the report content in text format, and the output data is important keywords and phrases. For example, keywords such as "late at night," "crying," and "anxiety" are extracted.

[0140] Step 4:

[0141] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates specific questions such as, "Please tell me more about it. How often is the person crying?" The input data is the vague report, and the output data is a specific question.

[0142] Step 5:

[0143] The server sends the generated follow-up question to the user as a LINE message, allowing the user to provide additional information. The input data is the generated question, and the output data is the LINE message sent to the user.

[0144] Step 6:

[0145] The user receives additional questions and answers them with detailed information in the LINE app. For example, the user might enter an answer such as, "I cry late every night, and my parents keep yelling at me." The input data is the user's answer, and the output data is the message sent again via the LINE app.

[0146] Step 7:

[0147] The terminal again receives a reply message from the user and transmits it as text data to the server. The input data is a message containing the user's detailed information, and the output data is the text data that is again transmitted to the server.

[0148] Step 8:

[0149] The generation AI scores the urgency of the call based on all detailed information, including additional information. The input data is text data containing detailed information, and the output data is an urgency score. For example, it takes into account information such as "crying every night" and "parents yelling" to generate an urgency score of 9 / 10.

[0150] Step 9:

[0151] The server saves the urgency score obtained from this generation AI in a database and updates the list of reports. The list of reports is sorted by urgency. The input data is the urgency score, and the output data is the updated list of reports.

[0152] Step 10:

[0153] The server displays the list of reports sorted by urgency to the child consultation center staff. The input data is a list of reports sorted by urgency, and the output data is a display screen that can be viewed by the child consultation center staff.

[0154] Step 11:

[0155] The terminal sends this response instruction as a notification to the person in charge at the child consultation center. For example, it sends a message such as "Response is required as a top priority." The input data is the response instruction, and the output data is the notification message.

[0156] Step 12:

[0157] After completing the response, the child consultation center staff member will feed back the results to the server. For example, they will input the response result such as "We visited the site and confirmed the safety of the child." The input data is the response result, and the output data is the feedback data to the server.

[0158] Step 13:

[0159] The server receives the feedback information and stores it in a database. This feedback information is used as training data for the generative AI. The input data is the feedback information, and the output data is stored as training data.

[0160] (Application example 1)

[0161] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0162] The purpose of this invention is to effectively receive reports of abuse, crime, disasters, etc., quickly and accurately determine the urgency of the report, and instruct the relevant authorities to take appropriate action. In particular, the goal is to realize a quick and accurate response and improve the safety of local communities by automating the detailed analysis of the report content and the scoring of the urgency based on that analysis.

[0163] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0164] In this invention, the server includes: [means for receiving a report via a communication means; [analysis means including a generation AI for analyzing the content of the received report;] [means for scoring the urgency of the report based on the analysis results;] [means for instructing the necessary response based on the score;] [means for feeding back the response results; and [means for automatically notifying relevant organizations according to the urgency.] This enables detailed analysis of the report content and advanced urgency scoring, allowing relevant organizations to respond quickly and appropriately.

[0165] "Means of communication"

[0166] A device is a means for receiving reports from users and forwarding them to a server, and specifically includes smartphones and chat apps.

[0167] "Generative AI that analyzes received reports"

[0168] is an artificial intelligence used to analyze the content of reports, and uses natural language processing technology to extract important keywords and phrases from the report content.

[0169] "Score the urgency of the report based on the analysis results"

[0170] This involves evaluating the urgency of the report content analyzed by the generative AI and assigning a numerical score.

[0171] "Instruct the necessary action"

[0172] This means instructing relevant agencies to take prompt and appropriate action depending on the urgency score.

[0173] "Provide feedback on the results of the response"

[0174] This means reporting the results of the response to the server and using them as learning data for the generative AI.

[0175] "Automatically notify relevant organizations depending on the level of urgency."

[0176] This means automatically sending notifications to relevant agencies (police, fire department, ambulance, etc.) depending on the urgency of the report.

[0177] "Generative AI that engages in chat-style conversations with users to gather details about reports"

[0178] is a generative AI designed to collect detailed information about report content through dialogue with users, and has the ability to generate questions and analyze user responses.

[0179] This invention is a system that effectively receives reports of abuse, crime, disasters, etc., analyzes them using a generative AI model, and scores the urgency of the report. This system includes a means for receiving reports via a communication means, an analysis means including a generative AI that analyzes the content of the received report, a means for scoring the urgency of the report based on the analysis results, a means for instructing the necessary response based on the score, a means for providing feedback on the response results, and a means for automatically notifying relevant organizations according to the urgency. Specific embodiments of the system are described below.

[0180] Report acceptance

[0181] Users use their smartphones to report situations such as abuse, crime, and disasters. Reports are sent via LINE or the app's own chat system. For example, users can send specific details such as, "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was behaving very suspiciously."

[0182] Receiving and analyzing reports

[0183] The server transfers the text data received from the user to the generation AI to analyze the report content. The generation AI uses natural language processing technology to extract important keywords and phrases from the report content. The generation AI model used is GPT-4 (a model from OpenAI (registered trademark)). For example, keywords such as "suspicious person," "park," and "suspicious behavior" are extracted.

[0184] Urgency Scoring

[0185] The server scores the urgency of the call based on the data analyzed by the AI ​​generator. Scoring is based on a comprehensive assessment of the importance of each keyword and phrase. Urgency is expressed as a score from 1 to 10, and for example, an urgency score of 8 / 10 is assigned based on information such as "suspicious person," "late at night," and "remote location."

[0186] Instructions on necessary actions

[0187] The server then instructs the necessary response based on the urgency score. For example, if the urgency is high, a notification is sent immediately to the relevant authorities (police, fire department, etc.) and they are instructed to take the appropriate action. Notification methods include SMS, email, and a dedicated notification app.

[0188] Feedback of response results

[0189] The server receives the response results from the relevant agencies. For example, they may send feedback such as, "We arrived at the scene and confirmed the identity of the suspicious person." The results are stored in a database and used as learning data for the generative AI. This feedback continuously improves the system's analysis accuracy and scoring accuracy.

[0190] Specific examples

[0191] Here are some example prompts for the AI ​​generator:

[0192] Please analyze the following message for emergency context and urgency:

[0193] "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was acting very suspiciously."

[0194] In this way, detailed and accurate analysis of the report content becomes possible, and appropriate responses are taken, improving the safety of the local community.

[0195] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0196] Step 1:

[0197] The user uses a smartphone to input the report details as a text message. For example, the user might type, "Last night, I saw a suspicious person in a park near my house. He suddenly ran away and was behaving very suspiciously," and then press the send button. The input data is the text message entered by the user. The output data is the text message that was sent.

[0198] Step 2:

[0199] The device (smartphone) sends the report content to the server using a communication method (LINE or the app's own chat system). The input data is the text message sent by the user in step 1. The output data is the text message sent to the server.

[0200] Step 3:

[0201] The server passes the received text message to the generation AI. Specifically, it uses a natural language processing model such as GPT-4 to generate a prompt that analyzes the text message. The input data is the text message of the received report. The output data is the prompt text that is sent to the generation AI. For example, the prompt text might be "Please analyze the following message for emergency context and urgency: 'Last night, I saw a suspicious person in the park near my house. He suddenly ran away and was behaving very suspiciously.'"

[0202] Step 4:

[0203] The generation AI analyzes the report content based on the sent prompt text and extracts important keywords and phrases. The input data is the prompt text and the report content text message. The output data is the analyzed keywords and phrases. For example, extracted words include "suspicious person," "park," and "suspicious behavior."

[0204] Step 5:

[0205] The server scores the urgency of the report based on the keywords and phrases obtained from the generation AI. Specifically, it evaluates the importance of each keyword and phrase and generates an overall score. The input data are the keywords and phrases extracted from the generation AI. The output data is the urgency score. For example, an urgency score of 8 / 10 is output.

[0206] Step 6:

[0207] The server instructs the necessary response based on the urgency score and automatically sends a notification to the relevant organizations. Notification methods include SMS, email, and a dedicated notification app. The input data is the urgency score, and the output data is the notification message sent to the relevant organizations. For example, a notification such as "High urgency report: Suspicious person spotted in park" may be sent.

[0208] Step 7:

[0209] After the relevant organizations have completed their response, they will feed back the results to the server. For example, the server may report something like, "We have arrived at the scene and confirmed the identity of the suspicious person." The input data is the feedback message from the relevant organizations. The output data is the feedback information that is stored in the database.

[0210] Step 8:

[0211] The server stores the feedback information in a database and uses it as learning data for the generation AI. This improves the accuracy of report analysis and scoring from the next time onwards. The input data is the feedback information, and the output data is the updated results of the analysis model used as learning data.

[0212] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0213] The present invention relates to a system that efficiently receives reports of abuse, analyzes them using a generative AI, scores the urgency of the reports, and also recognizes the user's emotions using an emotion engine, and reflects this information in the analysis and scoring of the reports. The following describes an embodiment of the present invention.

[0214] System Configuration

[0215] The system mainly consists of the following components:

[0216] 1. A user interface for reporting via communication means (e.g., LINE app).

[0217] 2. A generation AI that allows the server to receive and analyze the report content.

[0218] 3. A function for the emotion engine to analyze the user's emotional state.

[0219] 4. Storage for the database to store report content and analysis results.

[0220] 5. A function that allows the notification method to notify the child consultation center of the priority of the report based on the scoring results.

[0221] Report acceptance

[0222] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[0223] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[0224] Scrutiny of the report contents

[0225] The server passes the received report content to the generation AI. The generation AI uses natural language processing technology to analyze the report content and extract important keywords and phrases. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0226] emotion recognition

[0227] The emotion engine analyzes emotions from the content of messages sent by users. The emotion engine identifies emotional states (e.g., tension, anxiety, anger, etc.) from the user's language and context, as well as from keyword analysis.

[0228] Collecting detailed information about reports

[0229] The generative AI takes into account the results of the emotion engine and generates appropriate questions for the user if the report is unclear or if additional information is required. The questions are tailored to the user's emotional state. For example, if the user is nervous, the AI ​​will choose questions that will minimize stress.

[0230] The server sends the generated question to the user as a LINE message. If the user answers with additional information, the device sends this message back to the server. For example, specific information such as "I cry late every night and my parents keep yelling at me" can be added.

[0231] Priority Scoring

[0232] The generation AI scores the urgency of the report based on all detailed information and the results of the emotion engine. The urgency is determined comprehensively by evaluating the importance of each keyword and phrase and the user's emotional state. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[0233] The server stores the score obtained from the AI ​​generation in a database and adds it to a list of reports, which child consultation center staff can use to respond quickly and prioritize reports.

[0234] Generate response instructions

[0235] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[0236] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[0237] Get feedback

[0238] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[0239] The server uses this feedback as training data for the generative AI and emotion engine to continuously improve the accuracy of the system. In this way, the system enables the efficient operation of child consultation centers and strengthens efforts to protect children from abuse.

[0240] The processing flow will be explained below.

[0241] Step 1:

[0242] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[0243] Step 2:

[0244] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[0245] Step 3:

[0246] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[0247] Step 4:

[0248] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[0249] Step 5:

[0250] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[0251] Step 6:

[0252] The emotion engine analyzes the received message and identifies the user's emotional state, which can be categorized as tension, anxiety, anger, etc.

[0253] Step 7:

[0254] The generative AI takes the analysis results of the emotion engine and generates additional questions for the user if the report content is unclear. Example question: "Please tell me more about this. How often do you cry?"

[0255] Step 8:

[0256] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[0257] Step 9:

[0258] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[0259] Step 10:

[0260] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[0261] Step 11:

[0262] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[0263] Step 12:

[0264] The AI ​​generator performs a final analysis of the report and generates an urgency score, taking into account the results of the emotion engine. For example, it may rate the urgency as 9 / 10.

[0265] Step 13:

[0266] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[0267] Step 14:

[0268] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[0269] Step 15:

[0270] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[0271] Step 16:

[0272] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[0273] Step 17:

[0274] The server receives the feedback data and stores it as learning data for the generative AI and emotion engine, which improves the system's analysis accuracy.

[0275] Example 2

[0276] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0277] Conventional abuse reporting systems have difficulty accurately determining the urgency of reports, which can lead to delayed responses. Furthermore, because they analyze reports without taking the user's emotions into account, there is a risk of incorrectly assessing the urgency of the report. Furthermore, conventional systems ignore the user's emotional state when collecting details about the report, which can discourage users from providing additional information. A system that solves these problems and processes abuse reports more efficiently and accurately is needed.

[0278] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0279] In this invention, the server includes: [means for receiving a report via a communication means;] [analysis means including a generation AI for analyzing the content of the received report; and] [emotion recognition engine means for analyzing the emotional state of the user.] This makes it possible [to accurately score the urgency of the report based on the results of the generation AI and the emotion recognition engine, and to promptly instruct the necessary response].

[0280] The "communication means" is an interface for receiving a message from a user and transferring it to the server.

[0281] "Generative AI" is an artificial intelligence system that includes technology to analyze the content of received reports and extract important keywords and phrases.

[0282] "Analysis means" is a system component that has the function of analyzing the received report content using generation AI.

[0283] An "emotion recognition engine" is a system that includes technology to analyze the emotional state of a user's report and identify emotions such as anxiety, tension, and anger.

[0284] The "means for scoring urgency" is a system component that has the function of quantifying and evaluating the urgency of the report content based on the analysis results of the generation AI and emotion recognition engine.

[0285] The "means for instructing the necessary response" is an interface for instructing the person in charge on the appropriate response based on the urgency score of the report.

[0286] "Feedback means" refers to a system component that has a process and function for reporting the response results to the system.

[0287] "Generative AI that interacts in chat format" is a system that includes generative AI technology for collecting detailed information about the report content through dialogue with the user.

[0288] This invention is a system that not only efficiently accepts abuse reports, analyzes them using a generative AI, and scores the urgency of the reports, but also recognizes the user's emotions using an emotion engine and reflects this information in the analysis and scoring of the reports. The following describes in detail the embodiments of the invention.

[0289] System Configuration

[0290] The system mainly consists of the following components:

[0291] 1. A user interface for reporting via communication means (e.g., messaging app).

[0292] 2. A generation AI that allows the server to receive and analyze the report content.

[0293] 3. A function for the emotion engine to analyze the user's emotional state.

[0294] 4. Storage for the database to store report content and analysis results.

[0295] 5. A function for notification methods to notify the priority of reports based on scoring results.

[0296] Hardware and software used

[0297] Use a messaging app (e.g., LINE app) as a means of communication.

[0298] For generative AI, we use Python's NLTK library.

[0299] The emotion engine uses the Python DeepMoji library.

[0300] The server requires a machine to analyze, store, and notify data (e.g., a cloud server).

[0301] For the database, an RDBMS such as MySQL (registered trademark) or PostgreSQL can be used.

[0302] Report acceptance

[0303] Users report abuse using a messaging app, for example by sending a message like, "I heard a child in the neighborhood crying late into the night." The device receives this report and sends it to the server as text data.

[0304] Scrutiny of the report contents

[0305] The server passes the received report content to the generation AI, which uses Python's NLTK library to analyze the report content and extract important keywords and phrases.

[0306] emotion recognition

[0307] The emotion engine analyzes emotions from the content of user reports. The emotion engine uses Python's DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the content of reports.

[0308] Priority Scoring

[0309] The AI ​​generator scores the urgency of the call based on the detailed information and the results of the emotion engine. The server saves the generated urgency score in a database and adds it to the call list.

[0310] Collecting detailed information about reports

[0311] The generation AI considers the results of the emotion engine and asks the user additional questions if necessary. For example, it collects specific information such as, "I cry late every night, and my parents keep yelling at me." The server sends the generated questions to the user as messages in a messaging app and receives additional information from the user.

[0312] Generate response instructions

[0313] The list of calls is sorted by urgency score, and the server displays the calls with the highest score to the person in charge. The person in charge responds to the calls with the highest urgency first. The device then sends a notification to the person in charge, such as "Emergency response required."

[0314] Get feedback

[0315] After the person in charge completes the response, the result is fed back to the server. The response result (e.g., "We visited the site and confirmed the safety of the child") is reported to the server. The server uses this feedback as learning data for the generative AI and emotion engine.

[0316] Prompt Sentence Examples

[0317] Here are some examples of prompts to input to a generative AI model:

[0318] You are the AI ​​generator responsible for assessing the urgency level. Please rate the urgency score based on the report content and emotional state below:

[0319] Report: "I heard a child in the neighborhood crying late at night."

[0320] Emotional state: tension, anxiety

[0321] Please rate this report's urgency score from 0 to 10 and explain your reasoning.

[0322] In this way, the system efficiently carries out a series of processes from receiving reports to analysis, scoring, and feedback, helping child consultation centers respond quickly and accurately.

[0323] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0324] Step 1:

[0325] Users use a messaging app to send abuse reports, typing a specific text message such as "I heard a child in the neighborhood crying late at night," and then hitting send.

[0326] Input: The notification message entered by the user.

[0327] Output: Sending signal from the Messages app.

[0328] What happens: The messaging app receives a text message and generates a signal to send.

[0329] Step 2:

[0330] The terminal receives the report from the user and transfers it to the server as text data.

[0331] Input: Outgoing signal from the Messages app.

[0332] Output: Text data to send to the server.

[0333] Specific operation: The message is converted to text format using the messaging app's API and sent to the server.

[0334] Step 3:

[0335] The server passes the received report content to the generation AI, which then analyzes the report content using Python's NLTK library and extracts important keywords and phrases. For example, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0336] Input: Report text data sent from the terminal.

[0337] Output: Parsed keywords and phrases.

[0338] Specific operation: The report content is input into the generation AI, and natural language processing is performed using the NLTK library to extract important keywords.

[0339] Step 4:

[0340] The emotion recognition engine analyzes the emotions of users' reports, using the Python DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the report content.

[0341] Input: Report text data.

[0342] Output: Parsed emotional state.

[0343] Specific operation: Analyzes text data using the DeepMoji library and determines emotional state.

[0344] Step 5:

[0345] The server scores the urgency of the report based on the results of the generation AI and emotion recognition engine. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[0346] Input: Keywords extracted by the generative AI and emotional states from the emotion recognition engine.

[0347] Output: Urgency score.

[0348] Specific operation: The analysis results of the generative AI and emotion recognition engine are integrated, and an algorithm is executed to score the urgency level.

[0349] Step 6:

[0350] The server stores the generated urgency score in a database and adds it to a list of reports, which are sorted by score and made available to personnel in charge of handling them according to priority.

[0351] Input: Urgency score.

[0352] Output: Results saved in database and sorted report list.

[0353] Specific operation: Save the urgency score in the database and sort the report list by urgency score.

[0354] Step 7:

[0355] The server notifies the person in charge of high-priority calls based on the urgency score.

[0356] Input: A sorted list of notifications.

[0357] Output: Notification message to the person in charge.

[0358] Specific operation: Notify the person in charge of a high-urgency report and display a message such as "Emergency response required."

[0359] Step 8:

[0360] After the person in charge completes the response, they will feed back the results to the server. For example, they may report to the server that they have visited the site and confirmed that the child is safe.

[0361] Input: Report of response results.

[0362] Output: Feedback content is registered in a database.

[0363] Specific operation: The results reported by the person in charge are stored in a database and used as learning data for the generative AI and emotion recognition engine.

[0364] (Application example 2)

[0365] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0366] Conventional emergency call systems often lack sufficient analysis of call content and scoring of urgency, resulting in delayed appropriate responses. Additionally, because scoring is done without taking into account the user's emotional state, it is difficult to accurately assess the urgency, leading to issues such as inappropriate responses in tense situations.

[0367] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for receiving a report via a communication means, an analysis means including a generation AI that analyzes the received report content, a means for scoring the urgency of the report based on the analysis results, a means for instructing a necessary response based on the score, a means for providing feedback on the response results, an analysis means including an emotion recognition engine that analyzes the user's emotional state, and a means for extracting important information from the report content using the generation AI. This enables detailed analysis of the report content and accurate scoring of the urgency taking into account the user's emotional state. Furthermore, by promptly instructing an appropriate response and utilizing the response results as feedback, the accuracy and speed of the system can be improved.

[0368] "Communication means" refers to the interface or application used to receive notifications.

[0369] "Generation AI" is an artificial intelligence that analyzes the content of received reports.

[0370] "Analysis means" refers to functions for analyzing the content of reports, including generative AI and emotion recognition engines.

[0371] The "scoring means" is a function that scores the urgency of a report based on the analysis results.

[0372] The "instruction means" is a function that instructs the necessary response based on the score.

[0373] "Feedback means" is a function that collects the response results and feeds them back to the system.

[0374] An "emotion recognition engine" is a system for analyzing a user's emotional state.

[0375] "Means for extracting important information" refers to a function that uses generation AI to extract important keywords and phrases from the report content.

[0376] The "database" is a storage system for saving analyzed report content and scoring results.

[0377] This invention is a system that efficiently receives abuse and emergency calls, analyzes them using generative AI, and scores the urgency of the calls. It also features an emotion recognition engine that recognizes the user's emotions and reflects that information in the analysis and scoring of calls.

[0378] System Configuration

[0379] The system mainly consists of the following components:

[0380] 1. A user interface (such as a smartphone application) for receiving notifications via communication means.

[0381] 2. The analysis means, including the generation AI, analyzes the received report.

[0382] 3. A function that allows the emotion recognition engine to analyze the user's emotional state.

[0383] 4. The scoring means scores the urgency of the call based on the analysis results.

[0384] 5. The instruction tool will indicate the necessary action based on the score.

[0385] 6. The feedback means acquires the response results and feeds them back to the system.

[0386] 7. The database is a storage system for storing analysis results and report contents.

[0387] explanation

[0388] 1. The user makes an emergency call using the app via a communication method, inputting the call contents by voice or text.

[0389] Example: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared."

[0390] 2. The server receives the report and analyzes it using a generation AI, which uses natural language processing technology to extract important information and keywords.

[0391] 3. Analyze the user's emotional state using an emotion recognition engine. The emotion engine not only analyzes keywords but also identifies the user's emotional state (e.g., tension, anxiety, anger, etc.) from the user's vocabulary and context.

[0392] 4. The generation AI scores the urgency of the report based on the analyzed information and the results of the emotion recognition engine. The scoring results reflect important keywords in the report and the user's emotional state.

[0393] 5. The server then determines the necessary response based on the score. For example, if the score is high, it will automatically notify the security company or police.

[0394] Example prompt: "Please analyze the emotional state of the following text: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared.""

[0395] 6. After the response is completed, feedback is collected and used as training data for the generative AI and emotion recognition engine. For example, feedback such as "I visited the site and confirmed the situation" can be collected.

[0396] Hardware and software usage examples

[0397] Hardware: Servers, smartphones

[0398] Software: Google® Cloud Natural Language API, database system (e.g., MySQL)

[0399] This enables detailed analysis of the report content and accurate scoring of the urgency level that takes into account the user's emotional state.In addition, by promptly instructing appropriate responses and utilizing the response results as feedback, the system's accuracy and response speed can be improved.

[0400] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0401] Step 1:

[0402] The user inputs the report details. Using a smartphone app, the user makes an emergency call by voice or text. For example, the user might input, "I saw a suspicious person in my neighborhood. He's wearing black clothes and appears to be carrying a weapon. I'm very scared."

[0403] Input: The report entered by the user (voice or text).

[0404] Output: Received notification data.

[0405] Step 2:

[0406] The terminal sends the received report data to the server. The terminal transfers the report content to the server as text data.

[0407] Input: Received notification data.

[0408] Output: Report data transferred to the server.

[0409] Step 3:

[0410] The server passes the received report data to the generation AI, which uses natural language processing technology to analyze the report content and extract important information and keywords, such as "suspicious person," "black clothing," and "weapon."

[0411] Input: Received notification data.

[0412] Output: Parsed keywords and information.

[0413] Step 4:

[0414] The server analyzes the user's emotional state using an emotion recognition engine, which recognizes the user's emotional state (e.g., fear, anxiety, tension) from the wording and context of the message.

[0415] Input: Received notification data.

[0416] Output: Emotion analysis results (e.g. fear index, anxiety index).

[0417] Step 5:

[0418] The generation AI combines the analysis results and emotion recognition results to score the urgency of the call. For example, it may assign a score of "urgency 8 / 10" based on the call content and emotion analysis results.

[0419] Input: Analyzed keywords and information, sentiment analysis results.

[0420] Output: Urgency score.

[0421] Step 6:

[0422] The server will then instruct the necessary response based on the urgency score. For example, if the urgency score is high, the server will automatically send the report to the police or a security company.

[0423] Input: Urgency score.

[0424] Output: Instructions for the required action.

[0425] Step 7:

[0426] After the response is completed, the response results are collected through feedback channels, such as "The police arrived at the scene and assessed the situation."

[0427] Input: The match result.

[0428] Output: Collected feedback data.

[0429] Step 8:

[0430] The server uses the collected feedback data as training data for the generative AI and emotion recognition engine, which improves the accuracy of the system.

[0431] Input: Collected feedback data.

[0432] Output: Updated training data.

[0433] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0434] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0435] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0436] [Second embodiment]

[0437] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0438] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0439] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0440] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0441] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0442] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0443] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0444] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0445] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0446] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0447] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0448] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0449] The present invention relates to a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the reports. Hereinafter, an embodiment of the present invention will be described.

[0450] System Configuration

[0451] The system mainly consists of the following components:

[0452] 1. A means of communication for users to report (e.g., LINE app).

[0453] 2. A server that receives and analyzes the reports.

[0454] 3. Generative AI that scores the importance of reports and provides specific instructions on how to respond.

[0455] 4. A device that sends notifications based on the scoring results.

[0456] Report acceptance

[0457] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[0458] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[0459] Scrutiny of the report contents

[0460] The server passes the received report content to the generation AI. This generation AI analyzes the report content using natural language processing technology. As a result of the analysis, important keywords and phrases are extracted. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0461] Next, if the report is unclear or if additional information is needed, the AI ​​generates follow-up questions to the user. These questions are asked to understand the situation in more detail. For example, it generates specific questions such as, "Tell me more about this. How often does the person cry?"

[0462] The server sends the generated question to the user as a LINE message.

[0463] If the user responds with additional information, for example, "I cry late every night and my parents keep yelling at me," the device will again send this message to the server.

[0464] Priority Scoring

[0465] The AI ​​generator then scores the urgency of the call based on all the details. The urgency is determined by evaluating the importance of each keyword and phrase and making a comprehensive judgment. For example, information such as "crying every night" and "parents yelling" generates an urgency score of 9 / 10.

[0466] The server stores the score obtained from this generative AI in a database and adds it to a list of reports that child consultation center personnel can use to respond quickly and based on priority.

[0467] Generate response instructions

[0468] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[0469] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[0470] Get feedback

[0471] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[0472] The server uses this feedback as training data for the generative AI to continuously improve the accuracy of the system. In this way, the system will enable the efficient operation of child consultation centers and strengthen efforts to protect children from abuse.

[0473] The processing flow will be explained below.

[0474] Step 1:

[0475] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[0476] Step 2:

[0477] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[0478] Step 3:

[0479] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[0480] Step 4:

[0481] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[0482] Step 5:

[0483] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[0484] Step 6:

[0485] If the AI ​​generator is unclear about the content of the report, it generates additional questions for the user. Example question: "Please tell me more about it. How often do you cry?"

[0486] Step 7:

[0487] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[0488] Step 8:

[0489] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[0490] Step 9:

[0491] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[0492] Step 10:

[0493] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[0494] Step 11:

[0495] The AI ​​generator performs a final analysis of the report and generates an urgency score, such as an "urgency score of 9 / 10."

[0496] Step 12:

[0497] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[0498] Step 13:

[0499] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[0500] Step 14:

[0501] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[0502] Step 15:

[0503] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[0504] Step 16:

[0505] The server receives the feedback data and stores it as training data for the generative AI, which improves the system's analysis accuracy.

[0506] Example 1

[0507] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0508] In modern society, reports of abuse need to be responded to quickly and accurately, but existing systems often have difficulty properly determining the priority of the response. As a result, if the report is vague or lacks detailed information, the assessment of the urgency and response may be delayed. This creates the problem of not being able to provide prompt support to victims of abuse.

[0509] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0510] In this invention, the server includes a means for receiving reports via a communication means, an analysis means including a generation AI for analyzing the received report content, and a means for scoring the urgency of the report based on the analysis results. This makes it possible to automatically collect detailed information about the report even when the report content is vague, and quickly determine the priority of an appropriate response.

[0511] "Communication means" refers to a means for receiving notifications from users and transmitting them to other components in the system.

[0512] "Generative AI" is an artificial intelligence technology that analyzes the content of received reports, extracts important information, and scores the urgency of the report.

[0513] "Analysis means" refers to a means that includes a generation AI and has the function of analyzing the content of the received report based on natural language processing technology.

[0514] The "scoring method" is a method for evaluating the urgency of a report and assigning a score based on the report content analyzed by the generation AI.

[0515] The "response instruction means" is a means for instructing the necessary response based on the scored report content.

[0516] A "feedback method" is a means for collecting response results and using them as learning data for the generative AI.

[0517] The "means for generating additional questions" is a means by which the generation AI automatically generates additional questions when the content of the report is unclear and sends them to the user via a communication means.

[0518] The "database storage means" is a means of storing the report content and urgency score analyzed by the generation AI in a database and managing the list of reports.

[0519] This invention describes a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the report. This system includes a communication means, a generative AI, an analysis means, a scoring means, a response instruction means, a feedback means, a follow-up question generation means, and a database storage means.

[0520] System Configuration

[0521] The system consists of the following main components:

[0522] 1. Communication methods (e.g., messaging applications)

[0523] 2. Server with analysis tools

[0524] 3. Generative AI with scoring methods

[0525] 4. Terminal equipped with response instruction means

[0526] 5. Server with Feedback Mechanism

[0527] Report acceptance and processing flow

[0528] A user reports abuse using a messaging application (e.g., LINE). For example, the user might type, "I'm worried because my neighbor's child is crying late at night," and send it. The device receives this report message, formats it as text data, and sends it to the server.

[0529] The server passes the received text data to a generative AI model (e.g., GPT-4), which uses natural language processing technology to analyze the content of the call and extract important keywords and phrases such as "late at night," "crying," and "anxiety."

[0530] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates a question such as, "Please tell me more about it. How often does the child cry?" and the server sends this to the user via communication means. Once the user answers with more detailed information, additional information such as, "The child cries late into the night every night and their parents continue to yell at them" is sent to the server via the device.

[0531] Priority scoring and notification

[0532] The AI ​​generates a score for the urgency of the report based on all the details, including additional information. For example, based on the information "crying every night" and "parents yelling," the AI ​​might assign an urgency score of 9 / 10. The server stores this score in a database and updates the list of reports, which are sorted by urgency.

[0533] The server will then notify the child consultation center staff that reports with a high urgency score should be handled with the highest priority. The device will then display a message such as "This is a top priority" to prompt the child consultation center staff to take action.

[0534] Get feedback

[0535] After completing a case, the child consultation center staff member will provide feedback on the situation to the server. For example, they might enter feedback such as, "I visited the site and confirmed the safety of the child." This feedback information is received by the server and stored in a database as learning data for the generation AI. This feedback continuously improves the accuracy of the system.

[0536] Examples and prompts

[0537] For example, if a user reports on the LINE app that "a child in the neighborhood is crying late at night," the device will send this to the server. The server will then ask the generation AI to analyze the report and extract the keywords "late at night" and "crying." The generation AI will then generate a follow-up question, "How often does the child cry?", and send it to the user. The user will respond with "The child cries late every night, and their parents are constantly yelling at them," and the device will again send this to the server. The generation AI will then rate the urgency score as 9 / 10, and the server will store this score. The server will then issue a notification based on this information, allowing a child consultation center staff member to respond promptly.

[0538] Example prompt sentence:

[0539] User: "I'm worried because my neighbor's child is crying late at night."

[0540] Generator: "Thank you for reporting this. How often do you cry?"

[0541] In this way, the system can quickly and accurately analyze the contents of the report and take appropriate action.

[0542] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0543] Step 1:

[0544] The user opens the LINE app and enters an abuse report. For example, they can enter "I'm worried because a child in my neighborhood is crying late at night" and send it. The entry is completed by pressing the send button on the LINE app.

[0545] Step 2:

[0546] The device receives a report message from the LINE app. The received message is converted into text data and sent to the server. This process is performed automatically, and the report data is transferred to the server in real time. The input data is the user's message, and the output data is transferred to the server as text data.

[0547] Step 3:

[0548] The server passes the received text data to a generative AI model. The generative AI model (e.g., GPT-4) uses natural language processing technology to analyze the report content. The input data is the report content in text format, and the output data is important keywords and phrases. For example, keywords such as "late at night," "crying," and "anxiety" are extracted.

[0549] Step 4:

[0550] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates specific questions such as, "Please tell me more about it. How often is the person crying?" The input data is the vague report, and the output data is a specific question.

[0551] Step 5:

[0552] The server sends the generated follow-up question to the user as a LINE message, allowing the user to provide additional information. The input data is the generated question, and the output data is the LINE message sent to the user.

[0553] Step 6:

[0554] The user receives additional questions and answers them with detailed information in the LINE app. For example, the user might enter an answer such as, "I cry late every night, and my parents keep yelling at me." The input data is the user's answer, and the output data is the message sent again via the LINE app.

[0555] Step 7:

[0556] The terminal again receives a reply message from the user and transmits it as text data to the server. The input data is a message containing the user's detailed information, and the output data is the text data that is again transmitted to the server.

[0557] Step 8:

[0558] The generation AI scores the urgency of the call based on all detailed information, including additional information. The input data is text data containing detailed information, and the output data is an urgency score. For example, it takes into account information such as "crying every night" and "parents yelling" to generate an urgency score of 9 / 10.

[0559] Step 9:

[0560] The server saves the urgency score obtained from this generation AI in a database and updates the list of reports. The list of reports is sorted by urgency. The input data is the urgency score, and the output data is the updated list of reports.

[0561] Step 10:

[0562] The server displays the list of reports sorted by urgency to the child consultation center staff. The input data is a list of reports sorted by urgency, and the output data is a display screen that can be viewed by the child consultation center staff.

[0563] Step 11:

[0564] The terminal sends this response instruction as a notification to the person in charge at the child consultation center. For example, it sends a message such as "Response is required as a top priority." The input data is the response instruction, and the output data is the notification message.

[0565] Step 12:

[0566] After completing the response, the child consultation center staff member will feed back the results to the server. For example, they will input the response result such as "We visited the site and confirmed the safety of the child." The input data is the response result, and the output data is the feedback data to the server.

[0567] Step 13:

[0568] The server receives the feedback information and stores it in a database. This feedback information is used as training data for the generative AI. The input data is the feedback information, and the output data is stored as training data.

[0569] (Application example 1)

[0570] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0571] The purpose of this invention is to effectively receive reports of abuse, crime, disasters, etc., quickly and accurately determine the urgency of the report, and instruct the relevant authorities to take appropriate action. In particular, the goal is to realize a quick and accurate response and improve the safety of local communities by automating the detailed analysis of the report content and the scoring of the urgency based on that analysis.

[0572] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0573] In this invention, the server includes: [means for receiving a report via a communication means; [analysis means including a generation AI for analyzing the content of the received report;] [means for scoring the urgency of the report based on the analysis results;] [means for instructing the necessary response based on the score;] [means for feeding back the response results; and [means for automatically notifying relevant organizations according to the urgency.] This enables detailed analysis of the report content and advanced urgency scoring, allowing relevant organizations to respond quickly and appropriately.

[0574] "Means of communication"

[0575] A device is a means for receiving reports from users and forwarding them to a server, and specifically includes smartphones and chat apps.

[0576] "Generative AI that analyzes received reports"

[0577] is an artificial intelligence used to analyze the content of reports, and uses natural language processing technology to extract important keywords and phrases from the report content.

[0578] "Score the urgency of the report based on the analysis results"

[0579] This involves evaluating the urgency of the report content analyzed by the generative AI and assigning a numerical score.

[0580] "Instruct the necessary action"

[0581] This means instructing relevant agencies to take prompt and appropriate action depending on the urgency score.

[0582] "Provide feedback on the results of the response"

[0583] This means reporting the results of the response to the server and using them as learning data for the generative AI.

[0584] "Automatically notify relevant organizations depending on the level of urgency."

[0585] This means automatically sending notifications to relevant agencies (police, fire department, ambulance, etc.) depending on the urgency of the report.

[0586] "Generative AI that engages in chat-style conversations with users to gather details about reports"

[0587] is a generative AI designed to collect detailed information about report content through dialogue with users, and has the ability to generate questions and analyze user responses.

[0588] This invention is a system that effectively receives reports of abuse, crime, disasters, etc., analyzes them using a generative AI model, and scores the urgency of the report. This system includes a means for receiving reports via a communication means, an analysis means including a generative AI that analyzes the content of the received report, a means for scoring the urgency of the report based on the analysis results, a means for instructing the necessary response based on the score, a means for providing feedback on the response results, and a means for automatically notifying relevant organizations according to the urgency. Specific embodiments of the system are described below.

[0589] Report acceptance

[0590] Users use their smartphones to report situations such as abuse, crime, and disasters. Reports are sent via LINE or the app's own chat system. For example, users can send specific details such as, "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was behaving very suspiciously."

[0591] Receiving and analyzing reports

[0592] The server transfers the text data received from the user to the generation AI to analyze the report content. The generation AI uses natural language processing technology to extract important keywords and phrases from the report content. The generation AI model used is GPT-4 (a model from OpenAI). For example, keywords such as "suspicious person," "park," and "suspicious behavior" are extracted.

[0593] Urgency Scoring

[0594] The server scores the urgency of the call based on the data analyzed by the AI ​​generator. Scoring is based on a comprehensive assessment of the importance of each keyword and phrase. Urgency is expressed as a score from 1 to 10, and for example, an urgency score of 8 / 10 is assigned based on information such as "suspicious person," "late at night," and "remote location."

[0595] Instructions on necessary actions

[0596] The server then instructs the necessary response based on the urgency score. For example, if the urgency is high, a notification is sent immediately to the relevant authorities (police, fire department, etc.) and they are instructed to take the appropriate action. Notification methods include SMS, email, and a dedicated notification app.

[0597] Feedback of response results

[0598] The server receives the response results from the relevant agencies. For example, they may send feedback such as, "We arrived at the scene and confirmed the identity of the suspicious person." The results are stored in a database and used as learning data for the generative AI. This feedback continuously improves the system's analysis accuracy and scoring accuracy.

[0599] Specific examples

[0600] Here are some example prompts for the AI ​​generator:

[0601] Please analyze the following message for emergency context and urgency:

[0602] "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was acting very suspiciously."

[0603] In this way, detailed and accurate analysis of the report content becomes possible, and appropriate responses are taken, improving the safety of the local community.

[0604] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0605] Step 1:

[0606] The user uses a smartphone to input the report details as a text message. For example, the user might type, "Last night, I saw a suspicious person in a park near my house. He suddenly ran away and was behaving very suspiciously," and then press the send button. The input data is the text message entered by the user. The output data is the text message that was sent.

[0607] Step 2:

[0608] The device (smartphone) sends the report content to the server using a communication method (LINE or the app's own chat system). The input data is the text message sent by the user in step 1. The output data is the text message sent to the server.

[0609] Step 3:

[0610] The server passes the received text message to the generation AI. Specifically, it uses a natural language processing model such as GPT-4 to generate a prompt that analyzes the text message. The input data is the text message of the received report. The output data is the prompt text that is sent to the generation AI. For example, the prompt text might be "Please analyze the following message for emergency context and urgency: 'Last night, I saw a suspicious person in the park near my house. He suddenly ran away and was behaving very suspiciously.'"

[0611] Step 4:

[0612] The generation AI analyzes the report content based on the sent prompt text and extracts important keywords and phrases. The input data is the prompt text and the report content text message. The output data is the analyzed keywords and phrases. For example, extracted words include "suspicious person," "park," and "suspicious behavior."

[0613] Step 5:

[0614] The server scores the urgency of the report based on the keywords and phrases obtained from the generation AI. Specifically, it evaluates the importance of each keyword and phrase and generates an overall score. The input data are the keywords and phrases extracted from the generation AI. The output data is the urgency score. For example, an urgency score of 8 / 10 is output.

[0615] Step 6:

[0616] The server instructs the necessary response based on the urgency score and automatically sends a notification to the relevant organizations. Notification methods include SMS, email, and a dedicated notification app. The input data is the urgency score, and the output data is the notification message sent to the relevant organizations. For example, a notification such as "High urgency report: Suspicious person spotted in park" may be sent.

[0617] Step 7:

[0618] After the relevant organizations have completed their response, they will feed back the results to the server. For example, the server may report something like, "We have arrived at the scene and confirmed the identity of the suspicious person." The input data is the feedback message from the relevant organizations. The output data is the feedback information that is stored in the database.

[0619] Step 8:

[0620] The server stores the feedback information in a database and uses it as learning data for the generation AI. This improves the accuracy of report analysis and scoring from the next time onwards. The input data is the feedback information, and the output data is the updated results of the analysis model used as learning data.

[0621] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0622] The present invention relates to a system that efficiently receives reports of abuse, analyzes them using a generative AI, scores the urgency of the reports, and also recognizes the user's emotions using an emotion engine, and reflects this information in the analysis and scoring of the reports. The following describes an embodiment of the present invention.

[0623] System Configuration

[0624] The system mainly consists of the following components:

[0625] 1. A user interface for reporting via communication means (e.g., LINE app).

[0626] 2. A generation AI that allows the server to receive and analyze the report content.

[0627] 3. A function for the emotion engine to analyze the user's emotional state.

[0628] 4. Storage for the database to store report content and analysis results.

[0629] 5. A function that allows the notification method to notify the child consultation center of the priority of the report based on the scoring results.

[0630] Report acceptance

[0631] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[0632] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[0633] Scrutiny of the report contents

[0634] The server passes the received report content to the generation AI. The generation AI uses natural language processing technology to analyze the report content and extract important keywords and phrases. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0635] emotion recognition

[0636] The emotion engine analyzes emotions from the content of messages sent by users. The emotion engine identifies emotional states (e.g., tension, anxiety, anger, etc.) from the user's language and context, as well as from keyword analysis.

[0637] Collecting detailed information about reports

[0638] The generative AI takes into account the results of the emotion engine and generates appropriate questions for the user if the report is unclear or if additional information is required. The questions are tailored to the user's emotional state. For example, if the user is nervous, the AI ​​will choose questions that will minimize stress.

[0639] The server sends the generated question to the user as a LINE message. If the user answers with additional information, the device sends this message back to the server. For example, specific information such as "I cry late every night and my parents keep yelling at me" can be added.

[0640] Priority Scoring

[0641] The generation AI scores the urgency of the report based on all detailed information and the results of the emotion engine. The urgency is determined comprehensively by evaluating the importance of each keyword and phrase and the user's emotional state. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[0642] The server stores the score obtained from the AI ​​generation in a database and adds it to a list of reports, which child consultation center staff can use to respond quickly and prioritize reports.

[0643] Generate response instructions

[0644] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[0645] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[0646] Get feedback

[0647] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[0648] The server uses this feedback as training data for the generative AI and emotion engine to continuously improve the accuracy of the system. In this way, the system enables the efficient operation of child consultation centers and strengthens efforts to protect children from abuse.

[0649] The processing flow will be explained below.

[0650] Step 1:

[0651] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[0652] Step 2:

[0653] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[0654] Step 3:

[0655] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[0656] Step 4:

[0657] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[0658] Step 5:

[0659] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[0660] Step 6:

[0661] The emotion engine analyzes the received message and identifies the user's emotional state, which can be categorized as tension, anxiety, anger, etc.

[0662] Step 7:

[0663] The generative AI takes the analysis results of the emotion engine and generates additional questions for the user if the report content is unclear. Example question: "Please tell me more about this. How often do you cry?"

[0664] Step 8:

[0665] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[0666] Step 9:

[0667] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[0668] Step 10:

[0669] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[0670] Step 11:

[0671] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[0672] Step 12:

[0673] The AI ​​generator performs a final analysis of the report and generates an urgency score, taking into account the results of the emotion engine. For example, it may rate the urgency as 9 / 10.

[0674] Step 13:

[0675] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[0676] Step 14:

[0677] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[0678] Step 15:

[0679] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[0680] Step 16:

[0681] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[0682] Step 17:

[0683] The server receives the feedback data and stores it as learning data for the generative AI and emotion engine, which improves the system's analysis accuracy.

[0684] Example 2

[0685] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0686] Conventional abuse reporting systems have difficulty accurately determining the urgency of reports, which can lead to delayed responses. Furthermore, because they analyze reports without taking the user's emotions into account, there is a risk of incorrectly assessing the urgency of the report. Furthermore, conventional systems ignore the user's emotional state when collecting details about the report, which can discourage users from providing additional information. A system that solves these problems and processes abuse reports more efficiently and accurately is needed.

[0687] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0688] In this invention, the server includes: [means for receiving a report via a communication means;] [analysis means including a generation AI for analyzing the content of the received report; and] [emotion recognition engine means for analyzing the emotional state of the user.] This makes it possible [to accurately score the urgency of the report based on the results of the generation AI and the emotion recognition engine, and to promptly instruct the necessary response].

[0689] The "communication means" is an interface for receiving a message from a user and transferring it to the server.

[0690] "Generative AI" is an artificial intelligence system that includes technology to analyze the content of received reports and extract important keywords and phrases.

[0691] "Analysis means" is a system component that has the function of analyzing the received report content using generation AI.

[0692] An "emotion recognition engine" is a system that includes technology to analyze the emotional state of a user's report and identify emotions such as anxiety, tension, and anger.

[0693] The "means for scoring urgency" is a system component that has the function of quantifying and evaluating the urgency of the report content based on the analysis results of the generation AI and emotion recognition engine.

[0694] The "means for instructing the necessary response" is an interface for instructing the person in charge on the appropriate response based on the urgency score of the report.

[0695] "Feedback means" refers to a system component that has a process and function for reporting the response results to the system.

[0696] "Generative AI that interacts in chat format" is a system that includes generative AI technology for collecting detailed information about the report content through dialogue with the user.

[0697] This invention is a system that not only efficiently accepts abuse reports, analyzes them using a generative AI, and scores the urgency of the reports, but also recognizes the user's emotions using an emotion engine and reflects this information in the analysis and scoring of the reports. The following describes in detail the embodiments of the invention.

[0698] System Configuration

[0699] The system mainly consists of the following components:

[0700] 1. A user interface for reporting via communication means (e.g., messaging app).

[0701] 2. A generation AI that allows the server to receive and analyze the report content.

[0702] 3. A function for the emotion engine to analyze the user's emotional state.

[0703] 4. Storage for the database to store report content and analysis results.

[0704] 5. A function for notification methods to notify the priority of reports based on scoring results.

[0705] Hardware and software used

[0706] Use a messaging app (e.g., LINE app) as a means of communication.

[0707] For generative AI, we use Python's NLTK library.

[0708] The emotion engine uses the Python DeepMoji library.

[0709] The server requires a machine to analyze, store, and notify data (e.g., a cloud server).

[0710] The database can be an RDBMS such as MySQL or PostgreSQL.

[0711] Report acceptance

[0712] Users report abuse using a messaging app, for example by sending a message like, "I heard a child in the neighborhood crying late into the night." The device receives this report and sends it to the server as text data.

[0713] Scrutiny of the report contents

[0714] The server passes the received report content to the generation AI, which uses Python's NLTK library to analyze the report content and extract important keywords and phrases.

[0715] emotion recognition

[0716] The emotion engine analyzes emotions from the content of user reports. The emotion engine uses Python's DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the content of reports.

[0717] Priority Scoring

[0718] The AI ​​generator scores the urgency of the call based on the detailed information and the results of the emotion engine. The server saves the generated urgency score in a database and adds it to the call list.

[0719] Collecting detailed information about reports

[0720] The generation AI considers the results of the emotion engine and asks the user additional questions if necessary. For example, it collects specific information such as, "I cry late every night, and my parents keep yelling at me." The server sends the generated questions to the user as messages in a messaging app and receives additional information from the user.

[0721] Generate response instructions

[0722] The list of calls is sorted by urgency score, and the server displays the calls with the highest score to the person in charge. The person in charge responds to the calls with the highest urgency first. The device then sends a notification to the person in charge, such as "Emergency response required."

[0723] Get feedback

[0724] After the person in charge completes the response, the result is fed back to the server. The response result (e.g., "We visited the site and confirmed the safety of the child") is reported to the server. The server uses this feedback as learning data for the generative AI and emotion engine.

[0725] Prompt Sentence Examples

[0726] Here are some examples of prompts to input to a generative AI model:

[0727] You are the AI ​​generator responsible for assessing the urgency level. Please rate the urgency score based on the report content and emotional state below:

[0728] Report: "I heard a child in the neighborhood crying late at night."

[0729] Emotional state: tension, anxiety

[0730] Please rate this report's urgency score from 0 to 10 and explain your reasoning.

[0731] In this way, the system efficiently carries out a series of processes from receiving reports to analysis, scoring, and feedback, helping child consultation centers respond quickly and accurately.

[0732] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0733] Step 1:

[0734] Users use a messaging app to send abuse reports, typing a specific text message such as "I heard a child in the neighborhood crying late at night," and then hitting send.

[0735] Input: The notification message entered by the user.

[0736] Output: Sending signal from the Messages app.

[0737] What happens: The messaging app receives a text message and generates a signal to send.

[0738] Step 2:

[0739] The terminal receives the report from the user and transfers it to the server as text data.

[0740] Input: Outgoing signal from the Messages app.

[0741] Output: Text data to send to the server.

[0742] Specific operation: The message is converted to text format using the messaging app's API and sent to the server.

[0743] Step 3:

[0744] The server passes the received report content to the generation AI, which then analyzes the report content using Python's NLTK library and extracts important keywords and phrases. For example, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0745] Input: Report text data sent from the terminal.

[0746] Output: Parsed keywords and phrases.

[0747] Specific operation: The report content is input into the generation AI, and natural language processing is performed using the NLTK library to extract important keywords.

[0748] Step 4:

[0749] The emotion recognition engine analyzes the emotions of users' reports, using the Python DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the report content.

[0750] Input: Report text data.

[0751] Output: Parsed emotional state.

[0752] Specific operation: Analyzes text data using the DeepMoji library and determines emotional state.

[0753] Step 5:

[0754] The server scores the urgency of the report based on the results of the generation AI and emotion recognition engine. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[0755] Input: Keywords extracted by the generative AI and emotional states from the emotion recognition engine.

[0756] Output: Urgency score.

[0757] Specific operation: The analysis results of the generative AI and emotion recognition engine are integrated, and an algorithm is executed to score the urgency level.

[0758] Step 6:

[0759] The server stores the generated urgency score in a database and adds it to a list of reports, which are sorted by score and made available to personnel in charge of handling them according to priority.

[0760] Input: Urgency score.

[0761] Output: Results saved in database and sorted report list.

[0762] Specific operation: Save the urgency score in the database and sort the report list by urgency score.

[0763] Step 7:

[0764] The server notifies the person in charge of high-priority calls based on the urgency score.

[0765] Input: A sorted list of notifications.

[0766] Output: Notification message to the person in charge.

[0767] Specific operation: Notify the person in charge of a high-urgency report and display a message such as "Emergency response required."

[0768] Step 8:

[0769] After the person in charge completes the response, they will feed back the results to the server. For example, they may report to the server that they have visited the site and confirmed that the child is safe.

[0770] Input: Report of response results.

[0771] Output: Feedback content is registered in a database.

[0772] Specific operation: The results reported by the person in charge are stored in a database and used as learning data for the generative AI and emotion recognition engine.

[0773] (Application example 2)

[0774] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0775] Conventional emergency call systems often lack sufficient analysis of call content and scoring of urgency, resulting in delayed appropriate responses. Additionally, because scoring is done without taking into account the user's emotional state, it is difficult to accurately assess the urgency, leading to issues such as inappropriate responses in tense situations.

[0776] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for receiving a report via a communication means, an analysis means including a generation AI that analyzes the received report content, a means for scoring the urgency of the report based on the analysis results, a means for instructing a necessary response based on the score, a means for providing feedback on the response results, an analysis means including an emotion recognition engine that analyzes the user's emotional state, and a means for extracting important information from the report content using the generation AI. This enables detailed analysis of the report content and accurate scoring of the urgency taking into account the user's emotional state. Furthermore, by promptly instructing an appropriate response and utilizing the response results as feedback, the accuracy and speed of the system can be improved.

[0777] "Communication means" refers to the interface or application used to receive notifications.

[0778] "Generation AI" is an artificial intelligence that analyzes the content of received reports.

[0779] "Analysis means" refers to functions for analyzing the content of reports, including generative AI and emotion recognition engines.

[0780] The "scoring means" is a function that scores the urgency of a report based on the analysis results.

[0781] The "instruction means" is a function that instructs the necessary response based on the score.

[0782] "Feedback means" is a function that collects the response results and feeds them back to the system.

[0783] An "emotion recognition engine" is a system for analyzing a user's emotional state.

[0784] "Means for extracting important information" refers to a function that uses generation AI to extract important keywords and phrases from the report content.

[0785] The "database" is a storage system for saving analyzed report content and scoring results.

[0786] This invention is a system that efficiently receives abuse and emergency calls, analyzes them using generative AI, and scores the urgency of the calls. It also features an emotion recognition engine that recognizes the user's emotions and reflects that information in the analysis and scoring of calls.

[0787] System Configuration

[0788] The system mainly consists of the following components:

[0789] 1. A user interface (such as a smartphone application) for receiving notifications via communication means.

[0790] 2. The analysis means, including the generation AI, analyzes the received report.

[0791] 3. A function that allows the emotion recognition engine to analyze the user's emotional state.

[0792] 4. The scoring means scores the urgency of the call based on the analysis results.

[0793] 5. The instruction tool will indicate the necessary action based on the score.

[0794] 6. The feedback means acquires the response results and feeds them back to the system.

[0795] 7. The database is a storage system for storing analysis results and report contents.

[0796] explanation

[0797] 1. The user makes an emergency call using the app via a communication method, inputting the call contents by voice or text.

[0798] Example: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared."

[0799] 2. The server receives the report and analyzes it using a generation AI, which uses natural language processing technology to extract important information and keywords.

[0800] 3. Analyze the user's emotional state using an emotion recognition engine. The emotion engine not only analyzes keywords but also identifies the user's emotional state (e.g., tension, anxiety, anger, etc.) from the user's vocabulary and context.

[0801] 4. The generation AI scores the urgency of the report based on the analyzed information and the results of the emotion recognition engine. The scoring results reflect important keywords in the report and the user's emotional state.

[0802] 5. The server then determines the necessary response based on the score. For example, if the score is high, it will automatically notify the security company or police.

[0803] Example prompt: "Please analyze the emotional state of the following text: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared.""

[0804] 6. After the response is completed, feedback is collected and used as training data for the generative AI and emotion recognition engine. For example, feedback such as "I visited the site and confirmed the situation" can be collected.

[0805] Hardware and software usage examples

[0806] Hardware: Servers, smartphones

[0807] Software: Google Cloud Natural Language API, database system (e.g., MySQL)

[0808] This enables detailed analysis of the report content and accurate scoring of the urgency level that takes into account the user's emotional state.In addition, by promptly instructing appropriate responses and utilizing the response results as feedback, the system's accuracy and response speed can be improved.

[0809] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0810] Step 1:

[0811] The user inputs the report details. Using a smartphone app, the user makes an emergency call by voice or text. For example, the user might input, "I saw a suspicious person in my neighborhood. He's wearing black clothes and appears to be carrying a weapon. I'm very scared."

[0812] Input: The report entered by the user (voice or text).

[0813] Output: Received notification data.

[0814] Step 2:

[0815] The terminal sends the received report data to the server. The terminal transfers the report content to the server as text data.

[0816] Input: Received notification data.

[0817] Output: Report data transferred to the server.

[0818] Step 3:

[0819] The server passes the received report data to the generation AI, which uses natural language processing technology to analyze the report content and extract important information and keywords, such as "suspicious person," "black clothing," and "weapon."

[0820] Input: Received notification data.

[0821] Output: Parsed keywords and information.

[0822] Step 4:

[0823] The server analyzes the user's emotional state using an emotion recognition engine, which recognizes the user's emotional state (e.g., fear, anxiety, tension) from the wording and context of the message.

[0824] Input: Received notification data.

[0825] Output: Emotion analysis results (e.g. fear index, anxiety index).

[0826] Step 5:

[0827] The generation AI combines the analysis results and emotion recognition results to score the urgency of the call. For example, it may assign a score of "urgency 8 / 10" based on the call content and emotion analysis results.

[0828] Input: Analyzed keywords and information, sentiment analysis results.

[0829] Output: Urgency score.

[0830] Step 6:

[0831] The server will then instruct the necessary response based on the urgency score. For example, if the urgency score is high, the server will automatically send the report to the police or a security company.

[0832] Input: Urgency score.

[0833] Output: Instructions for the required action.

[0834] Step 7:

[0835] After the response is completed, the response results are collected through feedback channels, such as "The police arrived at the scene and assessed the situation."

[0836] Input: The match result.

[0837] Output: Collected feedback data.

[0838] Step 8:

[0839] The server uses the collected feedback data as training data for the generative AI and emotion recognition engine, which improves the accuracy of the system.

[0840] Input: Collected feedback data.

[0841] Output: Updated training data.

[0842] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0843] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0844] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0845] [Third embodiment]

[0846] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0847] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0848] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0849] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0850] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0851] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0852] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0853] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0854] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0855] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0856] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0857] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0858] The present invention relates to a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the reports. Hereinafter, an embodiment of the present invention will be described.

[0859] System Configuration

[0860] The system mainly consists of the following components:

[0861] 1. A means of communication for users to report (e.g., LINE app).

[0862] 2. A server that receives and analyzes the reports.

[0863] 3. Generative AI that scores the importance of reports and provides specific instructions on how to respond.

[0864] 4. A device that sends notifications based on the scoring results.

[0865] Report acceptance

[0866] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[0867] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[0868] Scrutiny of the report contents

[0869] The server passes the received report content to the generation AI. This generation AI analyzes the report content using natural language processing technology. As a result of the analysis, important keywords and phrases are extracted. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[0870] Next, if the report is unclear or if additional information is needed, the AI ​​generates follow-up questions to the user. These questions are asked to understand the situation in more detail. For example, it generates specific questions such as, "Tell me more about this. How often does the person cry?"

[0871] The server sends the generated question to the user as a LINE message.

[0872] If the user responds with additional information, for example, "I cry late every night and my parents keep yelling at me," the device will again send this message to the server.

[0873] Priority Scoring

[0874] The AI ​​generator then scores the urgency of the call based on all the details. The urgency is determined by evaluating the importance of each keyword and phrase and making a comprehensive judgment. For example, information such as "crying every night" and "parents yelling" generates an urgency score of 9 / 10.

[0875] The server stores the score obtained from this generative AI in a database and adds it to a list of reports that child consultation center personnel can use to respond quickly and based on priority.

[0876] Generate response instructions

[0877] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[0878] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[0879] Get feedback

[0880] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[0881] The server uses this feedback as training data for the generative AI to continuously improve the accuracy of the system. In this way, the system will enable the efficient operation of child consultation centers and strengthen efforts to protect children from abuse.

[0882] The processing flow will be explained below.

[0883] Step 1:

[0884] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[0885] Step 2:

[0886] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[0887] Step 3:

[0888] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[0889] Step 4:

[0890] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[0891] Step 5:

[0892] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[0893] Step 6:

[0894] If the AI ​​generator is unclear about the content of the report, it generates additional questions for the user. Example question: "Please tell me more about it. How often do you cry?"

[0895] Step 7:

[0896] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[0897] Step 8:

[0898] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[0899] Step 9:

[0900] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[0901] Step 10:

[0902] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[0903] Step 11:

[0904] The AI ​​generator performs a final analysis of the report and generates an urgency score, such as an "urgency score of 9 / 10."

[0905] Step 12:

[0906] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[0907] Step 13:

[0908] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[0909] Step 14:

[0910] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[0911] Step 15:

[0912] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[0913] Step 16:

[0914] The server receives the feedback data and stores it as training data for the generative AI, which improves the system's analysis accuracy.

[0915] Example 1

[0916] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0917] In modern society, reports of abuse need to be responded to quickly and accurately, but existing systems often have difficulty properly determining the priority of the response. As a result, if the report is vague or lacks detailed information, the assessment of the urgency and response may be delayed. This creates the problem of not being able to provide prompt support to victims of abuse.

[0918] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0919] In this invention, the server includes a means for receiving reports via a communication means, an analysis means including a generation AI for analyzing the received report content, and a means for scoring the urgency of the report based on the analysis results. This makes it possible to automatically collect detailed information about the report even when the report content is vague, and quickly determine the priority of an appropriate response.

[0920] "Communication means" refers to a means for receiving notifications from users and transmitting them to other components in the system.

[0921] "Generative AI" is an artificial intelligence technology that analyzes the content of received reports, extracts important information, and scores the urgency of the report.

[0922] "Analysis means" refers to a means that includes a generation AI and has the function of analyzing the content of the received report based on natural language processing technology.

[0923] The "scoring method" is a method for evaluating the urgency of a report and assigning a score based on the report content analyzed by the generation AI.

[0924] The "response instruction means" is a means for instructing the necessary response based on the scored report content.

[0925] A "feedback method" is a means for collecting response results and using them as learning data for the generative AI.

[0926] The "means for generating additional questions" is a means by which the generation AI automatically generates additional questions when the content of the report is unclear and sends them to the user via a communication means.

[0927] The "database storage means" is a means of storing the report content and urgency score analyzed by the generation AI in a database and managing the list of reports.

[0928] This invention describes a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the report. This system includes a communication means, a generative AI, an analysis means, a scoring means, a response instruction means, a feedback means, a follow-up question generation means, and a database storage means.

[0929] System Configuration

[0930] The system consists of the following main components:

[0931] 1. Communication methods (e.g., messaging applications)

[0932] 2. Server with analysis tools

[0933] 3. Generative AI with scoring methods

[0934] 4. Terminal equipped with response instruction means

[0935] 5. Server with Feedback Mechanism

[0936] Report acceptance and processing flow

[0937] A user reports abuse using a messaging application (e.g., LINE). For example, the user might type, "I'm worried because my neighbor's child is crying late at night," and send it. The device receives this report message, formats it as text data, and sends it to the server.

[0938] The server passes the received text data to a generative AI model (e.g., GPT-4), which uses natural language processing technology to analyze the content of the call and extract important keywords and phrases such as "late at night," "crying," and "anxiety."

[0939] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates a question such as, "Please tell me more about it. How often does the child cry?" and the server sends this to the user via communication means. Once the user answers with more detailed information, additional information such as, "The child cries late into the night every night and their parents continue to yell at them" is sent to the server via the device.

[0940] Priority scoring and notification

[0941] The AI ​​generates a score for the urgency of the report based on all the details, including additional information. For example, based on the information "crying every night" and "parents yelling," the AI ​​might assign an urgency score of 9 / 10. The server stores this score in a database and updates the list of reports, which are sorted by urgency.

[0942] The server will then notify the child consultation center staff that reports with a high urgency score should be handled with the highest priority. The device will then display a message such as "This is a top priority" to prompt the child consultation center staff to take action.

[0943] Get feedback

[0944] After completing a case, the child consultation center staff member will provide feedback on the situation to the server. For example, they might enter feedback such as, "I visited the site and confirmed the safety of the child." This feedback information is received by the server and stored in a database as learning data for the generation AI. This feedback continuously improves the accuracy of the system.

[0945] Examples and prompts

[0946] For example, if a user reports on the LINE app that "a child in the neighborhood is crying late at night," the device will send this to the server. The server will then ask the generation AI to analyze the report and extract the keywords "late at night" and "crying." The generation AI will then generate a follow-up question, "How often does the child cry?", and send it to the user. The user will respond with "The child cries late every night, and their parents are constantly yelling at them," and the device will again send this to the server. The generation AI will then rate the urgency score as 9 / 10, and the server will store this score. The server will then issue a notification based on this information, allowing a child consultation center staff member to respond promptly.

[0947] Example prompt sentence:

[0948] User: "I'm worried because my neighbor's child is crying late at night."

[0949] Generator: "Thank you for reporting this. How often do you cry?"

[0950] In this way, the system can quickly and accurately analyze the contents of the report and take appropriate action.

[0951] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0952] Step 1:

[0953] The user opens the LINE app and enters an abuse report. For example, they can enter "I'm worried because a child in my neighborhood is crying late at night" and send it. The entry is completed by pressing the send button on the LINE app.

[0954] Step 2:

[0955] The device receives a report message from the LINE app. The received message is converted into text data and sent to the server. This process is performed automatically, and the report data is transferred to the server in real time. The input data is the user's message, and the output data is transferred to the server as text data.

[0956] Step 3:

[0957] The server passes the received text data to a generative AI model. The generative AI model (e.g., GPT-4) uses natural language processing technology to analyze the report content. The input data is the report content in text format, and the output data is important keywords and phrases. For example, keywords such as "late at night," "crying," and "anxiety" are extracted.

[0958] Step 4:

[0959] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates specific questions such as, "Please tell me more about it. How often is the person crying?" The input data is the vague report, and the output data is a specific question.

[0960] Step 5:

[0961] The server sends the generated follow-up question to the user as a LINE message, allowing the user to provide additional information. The input data is the generated question, and the output data is the LINE message sent to the user.

[0962] Step 6:

[0963] The user receives additional questions and answers them with detailed information in the LINE app. For example, the user might enter an answer such as, "I cry late every night, and my parents keep yelling at me." The input data is the user's answer, and the output data is the message sent again via the LINE app.

[0964] Step 7:

[0965] The terminal again receives a reply message from the user and transmits it as text data to the server. The input data is a message containing the user's detailed information, and the output data is the text data that is again transmitted to the server.

[0966] Step 8:

[0967] The generation AI scores the urgency of the call based on all detailed information, including additional information. The input data is text data containing detailed information, and the output data is an urgency score. For example, it takes into account information such as "crying every night" and "parents yelling" to generate an urgency score of 9 / 10.

[0968] Step 9:

[0969] The server saves the urgency score obtained from this generation AI in a database and updates the list of reports. The list of reports is sorted by urgency. The input data is the urgency score, and the output data is the updated list of reports.

[0970] Step 10:

[0971] The server displays the list of reports sorted by urgency to the child consultation center staff. The input data is a list of reports sorted by urgency, and the output data is a display screen that can be viewed by the child consultation center staff.

[0972] Step 11:

[0973] The terminal sends this response instruction as a notification to the person in charge at the child consultation center. For example, it sends a message such as "Response is required as a top priority." The input data is the response instruction, and the output data is the notification message.

[0974] Step 12:

[0975] After completing the response, the child consultation center staff member will feed back the results to the server. For example, they will input the response result such as "We visited the site and confirmed the safety of the child." The input data is the response result, and the output data is the feedback data to the server.

[0976] Step 13:

[0977] The server receives the feedback information and stores it in a database. This feedback information is used as training data for the generative AI. The input data is the feedback information, and the output data is stored as training data.

[0978] (Application example 1)

[0979] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0980] The purpose of this invention is to effectively receive reports of abuse, crime, disasters, etc., quickly and accurately determine the urgency of the report, and instruct the relevant authorities to take appropriate action. In particular, the goal is to realize a quick and accurate response and improve the safety of local communities by automating the detailed analysis of the report content and the scoring of the urgency based on that analysis.

[0981] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0982] In this invention, the server includes: [means for receiving a report via a communication means; [analysis means including a generation AI for analyzing the content of the received report;] [means for scoring the urgency of the report based on the analysis results;] [means for instructing the necessary response based on the score;] [means for feeding back the response results; and [means for automatically notifying relevant organizations according to the urgency.] This enables detailed analysis of the report content and advanced urgency scoring, allowing relevant organizations to respond quickly and appropriately.

[0983] "Means of communication"

[0984] A device is a means for receiving reports from users and forwarding them to a server, and specifically includes smartphones and chat apps.

[0985] "Generative AI that analyzes received reports"

[0986] is an artificial intelligence used to analyze the content of reports, and uses natural language processing technology to extract important keywords and phrases from the report content.

[0987] "Score the urgency of the report based on the analysis results"

[0988] This involves evaluating the urgency of the report content analyzed by the generative AI and assigning a numerical score.

[0989] "Instruct the necessary action"

[0990] This means instructing relevant agencies to take prompt and appropriate action depending on the urgency score.

[0991] "Provide feedback on the results of the response"

[0992] This means reporting the results of the response to the server and using them as learning data for the generative AI.

[0993] "Automatically notify relevant organizations depending on the level of urgency."

[0994] This means automatically sending notifications to relevant agencies (police, fire department, ambulance, etc.) depending on the urgency of the report.

[0995] "Generative AI that engages in chat-style conversations with users to gather details about reports"

[0996] is a generative AI designed to collect detailed information about report content through dialogue with users, and has the ability to generate questions and analyze user responses.

[0997] This invention is a system that effectively receives reports of abuse, crime, disasters, etc., analyzes them using a generative AI model, and scores the urgency of the report. This system includes a means for receiving reports via a communication means, an analysis means including a generative AI that analyzes the content of the received report, a means for scoring the urgency of the report based on the analysis results, a means for instructing the necessary response based on the score, a means for providing feedback on the response results, and a means for automatically notifying relevant organizations according to the urgency. Specific embodiments of the system are described below.

[0998] Report acceptance

[0999] Users use their smartphones to report situations such as abuse, crime, and disasters. Reports are sent via LINE or the app's own chat system. For example, users can send specific details such as, "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was behaving very suspiciously."

[1000] Receiving and analyzing reports

[1001] The server transfers the text data received from the user to the generation AI to analyze the report content. The generation AI uses natural language processing technology to extract important keywords and phrases from the report content. The generation AI model used is GPT-4 (a model from OpenAI). For example, keywords such as "suspicious person," "park," and "suspicious behavior" are extracted.

[1002] Urgency Scoring

[1003] The server scores the urgency of the call based on the data analyzed by the AI ​​generator. Scoring is based on a comprehensive assessment of the importance of each keyword and phrase. Urgency is expressed as a score from 1 to 10, and for example, an urgency score of 8 / 10 is assigned based on information such as "suspicious person," "late at night," and "remote location."

[1004] Instructions on necessary actions

[1005] The server then instructs the necessary response based on the urgency score. For example, if the urgency is high, a notification is sent immediately to the relevant authorities (police, fire department, etc.) and they are instructed to take the appropriate action. Notification methods include SMS, email, and a dedicated notification app.

[1006] Feedback of response results

[1007] The server receives the response results from the relevant agencies. For example, they may send feedback such as, "We arrived at the scene and confirmed the identity of the suspicious person." The results are stored in a database and used as learning data for the generative AI. This feedback continuously improves the system's analysis accuracy and scoring accuracy.

[1008] Specific examples

[1009] Here are some example prompts for the AI ​​generator:

[1010] Please analyze the following message for emergency context and urgency:

[1011] "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was acting very suspiciously."

[1012] In this way, detailed and accurate analysis of the report content becomes possible, and appropriate responses are taken, improving the safety of the local community.

[1013] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1014] Step 1:

[1015] The user uses a smartphone to input the report details as a text message. For example, the user might type, "Last night, I saw a suspicious person in a park near my house. He suddenly ran away and was behaving very suspiciously," and then press the send button. The input data is the text message entered by the user. The output data is the text message that was sent.

[1016] Step 2:

[1017] The device (smartphone) sends the report content to the server using a communication method (LINE or the app's own chat system). The input data is the text message sent by the user in step 1. The output data is the text message sent to the server.

[1018] Step 3:

[1019] The server passes the received text message to the generation AI. Specifically, it uses a natural language processing model such as GPT-4 to generate a prompt that analyzes the text message. The input data is the text message of the received report. The output data is the prompt text that is sent to the generation AI. For example, the prompt text might be "Please analyze the following message for emergency context and urgency: 'Last night, I saw a suspicious person in the park near my house. He suddenly ran away and was behaving very suspiciously.'"

[1020] Step 4:

[1021] The generation AI analyzes the report content based on the sent prompt text and extracts important keywords and phrases. The input data is the prompt text and the report content text message. The output data is the analyzed keywords and phrases. For example, extracted words include "suspicious person," "park," and "suspicious behavior."

[1022] Step 5:

[1023] The server scores the urgency of the report based on the keywords and phrases obtained from the generation AI. Specifically, it evaluates the importance of each keyword and phrase and generates an overall score. The input data are the keywords and phrases extracted from the generation AI. The output data is the urgency score. For example, an urgency score of 8 / 10 is output.

[1024] Step 6:

[1025] The server instructs the necessary response based on the urgency score and automatically sends a notification to the relevant organizations. Notification methods include SMS, email, and a dedicated notification app. The input data is the urgency score, and the output data is the notification message sent to the relevant organizations. For example, a notification such as "High urgency report: Suspicious person spotted in park" may be sent.

[1026] Step 7:

[1027] After the relevant organizations have completed their response, they will feed back the results to the server. For example, the server may report something like, "We have arrived at the scene and confirmed the identity of the suspicious person." The input data is the feedback message from the relevant organizations. The output data is the feedback information that is stored in the database.

[1028] Step 8:

[1029] The server stores the feedback information in a database and uses it as learning data for the generation AI. This improves the accuracy of report analysis and scoring from the next time onwards. The input data is the feedback information, and the output data is the updated results of the analysis model used as learning data.

[1030] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1031] The present invention relates to a system that efficiently receives reports of abuse, analyzes them using a generative AI, scores the urgency of the reports, and also recognizes the user's emotions using an emotion engine, and reflects this information in the analysis and scoring of the reports. The following describes an embodiment of the present invention.

[1032] System Configuration

[1033] The system mainly consists of the following components:

[1034] 1. A user interface for reporting via communication means (e.g., LINE app).

[1035] 2. A generation AI that allows the server to receive and analyze the report content.

[1036] 3. A function for the emotion engine to analyze the user's emotional state.

[1037] 4. Storage for the database to store report content and analysis results.

[1038] 5. A function that allows the notification method to notify the child consultation center of the priority of the report based on the scoring results.

[1039] Report acceptance

[1040] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[1041] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[1042] Scrutiny of the report contents

[1043] The server passes the received report content to the generation AI. The generation AI uses natural language processing technology to analyze the report content and extract important keywords and phrases. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[1044] emotion recognition

[1045] The emotion engine analyzes emotions from the content of messages sent by users. The emotion engine identifies emotional states (e.g., tension, anxiety, anger, etc.) from the user's language and context, as well as from keyword analysis.

[1046] Collecting detailed information about reports

[1047] The generative AI takes into account the results of the emotion engine and generates appropriate questions for the user if the report is unclear or if additional information is required. The questions are tailored to the user's emotional state. For example, if the user is nervous, the AI ​​will choose questions that will minimize stress.

[1048] The server sends the generated question to the user as a LINE message. If the user answers with additional information, the device sends this message back to the server. For example, specific information such as "I cry late every night and my parents keep yelling at me" can be added.

[1049] Priority Scoring

[1050] The generation AI scores the urgency of the report based on all detailed information and the results of the emotion engine. The urgency is determined comprehensively by evaluating the importance of each keyword and phrase and the user's emotional state. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[1051] The server stores the score obtained from the AI ​​generation in a database and adds it to a list of reports, which child consultation center staff can use to respond quickly and prioritize reports.

[1052] Generate response instructions

[1053] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[1054] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[1055] Get feedback

[1056] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[1057] The server uses this feedback as training data for the generative AI and emotion engine to continuously improve the accuracy of the system. In this way, the system enables the efficient operation of child consultation centers and strengthens efforts to protect children from abuse.

[1058] The processing flow will be explained below.

[1059] Step 1:

[1060] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[1061] Step 2:

[1062] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[1063] Step 3:

[1064] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[1065] Step 4:

[1066] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[1067] Step 5:

[1068] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[1069] Step 6:

[1070] The emotion engine analyzes the received message and identifies the user's emotional state, which can be categorized as tension, anxiety, anger, etc.

[1071] Step 7:

[1072] The generative AI takes the analysis results of the emotion engine and generates additional questions for the user if the report content is unclear. Example question: "Please tell me more about this. How often do you cry?"

[1073] Step 8:

[1074] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[1075] Step 9:

[1076] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[1077] Step 10:

[1078] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[1079] Step 11:

[1080] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[1081] Step 12:

[1082] The AI ​​generator performs a final analysis of the report and generates an urgency score, taking into account the results of the emotion engine. For example, it may rate the urgency as 9 / 10.

[1083] Step 13:

[1084] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[1085] Step 14:

[1086] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[1087] Step 15:

[1088] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[1089] Step 16:

[1090] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[1091] Step 17:

[1092] The server receives the feedback data and stores it as learning data for the generative AI and emotion engine, which improves the system's analysis accuracy.

[1093] Example 2

[1094] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1095] Conventional abuse reporting systems have difficulty accurately determining the urgency of reports, which can lead to delayed responses. Furthermore, because they analyze reports without taking the user's emotions into account, there is a risk of incorrectly assessing the urgency of the report. Furthermore, conventional systems ignore the user's emotional state when collecting details about the report, which can discourage users from providing additional information. A system that solves these problems and processes abuse reports more efficiently and accurately is needed.

[1096] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1097] In this invention, the server includes: [means for receiving a report via a communication means;] [analysis means including a generation AI for analyzing the content of the received report; and] [emotion recognition engine means for analyzing the emotional state of the user.] This makes it possible [to accurately score the urgency of the report based on the results of the generation AI and the emotion recognition engine, and to promptly instruct the necessary response].

[1098] The "communication means" is an interface for receiving a message from a user and transferring it to the server.

[1099] "Generative AI" is an artificial intelligence system that includes technology to analyze the content of received reports and extract important keywords and phrases.

[1100] "Analysis means" is a system component that has the function of analyzing the received report content using generation AI.

[1101] An "emotion recognition engine" is a system that includes technology to analyze the emotional state of a user's report and identify emotions such as anxiety, tension, and anger.

[1102] The "means for scoring urgency" is a system component that has the function of quantifying and evaluating the urgency of the report content based on the analysis results of the generation AI and emotion recognition engine.

[1103] The "means for instructing the necessary response" is an interface for instructing the person in charge on the appropriate response based on the urgency score of the report.

[1104] "Feedback means" refers to a system component that has a process and function for reporting the response results to the system.

[1105] "Generative AI that interacts in chat format" is a system that includes generative AI technology for collecting detailed information about the report content through dialogue with the user.

[1106] This invention is a system that not only efficiently accepts abuse reports, analyzes them using a generative AI, and scores the urgency of the reports, but also recognizes the user's emotions using an emotion engine and reflects this information in the analysis and scoring of the reports. The following describes in detail the embodiments of the invention.

[1107] System Configuration

[1108] The system mainly consists of the following components:

[1109] 1. A user interface for reporting via communication means (e.g., messaging app).

[1110] 2. A generation AI that allows the server to receive and analyze the report content.

[1111] 3. A function for the emotion engine to analyze the user's emotional state.

[1112] 4. Storage for the database to store report content and analysis results.

[1113] 5. A function for notification methods to notify the priority of reports based on scoring results.

[1114] Hardware and software used

[1115] Use a messaging app (e.g., LINE app) as a means of communication.

[1116] For generative AI, we use Python's NLTK library.

[1117] The emotion engine uses the Python DeepMoji library.

[1118] The server requires a machine to analyze, store, and notify data (e.g., a cloud server).

[1119] The database can be an RDBMS such as MySQL or PostgreSQL.

[1120] Report acceptance

[1121] Users report abuse using a messaging app, for example by sending a message like, "I heard a child in the neighborhood crying late into the night." The device receives this report and sends it to the server as text data.

[1122] Scrutiny of the report contents

[1123] The server passes the received report content to the generation AI, which uses Python's NLTK library to analyze the report content and extract important keywords and phrases.

[1124] emotion recognition

[1125] The emotion engine analyzes emotions from the content of user reports. The emotion engine uses Python's DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the content of reports.

[1126] Priority Scoring

[1127] The AI ​​generator scores the urgency of the call based on the detailed information and the results of the emotion engine. The server saves the generated urgency score in a database and adds it to the call list.

[1128] Collecting detailed information about reports

[1129] The generation AI considers the results of the emotion engine and asks the user additional questions if necessary. For example, it collects specific information such as, "I cry late every night, and my parents keep yelling at me." The server sends the generated questions to the user as messages in a messaging app and receives additional information from the user.

[1130] Generate response instructions

[1131] The list of calls is sorted by urgency score, and the server displays the calls with the highest score to the person in charge. The person in charge responds to the calls with the highest urgency first. The device then sends a notification to the person in charge, such as "Emergency response required."

[1132] Get feedback

[1133] After the person in charge completes the response, the result is fed back to the server. The response result (e.g., "We visited the site and confirmed the safety of the child") is reported to the server. The server uses this feedback as learning data for the generative AI and emotion engine.

[1134] Prompt Sentence Examples

[1135] Here are some examples of prompts to input to a generative AI model:

[1136] You are the AI ​​generator responsible for assessing the urgency level. Please rate the urgency score based on the report content and emotional state below:

[1137] Report: "I heard a child in the neighborhood crying late at night."

[1138] Emotional state: tension, anxiety

[1139] Please rate this report's urgency score from 0 to 10 and explain your reasoning.

[1140] In this way, the system efficiently carries out a series of processes from receiving reports to analysis, scoring, and feedback, helping child consultation centers respond quickly and accurately.

[1141] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1142] Step 1:

[1143] Users use a messaging app to send abuse reports, typing a specific text message such as "I heard a child in the neighborhood crying late at night," and then hitting send.

[1144] Input: The notification message entered by the user.

[1145] Output: Sending signal from the Messages app.

[1146] What happens: The messaging app receives a text message and generates a signal to send.

[1147] Step 2:

[1148] The terminal receives the report from the user and transfers it to the server as text data.

[1149] Input: Outgoing signal from the Messages app.

[1150] Output: Text data to send to the server.

[1151] Specific operation: The message is converted to text format using the messaging app's API and sent to the server.

[1152] Step 3:

[1153] The server passes the received report content to the generation AI, which then analyzes the report content using Python's NLTK library and extracts important keywords and phrases. For example, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[1154] Input: Report text data sent from the terminal.

[1155] Output: Parsed keywords and phrases.

[1156] Specific operation: The report content is input into the generation AI, and natural language processing is performed using the NLTK library to extract important keywords.

[1157] Step 4:

[1158] The emotion recognition engine analyzes the emotions of users' reports, using the Python DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the report content.

[1159] Input: Report text data.

[1160] Output: Parsed emotional state.

[1161] Specific operation: Analyzes text data using the DeepMoji library and determines emotional state.

[1162] Step 5:

[1163] The server scores the urgency of the report based on the results of the generation AI and emotion recognition engine. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[1164] Input: Keywords extracted by the generative AI and emotional states from the emotion recognition engine.

[1165] Output: Urgency score.

[1166] Specific operation: The analysis results of the generative AI and emotion recognition engine are integrated, and an algorithm is executed to score the urgency level.

[1167] Step 6:

[1168] The server stores the generated urgency score in a database and adds it to a list of reports, which are sorted by score and made available to personnel in charge of handling them according to priority.

[1169] Input: Urgency score.

[1170] Output: Results saved in database and sorted report list.

[1171] Specific operation: Save the urgency score in the database and sort the report list by urgency score.

[1172] Step 7:

[1173] The server notifies the person in charge of high-priority calls based on the urgency score.

[1174] Input: A sorted list of notifications.

[1175] Output: Notification message to the person in charge.

[1176] Specific operation: Notify the person in charge of a high-urgency report and display a message such as "Emergency response required."

[1177] Step 8:

[1178] After the person in charge completes the response, they will feed back the results to the server. For example, they may report to the server that they have visited the site and confirmed that the child is safe.

[1179] Input: Report of response results.

[1180] Output: Feedback content is registered in a database.

[1181] Specific operation: The results reported by the person in charge are stored in a database and used as learning data for the generative AI and emotion recognition engine.

[1182] (Application example 2)

[1183] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1184] Conventional emergency call systems often lack sufficient analysis of call content and scoring of urgency, resulting in delayed appropriate responses. Additionally, because scoring is done without taking into account the user's emotional state, it is difficult to accurately assess the urgency, leading to issues such as inappropriate responses in tense situations.

[1185] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for receiving a report via a communication means, an analysis means including a generation AI that analyzes the received report content, a means for scoring the urgency of the report based on the analysis results, a means for instructing a necessary response based on the score, a means for providing feedback on the response results, an analysis means including an emotion recognition engine that analyzes the user's emotional state, and a means for extracting important information from the report content using the generation AI. This enables detailed analysis of the report content and accurate scoring of the urgency taking into account the user's emotional state. Furthermore, by promptly instructing an appropriate response and utilizing the response results as feedback, the accuracy and speed of the system can be improved.

[1186] "Communication means" refers to the interface or application used to receive notifications.

[1187] "Generation AI" is an artificial intelligence that analyzes the content of received reports.

[1188] "Analysis means" refers to functions for analyzing the content of reports, including generative AI and emotion recognition engines.

[1189] The "scoring means" is a function that scores the urgency of a report based on the analysis results.

[1190] The "instruction means" is a function that instructs the necessary response based on the score.

[1191] "Feedback means" is a function that collects the response results and feeds them back to the system.

[1192] An "emotion recognition engine" is a system for analyzing a user's emotional state.

[1193] "Means for extracting important information" refers to a function that uses generation AI to extract important keywords and phrases from the report content.

[1194] The "database" is a storage system for saving analyzed report content and scoring results.

[1195] This invention is a system that efficiently receives abuse and emergency calls, analyzes them using generative AI, and scores the urgency of the calls. It also features an emotion recognition engine that recognizes the user's emotions and reflects that information in the analysis and scoring of calls.

[1196] System Configuration

[1197] The system mainly consists of the following components:

[1198] 1. A user interface (such as a smartphone application) for receiving notifications via communication means.

[1199] 2. The analysis means, including the generation AI, analyzes the received report.

[1200] 3. A function that allows the emotion recognition engine to analyze the user's emotional state.

[1201] 4. The scoring means scores the urgency of the call based on the analysis results.

[1202] 5. The instruction tool will indicate the necessary action based on the score.

[1203] 6. The feedback means acquires the response results and feeds them back to the system.

[1204] 7. The database is a storage system for storing analysis results and report contents.

[1205] explanation

[1206] 1. The user makes an emergency call using the app via a communication method, inputting the call contents by voice or text.

[1207] Example: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared."

[1208] 2. The server receives the report and analyzes it using a generation AI, which uses natural language processing technology to extract important information and keywords.

[1209] 3. Analyze the user's emotional state using an emotion recognition engine. The emotion engine not only analyzes keywords but also identifies the user's emotional state (e.g., tension, anxiety, anger, etc.) from the user's vocabulary and context.

[1210] 4. The generation AI scores the urgency of the report based on the analyzed information and the results of the emotion recognition engine. The scoring results reflect important keywords in the report and the user's emotional state.

[1211] 5. The server then determines the necessary response based on the score. For example, if the score is high, it will automatically notify the security company or police.

[1212] Example prompt: "Please analyze the emotional state of the following text: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared.""

[1213] 6. After the response is completed, feedback is collected and used as training data for the generative AI and emotion recognition engine. For example, feedback such as "I visited the site and confirmed the situation" can be collected.

[1214] Hardware and software usage examples

[1215] Hardware: Servers, smartphones

[1216] Software: Google Cloud Natural Language API, database system (e.g., MySQL)

[1217] This enables detailed analysis of the report content and accurate scoring of the urgency level that takes into account the user's emotional state.In addition, by promptly instructing appropriate responses and utilizing the response results as feedback, the system's accuracy and response speed can be improved.

[1218] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1219] Step 1:

[1220] The user inputs the report details. Using a smartphone app, the user makes an emergency call by voice or text. For example, the user might input, "I saw a suspicious person in my neighborhood. He's wearing black clothes and appears to be carrying a weapon. I'm very scared."

[1221] Input: The report entered by the user (voice or text).

[1222] Output: Received notification data.

[1223] Step 2:

[1224] The terminal sends the received report data to the server. The terminal transfers the report content to the server as text data.

[1225] Input: Received notification data.

[1226] Output: Report data transferred to the server.

[1227] Step 3:

[1228] The server passes the received report data to the generation AI, which uses natural language processing technology to analyze the report content and extract important information and keywords, such as "suspicious person," "black clothing," and "weapon."

[1229] Input: Received notification data.

[1230] Output: Parsed keywords and information.

[1231] Step 4:

[1232] The server analyzes the user's emotional state using an emotion recognition engine, which recognizes the user's emotional state (e.g., fear, anxiety, tension) from the wording and context of the message.

[1233] Input: Received notification data.

[1234] Output: Emotion analysis results (e.g. fear index, anxiety index).

[1235] Step 5:

[1236] The generation AI combines the analysis results and emotion recognition results to score the urgency of the call. For example, it may assign a score of "urgency 8 / 10" based on the call content and emotion analysis results.

[1237] Input: Analyzed keywords and information, sentiment analysis results.

[1238] Output: Urgency score.

[1239] Step 6:

[1240] The server will then instruct the necessary response based on the urgency score. For example, if the urgency score is high, the server will automatically send the report to the police or a security company.

[1241] Input: Urgency score.

[1242] Output: Instructions for the required action.

[1243] Step 7:

[1244] After the response is completed, the response results are collected through feedback channels, such as "The police arrived at the scene and assessed the situation."

[1245] Input: The match result.

[1246] Output: Collected feedback data.

[1247] Step 8:

[1248] The server uses the collected feedback data as training data for the generative AI and emotion recognition engine, which improves the accuracy of the system.

[1249] Input: Collected feedback data.

[1250] Output: Updated training data.

[1251] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1252] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1253] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1254] [Fourth embodiment]

[1255] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1256] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1257] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1258] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1259] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1260] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1261] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1262] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1263] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1264] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1265] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1266] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1267] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1268] The present invention relates to a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the reports. Hereinafter, an embodiment of the present invention will be described.

[1269] System Configuration

[1270] The system mainly consists of the following components:

[1271] 1. A means of communication for users to report (e.g., LINE app).

[1272] 2. A server that receives and analyzes the reports.

[1273] 3. Generative AI that scores the importance of reports and provides specific instructions on how to respond.

[1274] 4. A device that sends notifications based on the scoring results.

[1275] Report acceptance

[1276] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[1277] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[1278] Scrutiny of the report contents

[1279] The server passes the received report content to the generation AI. This generation AI analyzes the report content using natural language processing technology. As a result of the analysis, important keywords and phrases are extracted. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[1280] Next, if the report is unclear or if additional information is needed, the AI ​​generates follow-up questions to the user. These questions are asked to understand the situation in more detail. For example, it generates specific questions such as, "Tell me more about this. How often does the person cry?"

[1281] The server sends the generated question to the user as a LINE message.

[1282] If the user responds with additional information, for example, "I cry late every night and my parents keep yelling at me," the device will again send this message to the server.

[1283] Priority Scoring

[1284] The AI ​​generator then scores the urgency of the call based on all the details. The urgency is determined by evaluating the importance of each keyword and phrase and making a comprehensive judgment. For example, information such as "crying every night" and "parents yelling" generates an urgency score of 9 / 10.

[1285] The server stores the score obtained from this generative AI in a database and adds it to a list of reports that child consultation center personnel can use to respond quickly and based on priority.

[1286] Generate response instructions

[1287] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[1288] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[1289] Get feedback

[1290] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[1291] The server uses this feedback as training data for the generative AI to continuously improve the accuracy of the system. In this way, the system will enable the efficient operation of child consultation centers and strengthen efforts to protect children from abuse.

[1292] The processing flow will be explained below.

[1293] Step 1:

[1294] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[1295] Step 2:

[1296] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[1297] Step 3:

[1298] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[1299] Step 4:

[1300] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[1301] Step 5:

[1302] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[1303] Step 6:

[1304] If the AI ​​generator is unclear about the content of the report, it generates additional questions for the user. Example question: "Please tell me more about it. How often do you cry?"

[1305] Step 7:

[1306] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[1307] Step 8:

[1308] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[1309] Step 9:

[1310] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[1311] Step 10:

[1312] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[1313] Step 11:

[1314] The AI ​​generator performs a final analysis of the report and generates an urgency score, such as an "urgency score of 9 / 10."

[1315] Step 12:

[1316] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[1317] Step 13:

[1318] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[1319] Step 14:

[1320] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[1321] Step 15:

[1322] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[1323] Step 16:

[1324] The server receives the feedback data and stores it as training data for the generative AI, which improves the system's analysis accuracy.

[1325] Example 1

[1326] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1327] In modern society, reports of abuse need to be responded to quickly and accurately, but existing systems often have difficulty properly determining the priority of the response. As a result, if the report is vague or lacks detailed information, the assessment of the urgency and response may be delayed. This creates the problem of not being able to provide prompt support to victims of abuse.

[1328] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1329] In this invention, the server includes a means for receiving reports via a communication means, an analysis means including a generation AI for analyzing the received report content, and a means for scoring the urgency of the report based on the analysis results. This makes it possible to automatically collect detailed information about the report even when the report content is vague, and quickly determine the priority of an appropriate response.

[1330] "Communication means" refers to a means for receiving notifications from users and transmitting them to other components in the system.

[1331] "Generative AI" is an artificial intelligence technology that analyzes the content of received reports, extracts important information, and scores the urgency of the report.

[1332] "Analysis means" refers to a means that includes a generation AI and has the function of analyzing the content of the received report based on natural language processing technology.

[1333] The "scoring method" is a method for evaluating the urgency of a report and assigning a score based on the report content analyzed by the generation AI.

[1334] The "response instruction means" is a means for instructing the necessary response based on the scored report content.

[1335] A "feedback method" is a means for collecting response results and using them as learning data for the generative AI.

[1336] The "means for generating additional questions" is a means by which the generation AI automatically generates additional questions when the content of the report is unclear and sends them to the user via a communication means.

[1337] The "database storage means" is a means of storing the report content and urgency score analyzed by the generation AI in a database and managing the list of reports.

[1338] This invention describes a system that effectively receives reports of abuse, analyzes them using a generative AI, and scores the urgency of the report. This system includes a communication means, a generative AI, an analysis means, a scoring means, a response instruction means, a feedback means, a follow-up question generation means, and a database storage means.

[1339] System Configuration

[1340] The system consists of the following main components:

[1341] 1. Communication methods (e.g., messaging applications)

[1342] 2. Server with analysis tools

[1343] 3. Generative AI with scoring methods

[1344] 4. Terminal equipped with response instruction means

[1345] 5. Server with Feedback Mechanism

[1346] Report acceptance and processing flow

[1347] A user reports abuse using a messaging application (e.g., LINE). For example, the user might type, "I'm worried because my neighbor's child is crying late at night," and send it. The device receives this report message, formats it as text data, and sends it to the server.

[1348] The server passes the received text data to a generative AI model (e.g., GPT-4), which uses natural language processing technology to analyze the content of the call and extract important keywords and phrases such as "late at night," "crying," and "anxiety."

[1349] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates a question such as, "Please tell me more about it. How often does the child cry?" and the server sends this to the user via communication means. Once the user answers with more detailed information, additional information such as, "The child cries late into the night every night and their parents continue to yell at them" is sent to the server via the device.

[1350] Priority scoring and notification

[1351] The AI ​​generates a score for the urgency of the report based on all the details, including additional information. For example, based on the information "crying every night" and "parents yelling," the AI ​​might assign an urgency score of 9 / 10. The server stores this score in a database and updates the list of reports, which are sorted by urgency.

[1352] The server will then notify the child consultation center staff that reports with a high urgency score should be handled with the highest priority. The device will then display a message such as "This is a top priority" to prompt the child consultation center staff to take action.

[1353] Get feedback

[1354] After completing a case, the child consultation center staff member will provide feedback on the situation to the server. For example, they might enter feedback such as, "I visited the site and confirmed the safety of the child." This feedback information is received by the server and stored in a database as learning data for the generation AI. This feedback continuously improves the accuracy of the system.

[1355] Examples and prompts

[1356] For example, if a user reports on the LINE app that "a child in the neighborhood is crying late at night," the device will send this to the server. The server will then ask the generation AI to analyze the report and extract the keywords "late at night" and "crying." The generation AI will then generate a follow-up question, "How often does the child cry?", and send it to the user. The user will respond with "The child cries late every night, and their parents are constantly yelling at them," and the device will again send this to the server. The generation AI will then rate the urgency score as 9 / 10, and the server will store this score. The server will then issue a notification based on this information, allowing a child consultation center staff member to respond promptly.

[1357] Example prompt sentence:

[1358] User: "I'm worried because my neighbor's child is crying late at night."

[1359] Generator: "Thank you for reporting this. How often do you cry?"

[1360] In this way, the system can quickly and accurately analyze the contents of the report and take appropriate action.

[1361] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1362] Step 1:

[1363] The user opens the LINE app and enters an abuse report. For example, they can enter "I'm worried because a child in my neighborhood is crying late at night" and send it. The entry is completed by pressing the send button on the LINE app.

[1364] Step 2:

[1365] The device receives a report message from the LINE app. The received message is converted into text data and sent to the server. This process is performed automatically, and the report data is transferred to the server in real time. The input data is the user's message, and the output data is transferred to the server as text data.

[1366] Step 3:

[1367] The server passes the received text data to a generative AI model. The generative AI model (e.g., GPT-4) uses natural language processing technology to analyze the report content. The input data is the report content in text format, and the output data is important keywords and phrases. For example, keywords such as "late at night," "crying," and "anxiety" are extracted.

[1368] Step 4:

[1369] If the report is unclear or additional information is required, the AI ​​automatically generates follow-up questions. For example, it generates specific questions such as, "Please tell me more about it. How often is the person crying?" The input data is the vague report, and the output data is a specific question.

[1370] Step 5:

[1371] The server sends the generated follow-up question to the user as a LINE message, allowing the user to provide additional information. The input data is the generated question, and the output data is the LINE message sent to the user.

[1372] Step 6:

[1373] The user receives additional questions and answers them with detailed information in the LINE app. For example, the user might enter an answer such as, "I cry late every night, and my parents keep yelling at me." The input data is the user's answer, and the output data is the message sent again via the LINE app.

[1374] Step 7:

[1375] The terminal again receives a reply message from the user and transmits it as text data to the server. The input data is a message containing the user's detailed information, and the output data is the text data that is again transmitted to the server.

[1376] Step 8:

[1377] The generation AI scores the urgency of the call based on all detailed information, including additional information. The input data is text data containing detailed information, and the output data is an urgency score. For example, it takes into account information such as "crying every night" and "parents yelling" to generate an urgency score of 9 / 10.

[1378] Step 9:

[1379] The server saves the urgency score obtained from this generation AI in a database and updates the list of reports. The list of reports is sorted by urgency. The input data is the urgency score, and the output data is the updated list of reports.

[1380] Step 10:

[1381] The server displays the list of reports sorted by urgency to the child consultation center staff. The input data is a list of reports sorted by urgency, and the output data is a display screen that can be viewed by the child consultation center staff.

[1382] Step 11:

[1383] The terminal sends this response instruction as a notification to the person in charge at the child consultation center. For example, it sends a message such as "Response is required as a top priority." The input data is the response instruction, and the output data is the notification message.

[1384] Step 12:

[1385] After completing the response, the child consultation center staff member will feed back the results to the server. For example, they will input the response result such as "We visited the site and confirmed the safety of the child." The input data is the response result, and the output data is the feedback data to the server.

[1386] Step 13:

[1387] The server receives the feedback information and stores it in a database. This feedback information is used as training data for the generative AI. The input data is the feedback information, and the output data is stored as training data.

[1388] (Application example 1)

[1389] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1390] The purpose of this invention is to effectively receive reports of abuse, crime, disasters, etc., quickly and accurately determine the urgency of the report, and instruct the relevant authorities to take appropriate action. In particular, the goal is to realize a quick and accurate response and improve the safety of local communities by automating the detailed analysis of the report content and the scoring of the urgency based on that analysis.

[1391] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1392] In this invention, the server includes: [means for receiving a report via a communication means; [analysis means including a generation AI for analyzing the content of the received report;] [means for scoring the urgency of the report based on the analysis results;] [means for instructing the necessary response based on the score;] [means for feeding back the response results; and [means for automatically notifying relevant organizations according to the urgency.] This enables detailed analysis of the report content and advanced urgency scoring, allowing relevant organizations to respond quickly and appropriately.

[1393] "Means of communication"

[1394] A device is a means for receiving reports from users and forwarding them to a server, and specifically includes smartphones and chat apps.

[1395] "Generative AI that analyzes received reports"

[1396] is an artificial intelligence used to analyze the content of reports, and uses natural language processing technology to extract important keywords and phrases from the report content.

[1397] "Score the urgency of the report based on the analysis results"

[1398] This involves evaluating the urgency of the report content analyzed by the generative AI and assigning a numerical score.

[1399] "Instruct the necessary action"

[1400] This means instructing relevant agencies to take prompt and appropriate action depending on the urgency score.

[1401] "Provide feedback on the results of the response"

[1402] This means reporting the results of the response to the server and using them as learning data for the generative AI.

[1403] "Automatically notify relevant organizations depending on the level of urgency."

[1404] This means automatically sending notifications to relevant agencies (police, fire department, ambulance, etc.) depending on the urgency of the report.

[1405] "Generative AI that engages in chat-style conversations with users to gather details about reports"

[1406] is a generative AI designed to collect detailed information about report content through dialogue with users, and has the ability to generate questions and analyze user responses.

[1407] This invention is a system that effectively receives reports of abuse, crime, disasters, etc., analyzes them using a generative AI model, and scores the urgency of the report. This system includes a means for receiving reports via a communication means, an analysis means including a generative AI that analyzes the content of the received report, a means for scoring the urgency of the report based on the analysis results, a means for instructing the necessary response based on the score, a means for providing feedback on the response results, and a means for automatically notifying relevant organizations according to the urgency. Specific embodiments of the system are described below.

[1408] Report acceptance

[1409] Users use their smartphones to report situations such as abuse, crime, and disasters. Reports are sent via LINE or the app's own chat system. For example, users can send specific details such as, "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was behaving very suspiciously."

[1410] Receiving and analyzing reports

[1411] The server transfers the text data received from the user to the generation AI to analyze the report content. The generation AI uses natural language processing technology to extract important keywords and phrases from the report content. The generation AI model used is GPT-4 (a model from OpenAI). For example, keywords such as "suspicious person," "park," and "suspicious behavior" are extracted.

[1412] Urgency Scoring

[1413] The server scores the urgency of the call based on the data analyzed by the AI ​​generator. Scoring is based on a comprehensive assessment of the importance of each keyword and phrase. Urgency is expressed as a score from 1 to 10, and for example, an urgency score of 8 / 10 is assigned based on information such as "suspicious person," "late at night," and "remote location."

[1414] Instructions on necessary actions

[1415] The server then instructs the necessary response based on the urgency score. For example, if the urgency is high, a notification is sent immediately to the relevant authorities (police, fire department, etc.) and they are instructed to take the appropriate action. Notification methods include SMS, email, and a dedicated notification app.

[1416] Feedback of response results

[1417] The server receives the response results from the relevant agencies. For example, they may send feedback such as, "We arrived at the scene and confirmed the identity of the suspicious person." The results are stored in a database and used as learning data for the generative AI. This feedback continuously improves the system's analysis accuracy and scoring accuracy.

[1418] Specific examples

[1419] Here are some example prompts for the AI ​​generator:

[1420] Please analyze the following message for emergency context and urgency:

[1421] "Last night, I saw a suspicious person in the park near my house. He suddenly ran off and was acting very suspiciously."

[1422] In this way, detailed and accurate analysis of the report content becomes possible, and appropriate responses are taken, improving the safety of the local community.

[1423] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1424] Step 1:

[1425] The user uses a smartphone to input the report details as a text message. For example, the user might type, "Last night, I saw a suspicious person in a park near my house. He suddenly ran away and was behaving very suspiciously," and then press the send button. The input data is the text message entered by the user. The output data is the text message that was sent.

[1426] Step 2:

[1427] The device (smartphone) sends the report content to the server using a communication method (LINE or the app's own chat system). The input data is the text message sent by the user in step 1. The output data is the text message sent to the server.

[1428] Step 3:

[1429] The server passes the received text message to the generation AI. Specifically, it uses a natural language processing model such as GPT-4 to generate a prompt that analyzes the text message. The input data is the text message of the received report. The output data is the prompt text that is sent to the generation AI. For example, the prompt text might be "Please analyze the following message for emergency context and urgency: 'Last night, I saw a suspicious person in the park near my house. He suddenly ran away and was behaving very suspiciously.'"

[1430] Step 4:

[1431] The generation AI analyzes the report content based on the sent prompt text and extracts important keywords and phrases. The input data is the prompt text and the report content text message. The output data is the analyzed keywords and phrases. For example, extracted words include "suspicious person," "park," and "suspicious behavior."

[1432] Step 5:

[1433] The server scores the urgency of the report based on the keywords and phrases obtained from the generation AI. Specifically, it evaluates the importance of each keyword and phrase and generates an overall score. The input data are the keywords and phrases extracted from the generation AI. The output data is the urgency score. For example, an urgency score of 8 / 10 is output.

[1434] Step 6:

[1435] The server instructs the necessary response based on the urgency score and automatically sends a notification to the relevant organizations. Notification methods include SMS, email, and a dedicated notification app. The input data is the urgency score, and the output data is the notification message sent to the relevant organizations. For example, a notification such as "High urgency report: Suspicious person spotted in park" may be sent.

[1436] Step 7:

[1437] After the relevant organizations have completed their response, they will feed back the results to the server. For example, the server may report something like, "We have arrived at the scene and confirmed the identity of the suspicious person." The input data is the feedback message from the relevant organizations. The output data is the feedback information that is stored in the database.

[1438] Step 8:

[1439] The server stores the feedback information in a database and uses it as learning data for the generation AI. This improves the accuracy of report analysis and scoring from the next time onwards. The input data is the feedback information, and the output data is the updated results of the analysis model used as learning data.

[1440] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1441] The present invention relates to a system that efficiently receives reports of abuse, analyzes them using a generative AI, scores the urgency of the reports, and also recognizes the user's emotions using an emotion engine, and reflects this information in the analysis and scoring of the reports. The following describes an embodiment of the present invention.

[1442] System Configuration

[1443] The system mainly consists of the following components:

[1444] 1. A user interface for reporting via communication means (e.g., LINE app).

[1445] 2. A generation AI that allows the server to receive and analyze the report content.

[1446] 3. A function for the emotion engine to analyze the user's emotional state.

[1447] 4. Storage for the database to store report content and analysis results.

[1448] 5. A function that allows the notification method to notify the child consultation center of the priority of the report based on the scoring results.

[1449] Report acceptance

[1450] Users report abuse using communication tools such as LINE, sending specific information such as "I heard a child in the neighborhood crying late into the night."

[1451] The terminal receives the message from the user and transmits it to the server as text data. The terminal automatically performs this reception process and immediately transfers the message data to the server.

[1452] Scrutiny of the report contents

[1453] The server passes the received report content to the generation AI. The generation AI uses natural language processing technology to analyze the report content and extract important keywords and phrases. For example, from the above report content, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[1454] emotion recognition

[1455] The emotion engine analyzes emotions from the content of messages sent by users. The emotion engine identifies emotional states (e.g., tension, anxiety, anger, etc.) from the user's language and context, as well as from keyword analysis.

[1456] Collecting detailed information about reports

[1457] The generative AI takes into account the results of the emotion engine and generates appropriate questions for the user if the report is unclear or if additional information is required. The questions are tailored to the user's emotional state. For example, if the user is nervous, the AI ​​will choose questions that will minimize stress.

[1458] The server sends the generated question to the user as a LINE message. If the user answers with additional information, the device sends this message back to the server. For example, specific information such as "I cry late every night and my parents keep yelling at me" can be added.

[1459] Priority Scoring

[1460] The generation AI scores the urgency of the report based on all detailed information and the results of the emotion engine. The urgency is determined comprehensively by evaluating the importance of each keyword and phrase and the user's emotional state. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[1461] The server stores the score obtained from the AI ​​generation in a database and adds it to a list of reports, which child consultation center staff can use to respond quickly and prioritize reports.

[1462] Generate response instructions

[1463] The list of reports is sorted by score, and the server displays the list of reports in descending order of score to the child consultation center staff. The staff responds to the reports with the highest urgency in the order displayed.

[1464] The device then sends this instruction to the child consultation center staff as a notification, displaying a message such as "This requires top priority."

[1465] Get feedback

[1466] After the child consultation center staff completes the response, they will feed back the results to the server. For example, the response results will be reported to the server as "We visited the site and confirmed the safety of the child."

[1467] The server uses this feedback as training data for the generative AI and emotion engine to continuously improve the accuracy of the system. In this way, the system enables the efficient operation of child consultation centers and strengthens efforts to protect children from abuse.

[1468] The processing flow will be explained below.

[1469] Step 1:

[1470] Users send abuse reports using LINE, providing specific information such as "I heard a child in the neighborhood crying late into the night."

[1471] Step 2:

[1472] The device receives a notification message from the user. The received message is processed as text data in JSON format or similar.

[1473] Step 3:

[1474] The device sends the received text data to the server. The data is transferred to the server using a protocol such as an HTTP POST request.

[1475] Step 4:

[1476] The server passes the received report content to the generation AI, which then analyzes the report content using natural language processing technology.

[1477] Step 5:

[1478] The AI ​​generator analyzes the content of the call and extracts important keywords and phrases, such as "late at night," "crying," and "neighborhood children."

[1479] Step 6:

[1480] The emotion engine analyzes the received message and identifies the user's emotional state, which can be categorized as tension, anxiety, anger, etc.

[1481] Step 7:

[1482] The generative AI takes the analysis results of the emotion engine and generates additional questions for the user if the report content is unclear. Example question: "Please tell me more about this. How often do you cry?"

[1483] Step 8:

[1484] The server sends the generated AI question to the user, and notifies the user of the question via LINE message.

[1485] Step 9:

[1486] The user responds with additional information, for example, "I cry late into the night every night and my parents keep yelling at me."

[1487] Step 10:

[1488] The device receives additional information from the user and sends it to the server again. A new message is forwarded to the server in JSON format or similar.

[1489] Step 11:

[1490] The server again passes new information to the generation AI, which continues to analyze the report. The generation AI then performs a detailed analysis based on the new information.

[1491] Step 12:

[1492] The AI ​​generator performs a final analysis of the report and generates an urgency score, taking into account the results of the emotion engine. For example, it may rate the urgency as 9 / 10.

[1493] Step 13:

[1494] The server saves the score received from the generated AI in a database and records additional information based on the score in the report list.

[1495] Step 14:

[1496] The server displays a list of reports in descending order of score to the child consultation center staff, who can then determine the priority of the reports based on the scores.

[1497] Step 15:

[1498] The device records and notifies the child consultation center staff, for example, by notifying them that "there is a report that requires top priority."

[1499] Step 16:

[1500] The child consultation center staff will take action and provide feedback on the results to the server, such as reporting, "We visited the site and confirmed the safety of the child."

[1501] Step 17:

[1502] The server receives the feedback data and stores it as learning data for the generative AI and emotion engine, which improves the system's analysis accuracy.

[1503] Example 2

[1504] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1505] Conventional abuse reporting systems have difficulty accurately determining the urgency of reports, which can lead to delayed responses. Furthermore, because they analyze reports without taking the user's emotions into account, there is a risk of incorrectly assessing the urgency of the report. Furthermore, conventional systems ignore the user's emotional state when collecting details about the report, which can discourage users from providing additional information. A system that solves these problems and processes abuse reports more efficiently and accurately is needed.

[1506] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1507] In this invention, the server includes: [means for receiving a report via a communication means;] [analysis means including a generation AI for analyzing the content of the received report; and] [emotion recognition engine means for analyzing the emotional state of the user.] This makes it possible [to accurately score the urgency of the report based on the results of the generation AI and the emotion recognition engine, and to promptly instruct the necessary response].

[1508] The "communication means" is an interface for receiving a message from a user and transferring it to the server.

[1509] "Generative AI" is an artificial intelligence system that includes technology to analyze the content of received reports and extract important keywords and phrases.

[1510] "Analysis means" is a system component that has the function of analyzing the received report content using generation AI.

[1511] An "emotion recognition engine" is a system that includes technology to analyze the emotional state of a user's report and identify emotions such as anxiety, tension, and anger.

[1512] The "means for scoring urgency" is a system component that has the function of quantifying and evaluating the urgency of the report content based on the analysis results of the generation AI and emotion recognition engine.

[1513] The "means for instructing the necessary response" is an interface for instructing the person in charge on the appropriate response based on the urgency score of the report.

[1514] "Feedback means" refers to a system component that has a process and function for reporting the response results to the system.

[1515] "Generative AI that interacts in chat format" is a system that includes generative AI technology for collecting detailed information about the report content through dialogue with the user.

[1516] This invention is a system that not only efficiently accepts abuse reports, analyzes them using a generative AI, and scores the urgency of the reports, but also recognizes the user's emotions using an emotion engine and reflects this information in the analysis and scoring of the reports. The following describes in detail the embodiments of the invention.

[1517] System Configuration

[1518] The system mainly consists of the following components:

[1519] 1. A user interface for reporting via communication means (e.g., messaging app).

[1520] 2. A generation AI that allows the server to receive and analyze the report content.

[1521] 3. A function for the emotion engine to analyze the user's emotional state.

[1522] 4. Storage for the database to store report content and analysis results.

[1523] 5. A function for notification methods to notify the priority of reports based on scoring results.

[1524] Hardware and software used

[1525] Use a messaging app (e.g., LINE app) as a means of communication.

[1526] For generative AI, we use Python's NLTK library.

[1527] The emotion engine uses the Python DeepMoji library.

[1528] The server requires a machine to analyze, store, and notify data (e.g., a cloud server).

[1529] The database can be an RDBMS such as MySQL or PostgreSQL.

[1530] Report acceptance

[1531] Users report abuse using a messaging app, for example by sending a message like, "I heard a child in the neighborhood crying late into the night." The device receives this report and sends it to the server as text data.

[1532] Scrutiny of the report contents

[1533] The server passes the received report content to the generation AI, which uses Python's NLTK library to analyze the report content and extract important keywords and phrases.

[1534] emotion recognition

[1535] The emotion engine analyzes emotions from the content of user reports. The emotion engine uses Python's DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the content of reports.

[1536] Priority Scoring

[1537] The AI ​​generator scores the urgency of the call based on the detailed information and the results of the emotion engine. The server saves the generated urgency score in a database and adds it to the call list.

[1538] Collecting detailed information about reports

[1539] The generation AI considers the results of the emotion engine and asks the user additional questions if necessary. For example, it collects specific information such as, "I cry late every night, and my parents keep yelling at me." The server sends the generated questions to the user as messages in a messaging app and receives additional information from the user.

[1540] Generate response instructions

[1541] The list of calls is sorted by urgency score, and the server displays the calls with the highest score to the person in charge. The person in charge responds to the calls with the highest urgency first. The device then sends a notification to the person in charge, such as "Emergency response required."

[1542] Get feedback

[1543] After the person in charge completes the response, the result is fed back to the server. The response result (e.g., "We visited the site and confirmed the safety of the child") is reported to the server. The server uses this feedback as learning data for the generative AI and emotion engine.

[1544] Prompt Sentence Examples

[1545] Here are some examples of prompts to input to a generative AI model:

[1546] You are the AI ​​generator responsible for assessing the urgency level. Please rate the urgency score based on the report content and emotional state below:

[1547] Report: "I heard a child in the neighborhood crying late at night."

[1548] Emotional state: tension, anxiety

[1549] Please rate this report's urgency score from 0 to 10 and explain your reasoning.

[1550] In this way, the system efficiently carries out a series of processes from receiving reports to analysis, scoring, and feedback, helping child consultation centers respond quickly and accurately.

[1551] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1552] Step 1:

[1553] Users use a messaging app to send abuse reports, typing a specific text message such as "I heard a child in the neighborhood crying late at night," and then hitting send.

[1554] Input: The notification message entered by the user.

[1555] Output: Sending signal from the Messages app.

[1556] What happens: The messaging app receives a text message and generates a signal to send.

[1557] Step 2:

[1558] The terminal receives the report from the user and transfers it to the server as text data.

[1559] Input: Outgoing signal from the Messages app.

[1560] Output: Text data to send to the server.

[1561] Specific operation: The message is converted to text format using the messaging app's API and sent to the server.

[1562] Step 3:

[1563] The server passes the received report content to the generation AI, which then analyzes the report content using Python's NLTK library and extracts important keywords and phrases. For example, keywords such as "late at night," "crying," and "neighborhood children" are extracted.

[1564] Input: Report text data sent from the terminal.

[1565] Output: Parsed keywords and phrases.

[1566] Specific operation: The report content is input into the generation AI, and natural language processing is performed using the NLTK library to extract important keywords.

[1567] Step 4:

[1568] The emotion recognition engine analyzes the emotions of users' reports, using the Python DeepMoji library to identify emotions such as "tension," "anxiety," and "anger" from the report content.

[1569] Input: Report text data.

[1570] Output: Parsed emotional state.

[1571] Specific operation: Analyzes text data using the DeepMoji library and determines emotional state.

[1572] Step 5:

[1573] The server scores the urgency of the report based on the results of the generation AI and emotion recognition engine. For example, an urgency score of 9 / 10 is generated based on information such as "crying every night," "parents yelling," and "user's high level of tension."

[1574] Input: Keywords extracted by the generative AI and emotional states from the emotion recognition engine.

[1575] Output: Urgency score.

[1576] Specific operation: The analysis results of the generative AI and emotion recognition engine are integrated, and an algorithm is executed to score the urgency level.

[1577] Step 6:

[1578] The server stores the generated urgency score in a database and adds it to a list of reports, which are sorted by score and made available to personnel in charge of handling them according to priority.

[1579] Input: Urgency score.

[1580] Output: Results saved in database and sorted report list.

[1581] Specific operation: Save the urgency score in the database and sort the report list by urgency score.

[1582] Step 7:

[1583] The server notifies the person in charge of high-priority calls based on the urgency score.

[1584] Input: A sorted list of notifications.

[1585] Output: Notification message to the person in charge.

[1586] Specific operation: Notify the person in charge of a high-urgency report and display a message such as "Emergency response required."

[1587] Step 8:

[1588] After the person in charge completes the response, they will feed back the results to the server. For example, they may report to the server that they have visited the site and confirmed that the child is safe.

[1589] Input: Report of response results.

[1590] Output: Feedback content is registered in a database.

[1591] Specific operation: The results reported by the person in charge are stored in a database and used as learning data for the generative AI and emotion recognition engine.

[1592] (Application example 2)

[1593] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1594] Conventional emergency call systems often lack sufficient analysis of call content and scoring of urgency, resulting in delayed appropriate responses. Additionally, because scoring is done without taking into account the user's emotional state, it is difficult to accurately assess the urgency, leading to issues such as inappropriate responses in tense situations.

[1595] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for receiving a report via a communication means, an analysis means including a generation AI that analyzes the received report content, a means for scoring the urgency of the report based on the analysis results, a means for instructing a necessary response based on the score, a means for providing feedback on the response results, an analysis means including an emotion recognition engine that analyzes the user's emotional state, and a means for extracting important information from the report content using the generation AI. This enables detailed analysis of the report content and accurate scoring of the urgency taking into account the user's emotional state. Furthermore, by promptly instructing an appropriate response and utilizing the response results as feedback, the accuracy and speed of the system can be improved.

[1596] "Communication means" refers to the interface or application used to receive notifications.

[1597] "Generation AI" is an artificial intelligence that analyzes the content of received reports.

[1598] "Analysis means" refers to functions for analyzing the content of reports, including generative AI and emotion recognition engines.

[1599] The "scoring means" is a function that scores the urgency of a report based on the analysis results.

[1600] The "instruction means" is a function that instructs the necessary response based on the score.

[1601] "Feedback means" is a function that collects the response results and feeds them back to the system.

[1602] An "emotion recognition engine" is a system for analyzing a user's emotional state.

[1603] "Means for extracting important information" refers to a function that uses generation AI to extract important keywords and phrases from the report content.

[1604] The "database" is a storage system for saving analyzed report content and scoring results.

[1605] This invention is a system that efficiently receives abuse and emergency calls, analyzes them using generative AI, and scores the urgency of the calls. It also features an emotion recognition engine that recognizes the user's emotions and reflects that information in the analysis and scoring of calls.

[1606] System Configuration

[1607] The system mainly consists of the following components:

[1608] 1. A user interface (such as a smartphone application) for receiving notifications via communication means.

[1609] 2. The analysis means, including the generation AI, analyzes the received report.

[1610] 3. A function that allows the emotion recognition engine to analyze the user's emotional state.

[1611] 4. The scoring means scores the urgency of the call based on the analysis results.

[1612] 5. The instruction tool will indicate the necessary action based on the score.

[1613] 6. The feedback means acquires the response results and feeds them back to the system.

[1614] 7. The database is a storage system for storing analysis results and report contents.

[1615] explanation

[1616] 1. The user makes an emergency call using the app via a communication method, inputting the call contents by voice or text.

[1617] Example: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared."

[1618] 2. The server receives the report and analyzes it using a generation AI, which uses natural language processing technology to extract important information and keywords.

[1619] 3. Analyze the user's emotional state using an emotion recognition engine. The emotion engine not only analyzes keywords but also identifies the user's emotional state (e.g., tension, anxiety, anger, etc.) from the user's vocabulary and context.

[1620] 4. The generation AI scores the urgency of the report based on the analyzed information and the results of the emotion recognition engine. The scoring results reflect important keywords in the report and the user's emotional state.

[1621] 5. The server then determines the necessary response based on the score. For example, if the score is high, it will automatically notify the security company or police.

[1622] Example prompt: "Please analyze the emotional state of the following text: "I saw a suspicious person in my neighborhood. He was wearing dark clothes and appeared to be carrying a weapon. I'm very scared.""

[1623] 6. After the response is completed, feedback is collected and used as training data for the generative AI and emotion recognition engine. For example, feedback such as "I visited the site and confirmed the situation" can be collected.

[1624] Hardware and software usage examples

[1625] Hardware: Servers, smartphones

[1626] Software: Google Cloud Natural Language API, database system (e.g., MySQL)

[1627] This enables detailed analysis of the report content and accurate scoring of the urgency level that takes into account the user's emotional state.In addition, by promptly instructing appropriate responses and utilizing the response results as feedback, the system's accuracy and response speed can be improved.

[1628] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1629] Step 1:

[1630] The user inputs the report details. Using a smartphone app, the user makes an emergency call by voice or text. For example, the user might input, "I saw a suspicious person in my neighborhood. He's wearing black clothes and appears to be carrying a weapon. I'm very scared."

[1631] Input: The report entered by the user (voice or text).

[1632] Output: Received notification data.

[1633] Step 2:

[1634] The terminal sends the received report data to the server. The terminal transfers the report content to the server as text data.

[1635] Input: Received notification data.

[1636] Output: Report data transferred to the server.

[1637] Step 3:

[1638] The server passes the received report data to the generation AI, which uses natural language processing technology to analyze the report content and extract important information and keywords, such as "suspicious person," "black clothing," and "weapon."

[1639] Input: Received notification data.

[1640] Output: Parsed keywords and information.

[1641] Step 4:

[1642] The server analyzes the user's emotional state using an emotion recognition engine, which recognizes the user's emotional state (e.g., fear, anxiety, tension) from the wording and context of the message.

[1643] Input: Received notification data.

[1644] Output: Emotion analysis results (e.g. fear index, anxiety index).

[1645] Step 5:

[1646] The generation AI combines the analysis results and emotion recognition results to score the urgency of the call. For example, it may assign a score of "urgency 8 / 10" based on the call content and emotion analysis results.

[1647] Input: Analyzed keywords and information, sentiment analysis results.

[1648] Output: Urgency score.

[1649] Step 6:

[1650] The server will then instruct the necessary response based on the urgency score. For example, if the urgency score is high, the server will automatically send the report to the police or a security company.

[1651] Input: Urgency score.

[1652] Output: Instructions for the required action.

[1653] Step 7:

[1654] After the response is completed, the response results are collected through feedback channels, such as "The police arrived at the scene and assessed the situation."

[1655] Input: The match result.

[1656] Output: Collected feedback data.

[1657] Step 8:

[1658] The server uses the collected feedback data as training data for the generative AI and emotion recognition engine, which improves the accuracy of the system.

[1659] Input: Collected feedback data.

[1660] Output: Updated training data.

[1661] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1662] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1663] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1664] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1665] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1666] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1667] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1668] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1669] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1670] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1671] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1672] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1673] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1674] 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.

[1675] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1676] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1677] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1678] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1679] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1680] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1681] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1682] The following is further disclosed regarding the above embodiment.

[1683] (Claim 1)

[1684] [Means for receiving notifications through communication means;

[1685] [Analysis means including a generation AI that analyzes the received report content],

[1686] [Means for scoring the urgency of the report based on the analysis results;

[1687] [Means of indicating necessary actions based on the score;

[1688] [Means for providing feedback on the results of the response,

[1689] A system including:

[1690] (Claim 2)

[1691] Including [generative AI that engages in chat-style conversations with users to gather details about reports].

[1692] 10. The system of claim 1.

[1693] (Claim 3)

[1694] [Including the means to store the report analyzed by the generating AI in a database and use it as learning data,

[1695] 10. The system of claim 1.

[1696] (Claim 4)

[1697] [The communication means includes means for receiving notifications via an internet connection,

[1698] 10. The system of claim 1.

[1699] (Claim 5)

[1700] Including [a generative AI that generates additional questions when the content of a received report is unclear]

[1701] 10. The system of claim 1.

[1702] "Example 1"

[1703] (Claim 1)

[1704] [Means for receiving notifications through communication means;

[1705] [Analysis means including a generation AI that analyzes the received report content],

[1706] [Means for scoring the urgency of the report based on the analysis results;

[1707] [Means of indicating necessary actions based on the score;

[1708] [Means for providing feedback on the results of the response,

[1709] [Means for the generation AI to generate additional questions when the content of the report is unclear and send them to the user via communication means;

[1710] [Means for storing the urgency scores of reports in a database and managing the list of reports based on response priority;

[1711] A system including:

[1712] (Claim 2)

[1713] The system of claim 1, including a generation AI that interacts with the user in a chat format to gather details about the report.

[1714] (Claim 3)

[1715] [The system according to claim 1, further comprising means for storing the report content analyzed by the generating AI in a database and using it as learning data.

[1716] "Application Example 1"

[1717] (Claim 1)

[1718] [Means for receiving notifications through communication means;

[1719] [Analysis means including a generation AI that analyzes the received report content],

[1720] [Means for scoring the urgency of the report based on the analysis results;

[1721] [Means of indicating necessary actions based on the score;

[1722] [Means for providing feedback on the results of the response,

[1723] [Means for automatically notifying relevant organizations depending on the level of urgency;

[1724] A system including:

[1725] (Claim 2)

[1726] Including [generative AI that engages in chat-style conversations with users to gather details about reports].

[1727] 10. The system of claim 1.

[1728] (Claim 3)

[1729] [Including the means to store the report analyzed by the generating AI in a database and use it as learning data,

[1730] 10. The system of claim 1.

[1731] "Example 2: Combining Emotion Engines"

[1732] (Claim 1)

[1733] [Means for receiving notifications through communication means;

[1734] [Analysis means including a generation AI that analyzes the received report content],

[1735] [an emotion recognition engine means for analyzing the user's emotional state;

[1736] [Means for scoring the urgency of the call based on the results of the generative AI and emotion recognition engine;

[1737] [Means of indicating necessary actions based on the score;

[1738] [Means for providing feedback on the results of the response,

[1739] A system including:

[1740] (Claim 2)

[1741] [Generative AI that interacts with users in chat to gather details about the report, taking into account the analysis results of the emotion recognition engine]

[1742] 10. The system of claim 1.

[1743] (Claim 3)

[1744] [Including the means to store the report content analyzed by the generation AI and the analysis results of the emotion recognition engine in a database and use them as learning data,

[1745] 10. The system of claim 1.

[1746] "Application example 2 when combining emotion engines"

[1747] (Claim 1)

[1748] [Means for receiving notifications through communication means;

[1749] [Analysis means including a generation AI that analyzes the received report content],

[1750] [Means for scoring the urgency of the report based on the analysis results;

[1751] [Means of indicating necessary actions based on the score;

[1752] [Means for providing feedback on the results of the response,

[1753] [analysis means including an emotion recognition engine for analyzing the user's emotional state];

[1754] [Method of extracting important information from report content by the generation AI],

[1755] A system including:

[1756] (Claim 2)

[1757] The system of claim 1, including a generation AI that interacts with the user in a chat format to gather details about the report.

[1758] (Claim 3)

[1759] [The system according to claim 1, further comprising means for storing the report content analyzed by the generating AI in a database and using it as learning data. [Explanation of symbols]

[1760] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a notification via a communication means; An analysis means including a generation AI that analyzes the received report content; A means for scoring the urgency of the report based on the analysis results; A means for indicating necessary actions based on the score; A means of providing feedback on the results of the response; A system including:

2. Including a generative AI that interacts with users in chat to gather details about the report. The system of claim 1 .

3. Including a means to store the report content analyzed by the generation AI in a database and use it as learning data. The system of claim 1 .

4. the communication means including means for receiving notifications via an internet connection; The system of claim 1 .

5. Including a generative AI that generates follow-up questions when the content of a received report is unclear. The system of claim 1 .

Images