system

Abstract

We provide the system. [Solution] A means of receiving text data containing personal information at a data input device, A means of using an information processing device that detects personal information contained in the aforementioned text data, A data processing means for removing or anonymizing and transforming the detected personal information, A means for inputting the converted data into a generation model and executing a predetermined task, Means for transmitting the output of the generation model to a data input device, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When utilizing generative AI, there is a problem that data input is restricted due to the protection of personal information and legal regulations. This problem is particularly prominent in the utilization of data involving personal information and communication secrets, and has become a factor hindering the practical application of generative AI. In addition, the operation of manually removing personal information requires time and labor and may require specialized knowledge. Due to these problems, it has become difficult to maximize the advantages of generative AI in many fields.

Means for Solving the Problems

[0005] This invention provides a system for automatically processing input data containing personal information. Specifically, it automatically detects text data containing personal information received by a data input device using an information processing device, and then processes it to remove or anonymize the personal information. This converts the data into a format suitable for a generative model, enabling the generative AI to perform predetermined tasks. Furthermore, by using a machine learning model in the information processing device, the identification of personal information is achieved with high accuracy, enabling the generative model to perform natural language processing tasks. This allows users to effectively utilize generative AI technology while complying with legal regulations.

[0006] A "data input device" is a device that receives and processes text data entered by a user.

[0007] "Text data" refers to a data format that contains character information, and is usually expressed as sentences or messages.

[0008] An "information processing device" is a device that has the function of analyzing, transforming, and processing data, and is particularly used for detecting personal information.

[0009] "Personal information" refers to information that can identify a specific individual, and includes names, addresses, telephone numbers, etc.

[0010] "Anonymization" refers to the process of removing or altering personal information so that a specific individual cannot be identified.

[0011] A "generative model" is an AI model that performs a specific task based on input data.

[0012] A "machine learning model" is an algorithm or system that learns from data and uses it to analyze and predict new data.

[0013] A "natural language processing task" refers to a series of operations and functions that use AI technology to understand and process human language. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, a labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention is implemented as a system for safely and efficiently processing data containing personal information. The system aims to automatically remove personal information from text data entered by the user and process the converted data using a generating AI.

[0036] The user first inputs text data using a data entry device. This data may include personal information such as customer information and inquiry details. The terminal receives the data entered by the user and transmits it to the server via the network.

[0037] The server passes the received data to an information processing device. This device has the capability to automatically detect personal information contained in the data using a machine learning model. The identified personal information is removed or anonymized so that the data cannot identify individuals. This process ensures the protection of data privacy.

[0038] The anonymized data is then fed into a generative model. This generative model is an AI model designed to perform natural language processing tasks, such as generating responses to user inquiries or creating summaries of long texts. The results from the generative model are then aggregated back onto the server.

[0039] The server sends the generated output results back to the terminal. By checking this output on their own terminal, users can obtain results that utilize the generated AI while protecting their personal information.

[0040] As a concrete example, consider a system for automatically processing customer inquiries in customer support. When a customer submits an inquiry containing personal information as feedback, this system first removes the personal information and then generates an automated response. This series of processes enables safe and rapid customer support.

[0041] Thus, the system based on the present invention makes it possible to maximize the convenience of generated AI while ensuring user privacy.

[0042] The following describes the processing flow.

[0043] Step 1:

[0044] Users enter text data that may contain personal information into a data entry device. The entered data is accepted as inquiries or customer feedback.

[0045] Step 2:

[0046] The terminal sends the user's input data to the server in its original form. During this process, the data is transferred via a secure protocol over the network.

[0047] Step 3:

[0048] The server passes the received data to the information processing unit. Here, the information processing unit uses a machine learning model to automatically identify personal information within the data. It detects identifiable information such as names, addresses, and phone numbers.

[0049] Step 4:

[0050] The data processing device removes or converts detected personal information into specific labels. This anonymizes the data, transforming it into a form that does not identify individuals.

[0051] Step 5:

[0052] The server sends anonymized data to a generative AI model. The generative AI model then uses the received data to perform predetermined tasks, including generating replies and creating summaries.

[0053] Step 6:

[0054] Once the generating AI model completes its processing, it returns the results to the server. The generated results do not contain any personal information.

[0055] Step 7:

[0056] The server sends the results obtained from the generated AI model to the terminal. The terminal presents the results to the user, who can then review the information.

[0057] Step 8:

[0058] Based on the output received by the user, additional modifications are made as needed, or actions are taken using the results. This process allows users to enjoy the convenience of generative AI while protecting their personal information.

[0059] (Example 1)

[0060] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0061] In modern society, the handling of data containing personal information is increasing, making privacy protection a critical issue. Traditional methods involve cumbersome manual anonymization and management of personal information, creating a need for technologies that process personal information efficiently and securely. Furthermore, in data processing using generative AI models, there is a need for technologies that generate automated responses and summaries while ensuring the protection of personal information.

[0062] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0063] In this invention, the server includes means for receiving document information containing personal identification information at a data input mechanism; means for using an information processing device to detect the personal identification information contained in the document information; data processing means for removing or anonymizing and transforming the detected personal identification information; means for inputting the transformed information into a generation algorithm and executing predetermined processing; means for transmitting the output of the generation algorithm to the data input mechanism; and means for using a secure communication protocol when the data input mechanism sends and receives data. This enables the secure and efficient processing of personal identification information and the generation of automated responses and summaries that take advantage of the convenience of the generation AI model.

[0064] A "data input mechanism" is a device or system for users to input text data and a means of receiving personally identifiable information and other information.

[0065] "Personally identifiable information" refers to information used to identify a specific individual, and includes names, addresses, telephone numbers, etc.

[0066] "Document information" refers to all information expressed in text format, and may include data containing personally identifiable information.

[0067] An "information processing device" is a computer system that detects personally identifiable information from received document information and processes that information.

[0068] A "machine learning algorithm" is a method for analyzing large amounts of data, learning patterns from it, and performing specific tasks.

[0069] "Anonymization" is the process of transforming personally identifiable information in data into a state where it cannot be identified, thereby protecting individual privacy.

[0070] A "generative algorithm" is a process or method for generating new information based on input data, and is particularly used to perform natural language processing tasks.

[0071] A "secure communication protocol" refers to technical specifications and procedures designed to protect sensitive information from being leaked during data transmission.

[0072] This invention is implemented as a system for securely and efficiently processing document information containing personally identifiable information. The user first uses a data input mechanism to input text data into the system. The data input mechanism may be a keyboard or a voice input device. This input text data is transmitted from the terminal to the server via a network.

[0073] The server uses an information processing device to detect personally identifiable information within the received document information. This process utilizes machine learning algorithms. Specifically, a model employing natural language processing technology automatically identifies personally identifiable information such as names, addresses, and phone numbers.

[0074] Any personally identifiable information detected is removed or anonymized by the server. During this process, masking or other methods may be applied to prevent the information from being linked to a specific individual. This process protects individual privacy.

[0075] The anonymized data is then sent from the server to a generation algorithm. This generation algorithm is designed to perform natural language processing tasks. For example, it can generate responses to user inquiries or create summaries of long texts.

[0076] The generated responses and summaries are sent back by the server to a terminal equipped with a data input mechanism. Users can then review and utilize these responses on the terminal. Data security and processing efficiency are ensured throughout this entire process.

[0077] As a concrete example, consider a system in customer support services that automatically processes customer feedback. If a customer submits an inquiry that includes personal information, this system will first remove the personal information and then generate an appropriate automated response to the customer's question. This process ensures safe and prompt customer support.

[0078] As an example of a prompt for a generative AI model, you can use something like, "Based on the inquiry 'Please tell me how to use product X,' please create a general guide on how to use the product."

[0079] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0080] Step 1:

[0081] Users enter text data containing personally identifiable information using a data entry mechanism. This data may include names, addresses, and inquiry details. This data is captured on the device and securely transmitted to a server via the network.

[0082] Step 2:

[0083] The server receives text data sent from the terminal. Based on the received data, the information processing device operates and uses a machine learning algorithm to detect personally identifiable information within the data. The input is raw text data, and the output is data with personally identifiable information identified. At this stage, the detected personal information is verified.

[0084] Step 3:

[0085] The server processes the data to remove or anonymize personally identifiable information. Personal information is securely processed through masking or deletion, resulting in anonymized data. The input is data with identified personal information, and the output is anonymized data. This process ensures that the data is in a privacy-conscious state.

[0086] Step 4:

[0087] The server inputs anonymized data into a generative AI model. The generative model performs natural language processing based on the prompt text and generates the necessary response text, summary, etc. The input is anonymized data, and the output is the generated response text and summary text. In this step, the AI ​​model constructs information according to the user's purpose.

[0088] Step 5:

[0089] The server aggregates the output results obtained from the generative model and sends them to the terminal. The output is in a text format that the user can read, and the terminal receives this information. The user can check the generated response on the terminal screen and take the corresponding action. In this step, the entire system's feedback loop is completed and information is provided to the user.

[0090] (Application Example 1)

[0091] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0092] In online services, users want to share information with peace of mind, but concerns about privacy violations remain. Furthermore, there is a need to process data containing personal information securely and effectively, while using generative AI to improve the user experience. In particular, there is a demand to automatically remove personal information during review submission on e-commerce sites, thereby protecting user privacy while generating value-added information.

[0093] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0094] In this invention, the server includes means for receiving document information containing identification information in a data acquisition unit, means for detecting the identification information contained in the document information using an analysis device, and data processing means for removing or anonymizing and transforming the detected identification information. This makes it possible to generate value-added information using generation AI while protecting user privacy.

[0095] The "data acquisition unit" is a component for receiving document information that includes identification information.

[0096] "Identifying information" refers to information that can identify an individual, such as an individual's name and address.

[0097] "Document information" refers to all text data expressed in an electronically processable format.

[0098] "Analysis device" refers to a device or system for detecting and analyzing identification information contained in received document information.

[0099] "Detection" refers to the process of identifying or finding specific information or data.

[0100] "Anonymization" refers to the process of transforming data so that individuals cannot be identified from their identifying information.

[0101] "Data processing means" refers to methods or devices for removing or anonymizing identifying information and transforming data.

[0102] A "generative model" refers to a machine learning model used to perform language processing tasks.

[0103] "User terminal" refers to a device that a user uses to receive and display information.

[0104] A system for carrying out this invention consists of a data acquisition unit, an analysis device, a data processing means, and a generation model.

[0105] First, the user inputs document information containing identification information using a device such as a smartphone or personal computer. The user's device then transmits this document information to the server via a data acquisition unit. After receiving this information, the server uses an analysis device to detect the identification information contained in the document. Natural language processing libraries such as SpaCy are used for the analysis.

[0106] Next, the server removes or anonymizes the detected identifiers to create transformed data. A Python program is used for this data processing. The anonymized data is then input into a generative model. The generative model uses natural language processing AI such as GPT-4(registered trademark) to perform specific information processing tasks, such as sentiment analysis and summarization.

[0107] The results generated by the generative model are sent from the server to the user's device and displayed to the user. Here, the user can view value-added information while their privacy is protected.

[0108] For example, if a user posts a review saying, "I tried out the new product and I really liked it. The delivery was fast," this system will detect and remove personal information such as the user's name and address. The generating AI can then summarize the review and display it as, "I'm satisfied with the product and the delivery was fast."

[0109] An example of a prompt to input into the generating AI model is as follows: "Remove personal information from the following review and summarize it: Review content."

[0110] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0111] Step 1:

[0112] The user inputs document information, including identification information, using a terminal. The text data entered by the user is sent from the terminal to the server via a data acquisition unit. At this time, the input data is processed in text format.

[0113] Step 2:

[0114] The server passes the document information received from the data acquisition unit to the analysis unit. The analysis unit uses a natural language processing library such as SpaCy to detect identifying information within the document. The input to this step is the received document information, and the output is a list of detected identifying information. Specifically, the process involves tokenizing the document and identifying patterns that are recognized as identifying information.

[0115] Step 3:

[0116] The server removes or anonymizes the detected identification information using data processing means. In this step, a list of identification information and the original document information are used as input, and anonymized document information is obtained as output. The data transformation involves replacing the identification information with a common noun such as "User X" based on specified rules.

[0117] Step 4:

[0118] The server inputs the converted document information into a generative model. The generative model used here is a natural language processing AI such as GPT-4, which performs information processing tasks according to the specified prompt sentences. The input in this case is anonymized document information, and the output is value-added information such as generated summaries and sentiment ratings. Specifically, it generates text according to the prompt sentences and returns the results to the server.

[0119] Step 5:

[0120] The server sends the output from the generative model to the user's terminal. The user's terminal displays this output, and the user views the information. Here, the input is value-added information obtained from the generative model, and the output is information converted into a format that the user can view on their terminal. Specifically, the function displays the received data on the user's screen.

[0121] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0122] This invention is implemented as a system for securely processing text data containing personal information, and further recognizing user emotions and generating corresponding responses. This system removes personal information from user input data, analyzes emotions, and then provides responses generated by AI.

[0123] A user enters text data using a data entry device. This data typically includes personal information and content that reflects emotions. The device then sends the entered data to a server. This data is transmitted to the server as quickly as possible for immediate processing.

[0124] The server hands over the received data to the information processing unit. The information processing unit first uses a machine learning model to automatically detect and remove or anonymize personal information. This process removes personally identifiable information, thereby enhancing data security.

[0125] Next, the emotion engine analyzes the user's emotions in the input data. The emotion engine has the ability to identify various emotional states—for example, anger, joy, and sadness—and based on this, it can determine the user's current emotion. This emotional information is then reflected in the subsequent response generation process.

[0126] Anonymized data and sentiment information are passed to a generative model. The generative model uses natural language processing techniques to automatically create responses that match the user's emotions. For example, if the user expresses dissatisfaction, it generates an empathetic response; if the user is happy, it generates a positive response.

[0127] The generated response is sent to the terminal via the server. The user can then view the response on their terminal. This mechanism allows users to receive more personalized responses while protecting their personal information and ensuring security.

[0128] A concrete example is a customer support scenario. When a customer sends a complaint about a product to a data entry device, the information processing device uses an emotion engine to identify the customer's anger while removing personal information. Based on this information, the generative model constructs a response that includes an apology and a solution, enabling a quick and appropriate response to the customer. This step contributes to the appropriate use of generative AI that addresses emotions while protecting personal information.

[0129] The following describes the processing flow.

[0130] Step 1:

[0131] Users use data entry devices to input text data as inquiries or feedback. This data may reflect emotions and potentially contain personal information.

[0132] Step 2:

[0133] The terminal receives input data and sends it to the server via a secure protocol. The data is transferred to the server in its original format.

[0134] Step 3:

[0135] The server provides the received data to the information processing device. The information processing device then uses a machine learning model to automatically detect personal information from the text and processes it by removing or anonymizing specific parts.

[0136] Step 4:

[0137] The server passes the data, with personal information removed, to the emotion engine. The emotion engine analyzes the input text to identify the user's emotional state. This includes emotion labels such as anger, joy, and sadness.

[0138] Step 5:

[0139] The results of the sentiment analysis and anonymized data are fed into the generative model. The generative model then enters the process of creating a response adapted to the emotion using natural language processing techniques.

[0140] Step 6:

[0141] The response generated by the generative model is sent back to the server. The server receives the response data and prepares for the next action.

[0142] Step 7:

[0143] The server sends the generated response back to the user's terminal. The terminal receives it and displays it to the user.

[0144] Step 8:

[0145] The user reviews the display on their device and, if necessary, decides on further actions to take regarding the generated response. This process allows the user to receive a personalized response that takes their emotions into consideration.

[0146] (Example 2)

[0147] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0148] There is a need to securely process text data containing personal information while appropriately recognizing user emotions and efficiently generating responses that correspond to those emotions. However, conventional systems have struggled to balance the protection of personal information with the generation of emotion-based responses. Therefore, there is a need for technology that can perform instant emotion analysis and provide emotion-based responses while ensuring the security of personal information.

[0149] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0150] In this invention, the server includes means for receiving data containing personal information and emotional information via a data input device, means for transferring the data to an information processing device using the server, and means for inputting the data into a generation algorithm to generate an appropriate response. This makes it possible to protect personal information and provide a humane response that responds to emotions.

[0151] A "data input device" is a device used by users to input information, and typically includes a keyboard or touch panel.

[0152] A "server" is a computer device that aggregates data on a network and transfers it to a processing unit.

[0153] An "information processing device" is a device that has the function of analyzing and transforming information within data, and performs sentiment analysis in addition to identifying and removing personal information.

[0154] A "machine learning algorithm" is a mathematical method that allows computers to learn specific patterns from data and perform predictions and classifications.

[0155] "Personal information" refers to information that can identify a specific individual, including information such as name and address.

[0156] An "emotion analysis engine" is a program or device that has the function of identifying human emotions from input data and determining their emotional state.

[0157] A "generative algorithm" is a means of automatically creating an appropriate response based on input data, and primarily utilizes natural language processing.

[0158] "Anonymization" is the process of transforming personal information so that it cannot be identified, and it is a measure taken to protect data privacy.

[0159] This invention begins with a user inputting text data using a data input device. The data input device typically uses a common input interface such as a keyboard or touch panel. The data entered by the user may include personal information or emotional data. The terminal transmits this input data to a server using a communication method such as the internet.

[0160] The server accurately transfers received data to the information processing unit based on the communication protocol. The information processing unit uses programming languages ​​such as Python and machine learning algorithms, specifically models such as TENSORFLOW® and Hugging Face, to detect personal information. This information processing unit protects privacy by anonymizing identifiable information such as names and addresses in the data.

[0161] Next, the information processing device analyzes emotional information in the data using an emotion analysis engine. Here, natural language processing technology is utilized to identify the user's emotional state. The identified emotional information and anonymized data are input into a generation algorithm. This algorithm uses a generation AI model such as OpenAI's GPT series to automatically generate an appropriate response that matches the user's emotions.

[0162] The generated response is sent back to the terminal via the server. The terminal displays the generated response to the user, who can then decide on their next action based on it. This entire process makes it possible to quickly provide more humane and empathetic responses while maintaining the protection of personal information.

[0163] A concrete example is a customer support scenario. For instance, if a user enters "I'm very upset because the product didn't arrive on schedule," the server passes this information to the information processing unit. The information processing unit then uses an analysis engine to identify the emotion of "confusion," and based on this, a generative AI model generates a response such as "We apologize, we will address this immediately." This response is sent to the user's terminal via the server and displayed on the screen.

[0164] Examples of prompt statements include the following:

[0165] User comment: "I'm very upset because the product didn't arrive on schedule."

[0166] Sentiment analysis result: "confused"

[0167] Response generation prompt: "Create a response that expresses empathy for the customer's distress, apologizes, and proposes a solution."

[0168] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0169] Step 1:

[0170] The user inputs text data into a data entry device. The user uses a keyboard or touch panel to input opinions, questions, and other information about the product. This process includes input data containing personal information and emotional information. The output is the user's raw text data.

[0171] Step 2:

[0172] The terminal sends user input data to the server. Here, the terminal uses a secure communication protocol to encrypt and securely transfer the data. The input is the raw text data generated in step 1, and the output is the encrypted data received by the server.

[0173] Step 3:

[0174] The server passes the received data to the information processing unit. The server checks the data for integrity, and if there are no problems, it passes the data to the information processing unit for processing. It receives encrypted text data as input and sends the integrity-checked text data as output to the information processing unit.

[0175] Step 4:

[0176] The information processing device detects, removes, or anonymizes personal information within the data using machine learning algorithms. In this step, for example, TensorFlow is used to identify names, addresses, etc., and anonymize them, such as by displaying only the "name." The input is consistent text data, and the output is text data with the personal information anonymized.

[0177] Step 5:

[0178] An information processing device analyzes the emotions of input data using an emotion analysis engine. Natural language processing techniques are used to identify emotions such as "joy" and "anger" from the input text. The input is anonymized text data, and the output is data with identified emotions.

[0179] Step 6:

[0180] An information processing device generates an emotion-appropriate response using a generation algorithm. The generation AI model constructs prompt sentences that create an appropriate response based on identified emotion information. The input consists of emotion-identified data and prompt sentences, and the output is the generated response.

[0181] Step 7:

[0182] The server sends the generated response to the terminal. The server encrypts the output text and sends it securely to the terminal. The input is the generated response, and the output is the response data received by the terminal.

[0183] Step 8:

[0184] The user confirms the response on their device. The device displays the received response on the screen in a format that is easy for the user to understand. The input is the response sent from the server, and the output is the response that the user visually confirms.

[0185] (Application Example 2)

[0186] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0187] In recent years, services using text data have required the protection of personal information and the provision of appropriate responses that respond to user emotions. To prevent the leakage of personal information and improve the user experience, it is necessary to integrate technologies that address each of these aspects, but such technologies are still under development. This invention aims to address this challenge by providing a system that handles personal information securely while providing prompt responses that are appropriate to the user's emotions.

[0188] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0189] In this invention, the server includes means for receiving text data containing personal information from a data input device, means for using information processing means to detect personal information contained in the text data, data processing means for removing or anonymizing and transforming the detected personal information, and means for inputting the transformed data and sentiment information into a generation model and performing a predetermined task. This makes it possible to automatically generate appropriate responses according to the user's sentiment while securely protecting personal information.

[0190] A "data input device" is a hardware or software means for a user to input text data, and includes smartphones, computers, and other similar devices.

[0191] "Personal information" refers to information used to identify a specific individual, and includes names, addresses, contact information, etc.

[0192] "Information processing means" refers to a device or program for identifying and processing personal information from received data, and may utilize machine learning algorithms.

[0193] "Data processing means" refers to a device or program for removing or anonymizing detected personal information and, if necessary, transforming the data.

[0194] "Converted data" refers to text data from which personal information has been removed or anonymized and converted into a secure format.

[0195] A "generative model" is a machine learning model that uses natural language processing techniques to generate output based on input data.

[0196] An "emotion engine" is software or an algorithm used to analyze and identify a user's emotions from text data.

[0197] A "response generation means" is a device or program that automatically creates an appropriate response based on the user's emotions using a generation model.

[0198] This invention begins with a user inputting text data using a data input device. This input data may contain personal information and the user's emotions. The terminal sends this data to a server for processing.

[0199] The server passes the received data to an information processing system. This information processing system uses machine learning models to detect and remove or anonymize personal information within the text data. Common data anonymization algorithms and libraries are used as tools. This process removes information that identifies specific individuals.

[0200] Next, the emotion engine analyzes the anonymized data and extracts emotions from the text entered by the user. This emotion engine uses natural language processing techniques to identify emotional states (e.g., joy, anxiety, anger) contained in the text and quantifies or categorizes the emotions.

[0201] Anonymized data and sentiment information are input into a generative model. This generative model automatically generates responses that correspond to the user's emotions. The generated responses take the user's emotions into consideration; for example, for data indicating anxiety, it selects words that provide reassurance to construct a response.

[0202] Finally, the generated response is sent to the terminal via the server. The user can then review the response on the terminal and obtain problem resolution or an answer to their inquiry. This process ensures both the protection of personal information and the provision of emotionally responsive responses.

[0203] For example, if a user enters the text, "I've encountered a payment error. I'm worried because I don't know how to fix it," the system will identify the user's anxiety and respond with something like, "We apologize for the inconvenience. We will guide you through the steps to resolve the issue."

[0204] An example of a prompt message is: "User input: A payment error occurred. I am worried because I don't know how to resolve it. Please provide sentiment analysis and an appropriate response."

[0205] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0206] Step 1:

[0207] Users input text data using a data entry device. This text data may include personal information and emotions. The data entered by the user is received by the terminal and immediately sent to the server.

[0208] Step 2:

[0209] The server passes the received text data to an information processing system. The information processing system uses a machine learning model to identify personal information within the text and removes or anonymizes it. Specifically, it extracts identifying information such as names and addresses and deletes or replaces it from the data to generate anonymized data. Through this step, the input becomes anonymized data.

[0210] Step 3:

[0211] The server passes anonymized data to the emotion engine. The emotion engine analyzes the sentiment of the text using natural language processing techniques. In this step, it identifies emotional states (e.g., anxiety, joy) from the text and outputs the sentiment information as a number or category. The sentiment information obtained here is used in the next step.

[0212] Step 4:

[0213] The server inputs anonymized data and sentiment information into a generative model. The generative model generates an appropriate response, taking the sentiment information into account. Specifically, it selects reassuring language structures to correspond to the user's emotions (e.g., anxiety), constructs a response in natural language, and outputs it. This response is tailored to the user's emotions.

[0214] Step 5:

[0215] The server sends the generated response to the terminal. The terminal presents the received response to the user. The user can then review the response and take appropriate action based on it.

[0216] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0217] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0218] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0219] [Second Embodiment]

[0220] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0221] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0222] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0223] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0224] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0225] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0226] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0227] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0228] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0229] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0230] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0231] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0232] This invention is implemented as a system for safely and efficiently processing data containing personal information. The system aims to automatically remove personal information from text data entered by the user and process the converted data using a generating AI.

[0233] The user first inputs text data using a data entry device. This data may include personal information such as customer information and inquiry details. The terminal receives the data entered by the user and transmits it to the server via the network.

[0234] The server passes the received data to an information processing device. This device has the capability to automatically detect personal information contained in the data using a machine learning model. The identified personal information is removed or anonymized so that the data cannot identify individuals. This process ensures the protection of data privacy.

[0235] The anonymized data is then fed into a generative model. This generative model is an AI model designed to perform natural language processing tasks, such as generating responses to user inquiries or creating summaries of long texts. The results from the generative model are then aggregated back onto the server.

[0236] The server sends the generated output results back to the terminal. By checking this output on their own terminal, users can obtain results that utilize the generated AI while protecting their personal information.

[0237] As a concrete example, consider a system for automatically processing customer inquiries in customer support. When a customer submits an inquiry containing personal information as feedback, this system first removes the personal information and then generates an automated response. This series of processes enables safe and rapid customer support.

[0238] Thus, the system based on the present invention makes it possible to maximize the convenience of generated AI while ensuring user privacy.

[0239] The following describes the processing flow.

[0240] Step 1:

[0241] Users enter text data that may contain personal information into a data entry device. The entered data is accepted as inquiries or customer feedback.

[0242] Step 2:

[0243] The terminal sends the user's input data to the server in its original form. During this process, the data is transferred via a secure protocol over the network.

[0244] Step 3:

[0245] The server passes the received data to the information processing unit. Here, the information processing unit uses a machine learning model to automatically identify personal information within the data. It detects identifiable information such as names, addresses, and phone numbers.

[0246] Step 4:

[0247] The data processing device removes or converts detected personal information into specific labels. This anonymizes the data, transforming it into a form that does not identify individuals.

[0248] Step 5:

[0249] The server sends anonymized data to a generative AI model. The generative AI model then uses the received data to perform predetermined tasks, including generating replies and creating summaries.

[0250] Step 6:

[0251] Once the generating AI model completes its processing, it returns the results to the server. The generated results do not contain any personal information.

[0252] Step 7:

[0253] The server sends the results obtained from the generated AI model to the terminal. The terminal presents the results to the user, who can then review the information.

[0254] Step 8:

[0255] Based on the output received by the user, additional modifications are made as needed, or actions are taken using the results. This process allows users to enjoy the convenience of generative AI while protecting their personal information.

[0256] (Example 1)

[0257] Next, we will describe Example 1. 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."

[0258] In modern society, the handling of data containing personal information is increasing, making privacy protection a critical issue. Traditional methods involve cumbersome manual anonymization and management of personal information, creating a need for technologies that process personal information efficiently and securely. Furthermore, in data processing using generative AI models, there is a need for technologies that generate automated responses and summaries while ensuring the protection of personal information.

[0259] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0260] In this invention, the server includes means for receiving document information containing personal identification information at a data input mechanism; means for using an information processing device to detect the personal identification information contained in the document information; data processing means for removing or anonymizing and transforming the detected personal identification information; means for inputting the transformed information into a generation algorithm and executing predetermined processing; means for transmitting the output of the generation algorithm to the data input mechanism; and means for using a secure communication protocol when the data input mechanism sends and receives data. This enables the secure and efficient processing of personal identification information and the generation of automated responses and summaries that take advantage of the convenience of the generation AI model.

[0261] A "data input mechanism" is a device or system for users to input text data and a means of receiving personally identifiable information and other information.

[0262] "Personally identifiable information" refers to information used to identify a specific individual, and includes names, addresses, telephone numbers, etc.

[0263] "Document information" refers to all information expressed in text format, and may include data containing personally identifiable information.

[0264] An "information processing device" is a computer system that detects personally identifiable information from received document information and processes that information.

[0265] A "machine learning algorithm" is a method for analyzing large amounts of data, learning patterns from it, and performing specific tasks.

[0266] "Anonymization" is the process of transforming personally identifiable information in data into a state where it cannot be identified, thereby protecting individual privacy.

[0267] A "generative algorithm" is a process or method for generating new information based on input data, and is particularly used to perform natural language processing tasks.

[0268] A "secure communication protocol" refers to technical specifications and procedures designed to protect sensitive information from being leaked during data transmission.

[0269] This invention is implemented as a system for securely and efficiently processing document information containing personally identifiable information. The user first uses a data input mechanism to input text data into the system. The data input mechanism may be a keyboard or a voice input device. This input text data is transmitted from the terminal to the server via a network.

[0270] The server uses an information processing device to detect personally identifiable information within the received document information. This process utilizes machine learning algorithms. Specifically, a model employing natural language processing technology automatically identifies personally identifiable information such as names, addresses, and phone numbers.

[0271] Any personally identifiable information detected is removed or anonymized by the server. During this process, masking or other methods may be applied to prevent the information from being linked to a specific individual. This process protects individual privacy.

[0272] The anonymized data is then sent from the server to a generation algorithm. This generation algorithm is designed to perform natural language processing tasks. For example, it can generate responses to user inquiries or create summaries of long texts.

[0273] The generated responses and summaries are sent back by the server to a terminal equipped with a data input mechanism. Users can then review and utilize these responses on the terminal. Data security and processing efficiency are ensured throughout this entire process.

[0274] As a concrete example, consider a system in customer support services that automatically processes customer feedback. If a customer submits an inquiry that includes personal information, this system will first remove the personal information and then generate an appropriate automated response to the customer's question. This process ensures safe and prompt customer support.

[0275] As an example of a prompt for a generative AI model, you can use something like, "Based on the inquiry 'Please tell me how to use product X,' please create a general guide on how to use the product."

[0276] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0277] Step 1:

[0278] The user uses the data input mechanism to input text data including personal identification information. The input data may include a name, an address, the content of an inquiry, etc. This data is captured by the terminal and securely transmitted to the server via the network.

[0279] Step 2:

[0280] The server receives the text data transmitted from the terminal. Based on the received data, the information processing device operates to detect the personal identification information in the data by means of a machine learning algorithm. The input is raw text data, and the output is the state in which the personal identification information is identified. At this stage, the detected personal information is verified.

[0281] Step 3:

[0282] The server performs data processing to remove or anonymize the personal identification information. The personal information is securely processed, such as by masking or deletion, to obtain anonymized data. The input is the data in which the personal information is identified, and the output is anonymized data. By this process, the data is made to be in a state that takes privacy into consideration.

[0283] Step 4:

[0284] The server inputs the anonymized data into the generative AI model. The generative model performs natural language processing based on the prompt text and generates the necessary response text, summary, etc. The input is anonymized data, and the output is the generated response or summary text. In this step, the AI model constructs information according to the user's purpose.

[0285] Step 5:

[0286] The server aggregates the output results obtained from the generation model and transmits them to the terminal. The output is in a text format that can be confirmed by the user, and the terminal receives this information. The user can view the generated response on the screen of the terminal and take corresponding actions. In this step, the overall feedback loop of the system is completed, and information is provided to the user.

[0287] (Application Example 1)

[0288] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0289] In online services, users want to share information with confidence, but there is a problem that they are concerned about privacy violations. Also, it is necessary to improve the user experience using generative AI while safely and effectively processing data including personal information. In particular, when entering reviews on an e-commerce site, it is required to generate valuable information while protecting the privacy of users by automatically removing personal information.

[0290] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0291] In this invention, the server includes means for receiving document information including identification information in a data acquisition unit, means for detecting the identification information included in the document information using an analysis device, and data processing means for removing or anonymizing and converting the detected identification information. Thereby, while protecting the privacy of users, it becomes possible to generate value-added information utilizing generative AI.

[0292] The "data acquisition unit" is a component for receiving document information including identification information.

[0293] "Identification information" is information that can identify an individual and refers to data such as an individual's name and address.

[0294] "Document information" refers to all text data expressed in an electronically processable format.

[0295] "Analysis device" refers to a device or system for detecting and analyzing identification information contained in received document information.

[0296] "Detection" refers to the process of identifying or finding specific information or data.

[0297] "Anonymization" refers to the process of transforming data so that individuals cannot be identified from their identifying information.

[0298] "Data processing means" refers to methods or devices for removing or anonymizing identifying information and transforming data.

[0299] A "generative model" refers to a machine learning model used to perform language processing tasks.

[0300] "User terminal" refers to a device that a user uses to receive and display information.

[0301] A system for carrying out this invention consists of a data acquisition unit, an analysis device, a data processing means, and a generation model.

[0302] First, the user inputs document information containing identification information using a device such as a smartphone or personal computer. The user's device then transmits this document information to the server via a data acquisition unit. After receiving this information, the server uses an analysis device to detect the identification information contained in the document. Natural language processing libraries such as SpaCy are used for the analysis.

[0303] Next, the server removes or anonymizes the detected identification information and creates the converted data. A program using Python is utilized for this data processing. The data after anonymization is input into the generation model. For the generation model, a natural language processing AI such as GPT-4 is used, and specific information processing tasks, such as sentiment analysis and summary generation, are performed.

[0304] The result generated by the generation model is transmitted from the server to the user's terminal and displayed to the user. Here, the user can view value-added information while their privacy is protected.

[0305] As a specific example, when the user posts a review saying "This is my impression of trying out a new product. I really like it. The delivery was fast," this system detects and removes personal information such as name and address. Then, the generative AI can summarize this review and display "Satisfied with the product and the delivery was also prompt."

[0306] An example of the prompt text input to the generative AI model is as follows. "Remove personal information from the following review and summarize: Review content."

[0307] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0308] Step 1:

[0309] The user inputs document information including identification information using the terminal. The text data input by the user is transmitted from the terminal to the server through the data acquisition unit. At this time, the input data is processed in text format.

[0310] Step 2:

[0311] The server passes the document information received from the data acquisition unit to the analysis unit. The analysis unit uses a natural language processing library such as SpaCy to detect identifying information within the document. The input to this step is the received document information, and the output is a list of detected identifying information. Specifically, the process involves tokenizing the document and identifying patterns that are recognized as identifying information.

[0312] Step 3:

[0313] The server removes or anonymizes the detected identification information using data processing means. In this step, a list of identification information and the original document information are used as input, and anonymized document information is obtained as output. The data transformation involves replacing the identification information with a common noun such as "User X" based on specified rules.

[0314] Step 4:

[0315] The server inputs the converted document information into a generative model. The generative model used here is a natural language processing AI such as GPT-4, which performs information processing tasks according to the specified prompt sentences. The input in this case is anonymized document information, and the output is value-added information such as generated summaries and sentiment ratings. Specifically, it generates text according to the prompt sentences and returns the results to the server.

[0316] Step 5:

[0317] The server sends the output from the generative model to the user's terminal. The user's terminal displays this output, and the user views the information. Here, the input is value-added information obtained from the generative model, and the output is information converted into a format that the user can view on their terminal. Specifically, the function displays the received data on the user's screen.

[0318] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0319] This invention is implemented as a system for securely processing text data containing personal information, and further recognizing user emotions and generating corresponding responses. This system removes personal information from user input data, analyzes emotions, and then provides responses generated by AI.

[0320] A user enters text data using a data entry device. This data typically includes personal information and content that reflects emotions. The device then sends the entered data to a server. This data is transmitted to the server as quickly as possible for immediate processing.

[0321] The server hands over the received data to the information processing unit. The information processing unit first uses a machine learning model to automatically detect and remove or anonymize personal information. This process removes personally identifiable information, thereby enhancing data security.

[0322] Next, the emotion engine analyzes the user's emotions in the input data. The emotion engine has the ability to identify various emotional states—for example, anger, joy, and sadness—and based on this, it can determine the user's current emotion. This emotional information is then reflected in the subsequent response generation process.

[0323] Anonymized data and sentiment information are passed to a generative model. The generative model uses natural language processing techniques to automatically create responses that match the user's emotions. For example, if the user expresses dissatisfaction, it generates an empathetic response; if the user is happy, it generates a positive response.

[0324] The generated response is sent to the terminal via the server. The user can then view the response on their terminal. This mechanism allows users to receive more personalized responses while protecting their personal information and ensuring security.

[0325] A concrete example is a customer support scenario. When a customer sends a complaint about a product to a data entry device, the information processing device uses an emotion engine to identify the customer's anger while removing personal information. Based on this information, the generative model constructs a response that includes an apology and a solution, enabling a quick and appropriate response to the customer. This step contributes to the appropriate use of generative AI that addresses emotions while protecting personal information.

[0326] The following describes the processing flow.

[0327] Step 1:

[0328] Users use data entry devices to input text data as inquiries or feedback. This data may reflect emotions and potentially contain personal information.

[0329] Step 2:

[0330] The terminal receives input data and sends it to the server via a secure protocol. The data is transferred to the server in its original format.

[0331] Step 3:

[0332] The server provides the received data to the information processing device. The information processing device then uses a machine learning model to automatically detect personal information from the text and processes it by removing or anonymizing specific parts.

[0333] Step 4:

[0334] The server passes the data, with personal information removed, to the emotion engine. The emotion engine analyzes the input text to identify the user's emotional state. This includes emotion labels such as anger, joy, and sadness.

[0335] Step 5:

[0336] The results of the sentiment analysis and anonymized data are fed into the generative model. The generative model then enters the process of creating a response adapted to the emotion using natural language processing techniques.

[0337] Step 6:

[0338] The response generated by the generative model is sent back to the server. The server receives the response data and prepares for the next action.

[0339] Step 7:

[0340] The server sends the generated response back to the user's terminal. The terminal receives it and displays it to the user.

[0341] Step 8:

[0342] The user reviews the display on their device and, if necessary, decides on further actions to take regarding the generated response. This process allows the user to receive a personalized response that takes their emotions into consideration.

[0343] (Example 2)

[0344] Next, we will describe Example 2. 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".

[0345] There is a need to securely process text data containing personal information while appropriately recognizing user emotions and efficiently generating responses that correspond to those emotions. However, conventional systems have struggled to balance the protection of personal information with the generation of emotion-based responses. Therefore, there is a need for technology that can perform instant emotion analysis and provide emotion-based responses while ensuring the security of personal information.

[0346] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0347] In this invention, the server includes means for receiving data containing personal information and emotional information via a data input device, means for transferring the data to an information processing device using the server, and means for inputting the data into a generation algorithm to generate an appropriate response. This makes it possible to protect personal information and provide a humane response that responds to emotions.

[0348] A "data input device" is a device used by users to input information, and typically includes a keyboard or touch panel.

[0349] A "server" is a computer device that aggregates data on a network and transfers it to a processing unit.

[0350] An "information processing device" is a device that has the function of analyzing and transforming information within data, and performs sentiment analysis in addition to identifying and removing personal information.

[0351] A "machine learning algorithm" is a mathematical method that allows computers to learn specific patterns from data and perform predictions and classifications.

[0352] "Personal information" refers to information that can identify a specific individual, including information such as name and address.

[0353] An "emotion analysis engine" is a program or device that has the function of identifying human emotions from input data and determining their emotional state.

[0354] A "generative algorithm" is a means of automatically creating an appropriate response based on input data, and primarily utilizes natural language processing.

[0355] "Anonymization" is the process of transforming personal information so that it cannot be identified, and it is a measure taken to protect data privacy.

[0356] This invention begins with a user inputting text data using a data input device. The data input device typically uses a common input interface such as a keyboard or touch panel. The data entered by the user may include personal information or emotional data. The terminal transmits this input data to a server using a communication method such as the internet.

[0357] The server accurately transfers the received data to the information processing unit based on the communication protocol. The information processing unit uses programming languages ​​such as Python and machine learning algorithms, specifically models like TensorFlow and Hugging Face, to detect personal information. This information processing unit protects privacy by anonymizing identifiable information such as names and addresses in the data.

[0358] Next, the information processing device analyzes emotional information in the data using an emotion analysis engine. Here, natural language processing technology is utilized to identify the user's emotional state. The identified emotional information and anonymized data are input into a generation algorithm. This algorithm uses a generative AI model such as OpenAI's GPT series to automatically generate an appropriate response that matches the user's emotions.

[0359] The generated response is sent back to the terminal via the server. The terminal displays the generated response to the user, who can then decide on their next action based on it. This entire process makes it possible to quickly provide more humane and empathetic responses while maintaining the protection of personal information.

[0360] A concrete example is a customer support scenario. For instance, if a user enters "I'm very upset because the product didn't arrive on schedule," the server passes this information to the information processing unit. The information processing unit then uses an analysis engine to identify the emotion of "confusion," and based on this, a generative AI model generates a response such as "We apologize, we will address this immediately." This response is sent to the user's terminal via the server and displayed on the screen.

[0361] Examples of prompt statements include the following:

[0362] User comment: "I'm very upset because the product didn't arrive on schedule."

[0363] Sentiment analysis result: "confused"

[0364] Response generation prompt: "Create a response that expresses empathy for the customer's distress, apologizes, and proposes a solution."

[0365] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0366] Step 1:

[0367] The user inputs text data into a data entry device. The user uses a keyboard or touch panel to input opinions, questions, and other information about the product. This process includes input data containing personal information and emotional information. The output is the user's raw text data.

[0368] Step 2:

[0369] The terminal sends user input data to the server. Here, the terminal uses a secure communication protocol to encrypt and securely transfer the data. The input is the raw text data generated in step 1, and the output is the encrypted data received by the server.

[0370] Step 3:

[0371] The server passes the received data to the information processing unit. The server checks the data for integrity, and if there are no problems, it passes the data to the information processing unit for processing. It receives encrypted text data as input and sends the integrity-checked text data as output to the information processing unit.

[0372] Step 4:

[0373] The information processing device detects, removes, or anonymizes personal information within the data using machine learning algorithms. In this step, for example, TensorFlow is used to identify names, addresses, etc., and anonymize them, such as by displaying only the "name." The input is consistent text data, and the output is text data with the personal information anonymized.

[0374] Step 5:

[0375] An information processing device analyzes the emotions of input data using an emotion analysis engine. Natural language processing techniques are used to identify emotions such as "joy" and "anger" from the input text. The input is anonymized text data, and the output is data with identified emotions.

[0376] Step 6:

[0377] An information processing device generates an emotion-appropriate response using a generation algorithm. The generation AI model constructs prompt sentences that create an appropriate response based on identified emotion information. The input consists of emotion-identified data and prompt sentences, and the output is the generated response.

[0378] Step 7:

[0379] The server sends the generated response to the terminal. The server encrypts the output text and sends it securely to the terminal. The input is the generated response, and the output is the response data received by the terminal.

[0380] Step 8:

[0381] The user confirms the response on their device. The device displays the received response on the screen in a format that is easy for the user to understand. The input is the response sent from the server, and the output is the response that the user visually confirms.

[0382] (Application Example 2)

[0383] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0384] In recent years, services using text data have required the protection of personal information and the provision of appropriate responses that respond to user emotions. To prevent the leakage of personal information and improve the user experience, it is necessary to integrate technologies that address each of these aspects, but such technologies are still under development. This invention aims to address this challenge by providing a system that handles personal information securely while providing prompt responses that are appropriate to the user's emotions.

[0385] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0386] In this invention, the server includes means for receiving text data containing personal information from a data input device, means for using information processing means to detect personal information contained in the text data, data processing means for removing or anonymizing and transforming the detected personal information, and means for inputting the transformed data and sentiment information into a generation model and performing a predetermined task. This makes it possible to automatically generate appropriate responses according to the user's sentiment while securely protecting personal information.

[0387] A "data input device" is a hardware or software means for a user to input text data, and includes smartphones, computers, and other similar devices.

[0388] "Personal information" refers to information used to identify a specific individual, and includes names, addresses, contact information, etc.

[0389] "Information processing means" refers to a device or program for identifying and processing personal information from received data, and may utilize machine learning algorithms.

[0390] "Data processing means" refers to a device or program for removing or anonymizing detected personal information and, if necessary, transforming the data.

[0391] "Converted data" refers to text data from which personal information has been removed or anonymized and converted into a secure format.

[0392] A "generative model" is a machine learning model that uses natural language processing techniques to generate output based on input data.

[0393] An "emotion engine" is software or an algorithm used to analyze and identify a user's emotions from text data.

[0394] A "response generation means" is a device or program that automatically creates an appropriate response based on the user's emotions using a generation model.

[0395] This invention begins with a user inputting text data using a data input device. This input data may contain personal information and the user's emotions. The terminal sends this data to a server for processing.

[0396] The server passes the received data to an information processing system. This information processing system uses machine learning models to detect and remove or anonymize personal information within the text data. Common data anonymization algorithms and libraries are used as tools. This process removes information that identifies specific individuals.

[0397] Next, the emotion engine analyzes the anonymized data and extracts emotions from the text entered by the user. This emotion engine uses natural language processing techniques to identify emotional states (e.g., joy, anxiety, anger) contained in the text and quantifies or categorizes the emotions.

[0398] Anonymized data and sentiment information are input into a generative model. This generative model automatically generates responses that correspond to the user's emotions. The generated responses take the user's emotions into consideration; for example, for data indicating anxiety, it selects words that provide reassurance to construct a response.

[0399] Finally, the generated response is sent to the terminal via the server. The user can then review the response on the terminal and obtain problem resolution or an answer to their inquiry. This process ensures both the protection of personal information and the provision of emotionally responsive responses.

[0400] For example, if a user enters the text, "I've encountered a payment error. I'm worried because I don't know how to fix it," the system will identify the user's anxiety and respond with something like, "We apologize for the inconvenience. We will guide you through the steps to resolve the issue."

[0401] An example of a prompt message is: "User input: A payment error occurred. I am worried because I don't know how to resolve it. Please provide sentiment analysis and an appropriate response."

[0402] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0403] Step 1:

[0404] Users input text data using a data entry device. This text data may include personal information and emotions. The data entered by the user is received by the terminal and immediately sent to the server.

[0405] Step 2:

[0406] The server passes the received text data to an information processing system. The information processing system uses a machine learning model to identify personal information within the text and removes or anonymizes it. Specifically, it extracts identifying information such as names and addresses and deletes or replaces it from the data to generate anonymized data. Through this step, the input becomes anonymized data.

[0407] Step 3:

[0408] The server passes anonymized data to the emotion engine. The emotion engine analyzes the sentiment of the text using natural language processing techniques. In this step, it identifies emotional states (e.g., anxiety, joy) from the text and outputs the sentiment information as a number or category. The sentiment information obtained here is used in the next step.

[0409] Step 4:

[0410] The server inputs anonymized data and sentiment information into a generative model. The generative model generates an appropriate response, taking the sentiment information into account. Specifically, it selects reassuring language structures to correspond to the user's emotions (e.g., anxiety), constructs a response in natural language, and outputs it. This response is tailored to the user's emotions.

[0411] Step 5:

[0412] The server sends the generated response to the terminal. The terminal presents the received response to the user. The user can then review the response and take appropriate action based on it.

[0413] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0414] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0415] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0416] [Third Embodiment]

[0417] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0418] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0419] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0420] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0421] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0422] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0423] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0424] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0425] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0426] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0427] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0428] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0429] This invention is implemented as a system for safely and efficiently processing data containing personal information. The system aims to automatically remove personal information from text data entered by the user and process the converted data using a generating AI.

[0430] The user first inputs text data using a data entry device. This data may include personal information such as customer information and inquiry details. The terminal receives the data entered by the user and transmits it to the server via the network.

[0431] The server passes the received data to an information processing device. This device has the capability to automatically detect personal information contained in the data using a machine learning model. The identified personal information is removed or anonymized so that the data cannot identify individuals. This process ensures the protection of data privacy.

[0432] The anonymized data is then fed into a generative model. This generative model is an AI model designed to perform natural language processing tasks, such as generating responses to user inquiries or creating summaries of long texts. The results from the generative model are then aggregated back onto the server.

[0433] The server sends the generated output results back to the terminal. By checking this output on their own terminal, users can obtain results that utilize the generated AI while protecting their personal information.

[0434] As a concrete example, consider a system for automatically processing customer inquiries in customer support. When a customer submits an inquiry containing personal information as feedback, this system first removes the personal information and then generates an automated response. This series of processes enables safe and rapid customer support.

[0435] Thus, the system based on the present invention makes it possible to maximize the convenience of generated AI while ensuring user privacy.

[0436] The following describes the processing flow.

[0437] Step 1:

[0438] Users enter text data that may contain personal information into a data entry device. The entered data is accepted as inquiries or customer feedback.

[0439] Step 2:

[0440] The terminal sends the user's input data to the server in its original form. During this process, the data is transferred via a secure protocol over the network.

[0441] Step 3:

[0442] The server passes the received data to the information processing unit. Here, the information processing unit uses a machine learning model to automatically identify personal information within the data. It detects identifiable information such as names, addresses, and phone numbers.

[0443] Step 4:

[0444] The data processing device removes or converts detected personal information into specific labels. This anonymizes the data, transforming it into a form that does not identify individuals.

[0445] Step 5:

[0446] The server sends anonymized data to a generative AI model. The generative AI model then uses the received data to perform predetermined tasks, including generating replies and creating summaries.

[0447] Step 6:

[0448] Once the generating AI model completes its processing, it returns the results to the server. The generated results do not contain any personal information.

[0449] Step 7:

[0450] The server sends the results obtained from the generated AI model to the terminal. The terminal presents the results to the user, who can then review the information.

[0451] Step 8:

[0452] Based on the output received by the user, additional modifications are made as needed, or actions are taken using the results. This process allows users to enjoy the convenience of generative AI while protecting their personal information.

[0453] (Example 1)

[0454] Next, we will describe Example 1. 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."

[0455] In modern society, the handling of data containing personal information is increasing, making privacy protection a critical issue. Traditional methods involve cumbersome manual anonymization and management of personal information, creating a need for technologies that process personal information efficiently and securely. Furthermore, in data processing using generative AI models, there is a need for technologies that generate automated responses and summaries while ensuring the protection of personal information.

[0456] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0457] In this invention, the server includes means for receiving document information containing personal identification information at a data input mechanism; means for using an information processing device to detect the personal identification information contained in the document information; data processing means for removing or anonymizing and transforming the detected personal identification information; means for inputting the transformed information into a generation algorithm and executing predetermined processing; means for transmitting the output of the generation algorithm to the data input mechanism; and means for using a secure communication protocol when the data input mechanism sends and receives data. This enables the secure and efficient processing of personal identification information and the generation of automated responses and summaries that take advantage of the convenience of the generation AI model.

[0458] A "data input mechanism" is a device or system for users to input text data and a means of receiving personally identifiable information and other information.

[0459] "Personally identifiable information" refers to information used to identify a specific individual, and includes names, addresses, telephone numbers, etc.

[0460] "Document information" refers to all information expressed in text format, and may include data containing personally identifiable information.

[0461] An "information processing device" is a computer system that detects personally identifiable information from received document information and processes that information.

[0462] A "machine learning algorithm" is a method for analyzing large amounts of data, learning patterns from it, and performing specific tasks.

[0463] "Anonymization" is the process of transforming personally identifiable information in data into a state where it cannot be identified, thereby protecting individual privacy.

[0464] A "generative algorithm" is a process or method for generating new information based on input data, and is particularly used to perform natural language processing tasks.

[0465] A "secure communication protocol" refers to technical specifications and procedures designed to protect sensitive information from being leaked during data transmission.

[0466] This invention is implemented as a system for securely and efficiently processing document information containing personally identifiable information. The user first uses a data input mechanism to input text data into the system. The data input mechanism may be a keyboard or a voice input device. This input text data is transmitted from the terminal to the server via a network.

[0467] The server uses an information processing device to detect personally identifiable information within the received document information. This process utilizes machine learning algorithms. Specifically, a model employing natural language processing technology automatically identifies personally identifiable information such as names, addresses, and phone numbers.

[0468] Any personally identifiable information detected is removed or anonymized by the server. During this process, masking or other methods may be applied to prevent the information from being linked to a specific individual. This process protects individual privacy.

[0469] The anonymized data is then sent from the server to a generation algorithm. This generation algorithm is designed to perform natural language processing tasks. For example, it can generate responses to user inquiries or create summaries of long texts.

[0470] The generated responses and summaries are sent back by the server to a terminal equipped with a data input mechanism. Users can then review and utilize these responses on the terminal. Data security and processing efficiency are ensured throughout this entire process.

[0471] As a concrete example, consider a system in customer support services that automatically processes customer feedback. If a customer submits an inquiry that includes personal information, this system will first remove the personal information and then generate an appropriate automated response to the customer's question. This process ensures safe and prompt customer support.

[0472] As an example of a prompt for a generative AI model, you can use something like, "Based on the inquiry 'Please tell me how to use product X,' please create a general guide on how to use the product."

[0473] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0474] Step 1:

[0475] Users enter text data containing personally identifiable information using a data entry mechanism. This data may include names, addresses, and inquiry details. This data is captured on the device and securely transmitted to a server via the network.

[0476] Step 2:

[0477] The server receives text data sent from the terminal. Based on the received data, the information processing device operates and uses a machine learning algorithm to detect personally identifiable information within the data. The input is raw text data, and the output is data with personally identifiable information identified. At this stage, the detected personal information is verified.

[0478] Step 3:

[0479] The server processes the data to remove or anonymize personally identifiable information. Personal information is securely processed through masking or deletion, resulting in anonymized data. The input is data with identified personal information, and the output is anonymized data. This process ensures that the data is in a privacy-conscious state.

[0480] Step 4:

[0481] The server inputs anonymized data into a generative AI model. The generative model performs natural language processing based on the prompt text and generates the necessary response text, summary, etc. The input is anonymized data, and the output is the generated response text and summary text. In this step, the AI ​​model constructs information according to the user's purpose.

[0482] Step 5:

[0483] The server aggregates the output results obtained from the generative model and sends them to the terminal. The output is in a text format that the user can read, and the terminal receives this information. The user can check the generated response on the terminal screen and take the corresponding action. In this step, the entire system's feedback loop is completed and information is provided to the user.

[0484] (Application Example 1)

[0485] Next, we will explain Application Example 1. In the following explanation, 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."

[0486] In online services, users want to share information with peace of mind, but concerns about privacy violations remain. Furthermore, there is a need to process data containing personal information securely and effectively, while using generative AI to improve the user experience. In particular, there is a demand to automatically remove personal information during review submission on e-commerce sites, thereby protecting user privacy while generating value-added information.

[0487] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0488] In this invention, the server includes means for receiving document information containing identification information in a data acquisition unit, means for detecting the identification information contained in the document information using an analysis device, and data processing means for removing or anonymizing and transforming the detected identification information. This makes it possible to generate value-added information using generation AI while protecting user privacy.

[0489] The "data acquisition unit" is a component for receiving document information that includes identification information.

[0490] "Identifying information" refers to information that can identify an individual, such as an individual's name and address.

[0491] "Document information" refers to all text data expressed in an electronically processable format.

[0492] "Analysis device" refers to a device or system for detecting and analyzing identification information contained in received document information.

[0493] "Detection" refers to the process of identifying or finding specific information or data.

[0494] "Anonymization" refers to the process of transforming data so that individuals cannot be identified from their identifying information.

[0495] "Data processing means" refers to methods or devices for removing or anonymizing identifying information and transforming data.

[0496] A "generative model" refers to a machine learning model used to perform language processing tasks.

[0497] "User terminal" refers to a device that a user uses to receive and display information.

[0498] A system for carrying out this invention consists of a data acquisition unit, an analysis device, a data processing means, and a generation model.

[0499] First, the user inputs document information containing identification information using a device such as a smartphone or personal computer. The user's device then transmits this document information to the server via a data acquisition unit. After receiving this information, the server uses an analysis device to detect the identification information contained in the document. Natural language processing libraries such as SpaCy are used for the analysis.

[0500] Next, the server removes or anonymizes the detected identifiers to create transformed data. A Python program is used for this data processing. The anonymized data is then input into a generative model. The generative model uses a natural language processing AI such as GPT-4 to perform specific information processing tasks, such as sentiment analysis and summarization.

[0501] The results generated by the generative model are sent from the server to the user's device and displayed to the user. Here, the user can view value-added information while their privacy is protected.

[0502] For example, if a user posts a review saying, "I tried out the new product and I really liked it. The delivery was fast," this system will detect and remove personal information such as the user's name and address. The generating AI can then summarize the review and display it as, "I'm satisfied with the product and the delivery was fast."

[0503] An example of a prompt to input into the generating AI model is as follows: "Remove personal information from the following review and summarize it: Review content."

[0504] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0505] Step 1:

[0506] The user inputs document information, including identification information, using a terminal. The text data entered by the user is sent from the terminal to the server via a data acquisition unit. At this time, the input data is processed in text format.

[0507] Step 2:

[0508] The server passes the document information received from the data acquisition unit to the analysis unit. The analysis unit uses a natural language processing library such as SpaCy to detect identifying information within the document. The input to this step is the received document information, and the output is a list of detected identifying information. Specifically, the process involves tokenizing the document and identifying patterns that are recognized as identifying information.

[0509] Step 3:

[0510] The server removes or anonymizes the detected identification information using data processing means. In this step, a list of identification information and the original document information are used as input, and anonymized document information is obtained as output. The data transformation involves replacing the identification information with a common noun such as "User X" based on specified rules.

[0511] Step 4:

[0512] The server inputs the converted document information into a generative model. The generative model used here is a natural language processing AI such as GPT-4, which performs information processing tasks according to the specified prompt sentences. The input in this case is anonymized document information, and the output is value-added information such as generated summaries and sentiment ratings. Specifically, it generates text according to the prompt sentences and returns the results to the server.

[0513] Step 5:

[0514] The server sends the output from the generative model to the user's terminal. The user's terminal displays this output, and the user views the information. Here, the input is value-added information obtained from the generative model, and the output is information converted into a format that the user can view on their terminal. Specifically, the function displays the received data on the user's screen.

[0515] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0516] This invention is implemented as a system for securely processing text data containing personal information, and further recognizing user emotions and generating corresponding responses. This system removes personal information from user input data, analyzes emotions, and then provides responses generated by AI.

[0517] A user enters text data using a data entry device. This data typically includes personal information and content that reflects emotions. The device then sends the entered data to a server. This data is transmitted to the server as quickly as possible for immediate processing.

[0518] The server hands over the received data to the information processing unit. The information processing unit first uses a machine learning model to automatically detect and remove or anonymize personal information. This process removes personally identifiable information, thereby enhancing data security.

[0519] Next, the emotion engine analyzes the user's emotions in the input data. The emotion engine has the ability to identify various emotional states—for example, anger, joy, and sadness—and based on this, it can determine the user's current emotion. This emotional information is then reflected in the subsequent response generation process.

[0520] Anonymized data and sentiment information are passed to a generative model. The generative model uses natural language processing techniques to automatically create responses that match the user's emotions. For example, if the user expresses dissatisfaction, it generates an empathetic response; if the user is happy, it generates a positive response.

[0521] The generated response is sent to the terminal via the server. The user can then view the response on their terminal. This mechanism allows users to receive more personalized responses while protecting their personal information and ensuring security.

[0522] A concrete example is a customer support scenario. When a customer sends a complaint about a product to a data entry device, the information processing device uses an emotion engine to identify the customer's anger while removing personal information. Based on this information, the generative model constructs a response that includes an apology and a solution, enabling a quick and appropriate response to the customer. This step contributes to the appropriate use of generative AI that addresses emotions while protecting personal information.

[0523] The following describes the processing flow.

[0524] Step 1:

[0525] Users use data entry devices to input text data as inquiries or feedback. This data may reflect emotions and potentially contain personal information.

[0526] Step 2:

[0527] The terminal receives input data and sends it to the server via a secure protocol. The data is transferred to the server in its original format.

[0528] Step 3:

[0529] The server provides the received data to the information processing device. The information processing device then uses a machine learning model to automatically detect personal information from the text and processes it by removing or anonymizing specific parts.

[0530] Step 4:

[0531] The server passes the data, with personal information removed, to the emotion engine. The emotion engine analyzes the input text to identify the user's emotional state. This includes emotion labels such as anger, joy, and sadness.

[0532] Step 5:

[0533] The results of the sentiment analysis and anonymized data are fed into the generative model. The generative model then enters the process of creating a response adapted to the emotion using natural language processing techniques.

[0534] Step 6:

[0535] The response generated by the generative model is sent back to the server. The server receives the response data and prepares for the next action.

[0536] Step 7:

[0537] The server sends the generated response back to the user's terminal. The terminal receives it and displays it to the user.

[0538] Step 8:

[0539] The user reviews the display on their device and, if necessary, decides on further actions to take regarding the generated response. This process allows the user to receive a personalized response that takes their emotions into consideration.

[0540] (Example 2)

[0541] Next, we will describe Example 2. 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."

[0542] There is a need to securely process text data containing personal information while appropriately recognizing user emotions and efficiently generating responses that correspond to those emotions. However, conventional systems have struggled to balance the protection of personal information with the generation of emotion-based responses. Therefore, there is a need for technology that can perform instant emotion analysis and provide emotion-based responses while ensuring the security of personal information.

[0543] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0544] In this invention, the server includes means for receiving data containing personal information and emotional information via a data input device, means for transferring the data to an information processing device using the server, and means for inputting the data into a generation algorithm to generate an appropriate response. This makes it possible to protect personal information and provide a humane response that responds to emotions.

[0545] A "data input device" is a device used by users to input information, and typically includes a keyboard or touch panel.

[0546] A "server" is a computer device that aggregates data on a network and transfers it to a processing unit.

[0547] An "information processing device" is a device that has the function of analyzing and transforming information within data, and performs sentiment analysis in addition to identifying and removing personal information.

[0548] A "machine learning algorithm" is a mathematical method that allows computers to learn specific patterns from data and perform predictions and classifications.

[0549] "Personal information" refers to information that can identify a specific individual, including information such as name and address.

[0550] An "emotion analysis engine" is a program or device that has the function of identifying human emotions from input data and determining their emotional state.

[0551] A "generative algorithm" is a means of automatically creating an appropriate response based on input data, and primarily utilizes natural language processing.

[0552] "Anonymization" is the process of transforming personal information so that it cannot be identified, and it is a measure taken to protect data privacy.

[0553] This invention begins with a user inputting text data using a data input device. The data input device typically uses a common input interface such as a keyboard or touch panel. The data entered by the user may include personal information or emotional data. The terminal transmits this input data to a server using a communication method such as the internet.

[0554] The server accurately transfers the received data to the information processing unit based on the communication protocol. The information processing unit uses programming languages ​​such as Python and machine learning algorithms, specifically models like TensorFlow and Hugging Face, to detect personal information. This information processing unit protects privacy by anonymizing identifiable information such as names and addresses in the data.

[0555] Next, the information processing device analyzes emotional information in the data using an emotion analysis engine. Here, natural language processing technology is utilized to identify the user's emotional state. The identified emotional information and anonymized data are input into a generation algorithm. This algorithm uses a generative AI model such as OpenAI's GPT series to automatically generate an appropriate response that matches the user's emotions.

[0556] The generated response is sent back to the terminal via the server. The terminal displays the generated response to the user, who can then decide on their next action based on it. This entire process makes it possible to quickly provide more humane and empathetic responses while maintaining the protection of personal information.

[0557] A concrete example is a customer support scenario. For instance, if a user enters "I'm very upset because the product didn't arrive on schedule," the server passes this information to the information processing unit. The information processing unit then uses an analysis engine to identify the emotion of "confusion," and based on this, a generative AI model generates a response such as "We apologize, we will address this immediately." This response is sent to the user's terminal via the server and displayed on the screen.

[0558] Examples of prompt statements include the following:

[0559] User comment: "I'm very upset because the product didn't arrive on schedule."

[0560] Sentiment analysis result: "confused"

[0561] Response generation prompt: "Create a response that expresses empathy for the customer's distress, apologizes, and proposes a solution."

[0562] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0563] Step 1:

[0564] The user inputs text data into a data entry device. The user uses a keyboard or touch panel to input opinions, questions, and other information about the product. This process includes input data containing personal information and emotional information. The output is the user's raw text data.

[0565] Step 2:

[0566] The terminal sends user input data to the server. Here, the terminal uses a secure communication protocol to encrypt and securely transfer the data. The input is the raw text data generated in step 1, and the output is the encrypted data received by the server.

[0567] Step 3:

[0568] The server passes the received data to the information processing unit. The server checks the data for integrity, and if there are no problems, it passes the data to the information processing unit for processing. It receives encrypted text data as input and sends the integrity-checked text data as output to the information processing unit.

[0569] Step 4:

[0570] The information processing device detects, removes, or anonymizes personal information within the data using machine learning algorithms. In this step, for example, TensorFlow is used to identify names, addresses, etc., and anonymize them, such as by displaying only the "name." The input is consistent text data, and the output is text data with the personal information anonymized.

[0571] Step 5:

[0572] An information processing device analyzes the emotions of input data using an emotion analysis engine. Natural language processing techniques are used to identify emotions such as "joy" and "anger" from the input text. The input is anonymized text data, and the output is data with identified emotions.

[0573] Step 6:

[0574] An information processing device generates an emotion-appropriate response using a generation algorithm. The generation AI model constructs prompt sentences that create an appropriate response based on identified emotion information. The input consists of emotion-identified data and prompt sentences, and the output is the generated response.

[0575] Step 7:

[0576] The server sends the generated response to the terminal. The server encrypts the output text and sends it securely to the terminal. The input is the generated response, and the output is the response data received by the terminal.

[0577] Step 8:

[0578] The user confirms the response on their device. The device displays the received response on the screen in a format that is easy for the user to understand. The input is the response sent from the server, and the output is the response that the user visually confirms.

[0579] (Application Example 2)

[0580] Next, we will explain application example 2. In the following explanation, 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."

[0581] In recent years, services using text data have required the protection of personal information and the provision of appropriate responses that respond to user emotions. To prevent the leakage of personal information and improve the user experience, it is necessary to integrate technologies that address each of these aspects, but such technologies are still under development. This invention aims to address this challenge by providing a system that handles personal information securely while providing prompt responses that are appropriate to the user's emotions.

[0582] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0583] In this invention, the server includes means for receiving text data containing personal information from a data input device, means for using information processing means to detect personal information contained in the text data, data processing means for removing or anonymizing and transforming the detected personal information, and means for inputting the transformed data and sentiment information into a generation model and performing a predetermined task. This makes it possible to automatically generate appropriate responses according to the user's sentiment while securely protecting personal information.

[0584] A "data input device" is a hardware or software means for a user to input text data, and includes smartphones, computers, and other similar devices.

[0585] "Personal information" refers to information used to identify a specific individual, and includes names, addresses, contact information, etc.

[0586] "Information processing means" refers to a device or program for identifying and processing personal information from received data, and may utilize machine learning algorithms.

[0587] "Data processing means" refers to a device or program for removing or anonymizing detected personal information and, if necessary, transforming the data.

[0588] "Converted data" refers to text data from which personal information has been removed or anonymized and converted into a secure format.

[0589] A "generative model" is a machine learning model that uses natural language processing techniques to generate output based on input data.

[0590] An "emotion engine" is software or an algorithm used to analyze and identify a user's emotions from text data.

[0591] A "response generation means" is a device or program that automatically creates an appropriate response based on the user's emotions using a generation model.

[0592] This invention begins with a user inputting text data using a data input device. This input data may contain personal information and the user's emotions. The terminal sends this data to a server for processing.

[0593] The server passes the received data to an information processing system. This information processing system uses machine learning models to detect and remove or anonymize personal information within the text data. Common data anonymization algorithms and libraries are used as tools. This process removes information that identifies specific individuals.

[0594] Next, the emotion engine analyzes the anonymized data and extracts emotions from the text entered by the user. This emotion engine uses natural language processing techniques to identify emotional states (e.g., joy, anxiety, anger) contained in the text and quantifies or categorizes the emotions.

[0595] Anonymized data and sentiment information are input into a generative model. This generative model automatically generates responses that correspond to the user's emotions. The generated responses take the user's emotions into consideration; for example, for data indicating anxiety, it selects words that provide reassurance to construct a response.

[0596] Finally, the generated response is sent to the terminal via the server. The user can then review the response on the terminal and obtain problem resolution or an answer to their inquiry. This process ensures both the protection of personal information and the provision of emotionally responsive responses.

[0597] For example, if a user enters the text, "I've encountered a payment error. I'm worried because I don't know how to fix it," the system will identify the user's anxiety and respond with something like, "We apologize for the inconvenience. We will guide you through the steps to resolve the issue."

[0598] An example of a prompt message is: "User input: A payment error occurred. I am worried because I don't know how to resolve it. Please provide sentiment analysis and an appropriate response."

[0599] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0600] Step 1:

[0601] Users input text data using a data entry device. This text data may include personal information and emotions. The data entered by the user is received by the terminal and immediately sent to the server.

[0602] Step 2:

[0603] The server passes the received text data to an information processing system. The information processing system uses a machine learning model to identify personal information within the text and removes or anonymizes it. Specifically, it extracts identifying information such as names and addresses and deletes or replaces it from the data to generate anonymized data. Through this step, the input becomes anonymized data.

[0604] Step 3:

[0605] The server passes anonymized data to the emotion engine. The emotion engine analyzes the sentiment of the text using natural language processing techniques. In this step, it identifies emotional states (e.g., anxiety, joy) from the text and outputs the sentiment information as a number or category. The sentiment information obtained here is used in the next step.

[0606] Step 4:

[0607] The server inputs anonymized data and sentiment information into a generative model. The generative model generates an appropriate response, taking the sentiment information into account. Specifically, it selects reassuring language structures to correspond to the user's emotions (e.g., anxiety), constructs a response in natural language, and outputs it. This response is tailored to the user's emotions.

[0608] Step 5:

[0609] The server sends the generated response to the terminal. The terminal presents the received response to the user. The user can then review the response and take appropriate action based on it.

[0610] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0611] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0612] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0613] [Fourth Embodiment]

[0614] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0615] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0616] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0617] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0618] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0619] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0620] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0621] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0622] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0623] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0624] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0625] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0626] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0627] This invention is implemented as a system for safely and efficiently processing data containing personal information. The system aims to automatically remove personal information from text data entered by the user and process the converted data using a generating AI.

[0628] The user first inputs text data using a data entry device. This data may include personal information such as customer information and inquiry details. The terminal receives the data entered by the user and transmits it to the server via the network.

[0629] The server passes the received data to an information processing device. This device has the capability to automatically detect personal information contained in the data using a machine learning model. The identified personal information is removed or anonymized so that the data cannot identify individuals. This process ensures the protection of data privacy.

[0630] The anonymized data is then fed into a generative model. This generative model is an AI model designed to perform natural language processing tasks, such as generating responses to user inquiries or creating summaries of long texts. The results from the generative model are then aggregated back onto the server.

[0631] The server sends the generated output results back to the terminal. By checking this output on their own terminal, users can obtain results that utilize the generated AI while protecting their personal information.

[0632] As a concrete example, consider a system for automatically processing customer inquiries in customer support. When a customer submits an inquiry containing personal information as feedback, this system first removes the personal information and then generates an automated response. This series of processes enables safe and rapid customer support.

[0633] Thus, the system based on the present invention makes it possible to maximize the convenience of generated AI while ensuring user privacy.

[0634] The following describes the processing flow.

[0635] Step 1:

[0636] Users enter text data that may contain personal information into a data entry device. The entered data is accepted as inquiries or customer feedback.

[0637] Step 2:

[0638] The terminal sends the user's input data to the server in its original form. During this process, the data is transferred via a secure protocol over the network.

[0639] Step 3:

[0640] The server passes the received data to the information processing unit. Here, the information processing unit uses a machine learning model to automatically identify personal information within the data. It detects identifiable information such as names, addresses, and phone numbers.

[0641] Step 4:

[0642] The data processing device removes or converts detected personal information into specific labels. This anonymizes the data, transforming it into a form that does not identify individuals.

[0643] Step 5:

[0644] The server sends anonymized data to a generative AI model. The generative AI model then uses the received data to perform predetermined tasks, including generating replies and creating summaries.

[0645] Step 6:

[0646] Once the generating AI model completes its processing, it returns the results to the server. The generated results do not contain any personal information.

[0647] Step 7:

[0648] The server sends the results obtained from the generated AI model to the terminal. The terminal presents the results to the user, who can then review the information.

[0649] Step 8:

[0650] Based on the output received by the user, additional modifications are made as needed, or actions are taken using the results. This process allows users to enjoy the convenience of generative AI while protecting their personal information.

[0651] (Example 1)

[0652] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0653] In modern society, the handling of data containing personal information is increasing, making privacy protection a critical issue. Traditional methods involve cumbersome manual anonymization and management of personal information, creating a need for technologies that process personal information efficiently and securely. Furthermore, in data processing using generative AI models, there is a need for technologies that generate automated responses and summaries while ensuring the protection of personal information.

[0654] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0655] In this invention, the server includes means for receiving document information containing personal identification information at a data input mechanism; means for using an information processing device to detect the personal identification information contained in the document information; data processing means for removing or anonymizing and transforming the detected personal identification information; means for inputting the transformed information into a generation algorithm and executing predetermined processing; means for transmitting the output of the generation algorithm to the data input mechanism; and means for using a secure communication protocol when the data input mechanism sends and receives data. This enables the secure and efficient processing of personal identification information and the generation of automated responses and summaries that take advantage of the convenience of the generation AI model.

[0656] A "data input mechanism" is a device or system for users to input text data and a means of receiving personally identifiable information and other information.

[0657] "Personally identifiable information" refers to information used to identify a specific individual, and includes names, addresses, telephone numbers, etc.

[0658] "Document information" refers to all information expressed in text format, and may include data containing personally identifiable information.

[0659] An "information processing device" is a computer system that detects personally identifiable information from received document information and processes that information.

[0660] A "machine learning algorithm" is a method for analyzing large amounts of data, learning patterns from it, and performing specific tasks.

[0661] "Anonymization" is the process of transforming personally identifiable information in data into a state where it cannot be identified, thereby protecting individual privacy.

[0662] A "generative algorithm" is a process or method for generating new information based on input data, and is particularly used to perform natural language processing tasks.

[0663] A "secure communication protocol" refers to technical specifications and procedures designed to protect sensitive information from being leaked during data transmission.

[0664] This invention is implemented as a system for securely and efficiently processing document information containing personally identifiable information. The user first uses a data input mechanism to input text data into the system. The data input mechanism may be a keyboard or a voice input device. This input text data is transmitted from the terminal to the server via a network.

[0665] The server uses an information processing device to detect personally identifiable information within the received document information. This process utilizes machine learning algorithms. Specifically, a model employing natural language processing technology automatically identifies personally identifiable information such as names, addresses, and phone numbers.

[0666] Any personally identifiable information detected is removed or anonymized by the server. During this process, masking or other methods may be applied to prevent the information from being linked to a specific individual. This process protects individual privacy.

[0667] The anonymized data is then sent from the server to a generation algorithm. This generation algorithm is designed to perform natural language processing tasks. For example, it can generate responses to user inquiries or create summaries of long texts.

[0668] The generated responses and summaries are sent back by the server to a terminal equipped with a data input mechanism. Users can then review and utilize these responses on the terminal. Data security and processing efficiency are ensured throughout this entire process.

[0669] As a concrete example, consider a system in customer support services that automatically processes customer feedback. If a customer submits an inquiry that includes personal information, this system will first remove the personal information and then generate an appropriate automated response to the customer's question. This process ensures safe and prompt customer support.

[0670] As an example of a prompt for a generative AI model, you can use something like, "Based on the inquiry 'Please tell me how to use product X,' please create a general guide on how to use the product."

[0671] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0672] Step 1:

[0673] Users enter text data containing personally identifiable information using a data entry mechanism. This data may include names, addresses, and inquiry details. This data is captured on the device and securely transmitted to a server via the network.

[0674] Step 2:

[0675] The server receives text data sent from the terminal. Based on the received data, the information processing device operates and uses a machine learning algorithm to detect personally identifiable information within the data. The input is raw text data, and the output is data with personally identifiable information identified. At this stage, the detected personal information is verified.

[0676] Step 3:

[0677] The server processes the data to remove or anonymize personally identifiable information. Personal information is securely processed through masking or deletion, resulting in anonymized data. The input is data with identified personal information, and the output is anonymized data. This process ensures that the data is in a privacy-conscious state.

[0678] Step 4:

[0679] The server inputs anonymized data into a generative AI model. The generative model performs natural language processing based on the prompt text and generates the necessary response text, summary, etc. The input is anonymized data, and the output is the generated response text and summary text. In this step, the AI ​​model constructs information according to the user's purpose.

[0680] Step 5:

[0681] The server aggregates the output results obtained from the generative model and sends them to the terminal. The output is in a text format that the user can read, and the terminal receives this information. The user can check the generated response on the terminal screen and take the corresponding action. In this step, the entire system's feedback loop is completed and information is provided to the user.

[0682] (Application Example 1)

[0683] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0684] In online services, users want to share information with peace of mind, but concerns about privacy violations remain. Furthermore, there is a need to process data containing personal information securely and effectively, while using generative AI to improve the user experience. In particular, there is a demand to automatically remove personal information during review submission on e-commerce sites, thereby protecting user privacy while generating value-added information.

[0685] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0686] In this invention, the server includes means for receiving document information containing identification information in a data acquisition unit, means for detecting the identification information contained in the document information using an analysis device, and data processing means for removing or anonymizing and transforming the detected identification information. This makes it possible to generate value-added information using generation AI while protecting user privacy.

[0687] The "data acquisition unit" is a component for receiving document information that includes identification information.

[0688] "Identifying information" refers to information that can identify an individual, such as an individual's name and address.

[0689] "Document information" refers to all text data expressed in an electronically processable format.

[0690] "Analysis device" refers to a device or system for detecting and analyzing identification information contained in received document information.

[0691] "Detection" refers to the process of identifying or finding specific information or data.

[0692] "Anonymization" refers to the process of transforming data so that individuals cannot be identified from their identifying information.

[0693] "Data processing means" refers to methods or devices for removing or anonymizing identifying information and transforming data.

[0694] A "generative model" refers to a machine learning model used to perform language processing tasks.

[0695] "User terminal" refers to a device that a user uses to receive and display information.

[0696] A system for carrying out this invention consists of a data acquisition unit, an analysis device, a data processing means, and a generation model.

[0697] First, the user inputs document information containing identification information using a device such as a smartphone or personal computer. The user's device then transmits this document information to the server via a data acquisition unit. After receiving this information, the server uses an analysis device to detect the identification information contained in the document. Natural language processing libraries such as SpaCy are used for the analysis.

[0698] Next, the server removes or anonymizes the detected identifiers to create transformed data. A Python program is used for this data processing. The anonymized data is then input into a generative model. The generative model uses a natural language processing AI such as GPT-4 to perform specific information processing tasks, such as sentiment analysis and summarization.

[0699] The results generated by the generative model are sent from the server to the user's device and displayed to the user. Here, the user can view value-added information while their privacy is protected.

[0700] For example, if a user posts a review saying, "I tried out the new product and I really liked it. The delivery was fast," this system will detect and remove personal information such as the user's name and address. The generating AI can then summarize the review and display it as, "I'm satisfied with the product and the delivery was fast."

[0701] An example of a prompt to input into the generating AI model is as follows: "Remove personal information from the following review and summarize it: Review content."

[0702] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0703] Step 1:

[0704] The user inputs document information, including identification information, using a terminal. The text data entered by the user is sent from the terminal to the server via a data acquisition unit. At this time, the input data is processed in text format.

[0705] Step 2:

[0706] The server passes the document information received from the data acquisition unit to the analysis unit. The analysis unit uses a natural language processing library such as SpaCy to detect identifying information within the document. The input to this step is the received document information, and the output is a list of detected identifying information. Specifically, the process involves tokenizing the document and identifying patterns that are recognized as identifying information.

[0707] Step 3:

[0708] The server removes or anonymizes the detected identification information using data processing means. In this step, a list of identification information and the original document information are used as input, and anonymized document information is obtained as output. The data transformation involves replacing the identification information with a common noun such as "User X" based on specified rules.

[0709] Step 4:

[0710] The server inputs the converted document information into a generative model. The generative model used here is a natural language processing AI such as GPT-4, which performs information processing tasks according to the specified prompt sentences. The input in this case is anonymized document information, and the output is value-added information such as generated summaries and sentiment ratings. Specifically, it generates text according to the prompt sentences and returns the results to the server.

[0711] Step 5:

[0712] The server sends the output from the generative model to the user's terminal. The user's terminal displays this output, and the user views the information. Here, the input is value-added information obtained from the generative model, and the output is information converted into a format that the user can view on their terminal. Specifically, the function displays the received data on the user's screen.

[0713] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0714] This invention is implemented as a system for securely processing text data containing personal information, and further recognizing user emotions and generating corresponding responses. This system removes personal information from user input data, analyzes emotions, and then provides responses generated by AI.

[0715] A user enters text data using a data entry device. This data typically includes personal information and content that reflects emotions. The device then sends the entered data to a server. This data is transmitted to the server as quickly as possible for immediate processing.

[0716] The server hands over the received data to the information processing unit. The information processing unit first uses a machine learning model to automatically detect and remove or anonymize personal information. This process removes personally identifiable information, thereby enhancing data security.

[0717] Next, the emotion engine analyzes the user's emotions in the input data. The emotion engine has the ability to identify various emotional states—for example, anger, joy, and sadness—and based on this, it can determine the user's current emotion. This emotional information is then reflected in the subsequent response generation process.

[0718] Anonymized data and sentiment information are passed to a generative model. The generative model uses natural language processing techniques to automatically create responses that match the user's emotions. For example, if the user expresses dissatisfaction, it generates an empathetic response; if the user is happy, it generates a positive response.

[0719] The generated response is sent to the terminal via the server. The user can then view the response on their terminal. This mechanism allows users to receive more personalized responses while protecting their personal information and ensuring security.

[0720] A concrete example is a customer support scenario. When a customer sends a complaint about a product to a data entry device, the information processing device uses an emotion engine to identify the customer's anger while removing personal information. Based on this information, the generative model constructs a response that includes an apology and a solution, enabling a quick and appropriate response to the customer. This step contributes to the appropriate use of generative AI that addresses emotions while protecting personal information.

[0721] The following describes the processing flow.

[0722] Step 1:

[0723] Users use data entry devices to input text data as inquiries or feedback. This data may reflect emotions and potentially contain personal information.

[0724] Step 2:

[0725] The terminal receives input data and sends it to the server via a secure protocol. The data is transferred to the server in its original format.

[0726] Step 3:

[0727] The server provides the received data to the information processing device. The information processing device then uses a machine learning model to automatically detect personal information from the text and processes it by removing or anonymizing specific parts.

[0728] Step 4:

[0729] The server passes the data, with personal information removed, to the emotion engine. The emotion engine analyzes the input text to identify the user's emotional state. This includes emotion labels such as anger, joy, and sadness.

[0730] Step 5:

[0731] The results of the sentiment analysis and anonymized data are fed into the generative model. The generative model then enters the process of creating a response adapted to the emotion using natural language processing techniques.

[0732] Step 6:

[0733] The response generated by the generative model is sent back to the server. The server receives the response data and prepares for the next action.

[0734] Step 7:

[0735] The server sends the generated response back to the user's terminal. The terminal receives it and displays it to the user.

[0736] Step 8:

[0737] The user reviews the display on their device and, if necessary, decides on further actions to take regarding the generated response. This process allows the user to receive a personalized response that takes their emotions into consideration.

[0738] (Example 2)

[0739] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0740] There is a need to securely process text data containing personal information while appropriately recognizing user emotions and efficiently generating responses that correspond to those emotions. However, conventional systems have struggled to balance the protection of personal information with the generation of emotion-based responses. Therefore, there is a need for technology that can perform instant emotion analysis and provide emotion-based responses while ensuring the security of personal information.

[0741] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0742] In this invention, the server includes means for receiving data containing personal information and emotional information via a data input device, means for transferring the data to an information processing device using the server, and means for inputting the data into a generation algorithm to generate an appropriate response. This makes it possible to protect personal information and provide a humane response that responds to emotions.

[0743] A "data input device" is a device used by users to input information, and typically includes a keyboard or touch panel.

[0744] A "server" is a computer device that aggregates data on a network and transfers it to a processing unit.

[0745] An "information processing device" is a device that has the function of analyzing and transforming information within data, and performs sentiment analysis in addition to identifying and removing personal information.

[0746] A "machine learning algorithm" is a mathematical method that allows computers to learn specific patterns from data and perform predictions and classifications.

[0747] "Personal information" refers to information that can identify a specific individual, including information such as name and address.

[0748] An "emotion analysis engine" is a program or device that has the function of identifying human emotions from input data and determining their emotional state.

[0749] A "generative algorithm" is a means of automatically creating an appropriate response based on input data, and primarily utilizes natural language processing.

[0750] "Anonymization" is the process of transforming personal information so that it cannot be identified, and it is a measure taken to protect data privacy.

[0751] This invention begins with a user inputting text data using a data input device. The data input device typically uses a common input interface such as a keyboard or touch panel. The data entered by the user may include personal information or emotional data. The terminal transmits this input data to a server using a communication method such as the internet.

[0752] The server accurately transfers the received data to the information processing unit based on the communication protocol. The information processing unit uses programming languages ​​such as Python and machine learning algorithms, specifically models like TensorFlow and Hugging Face, to detect personal information. This information processing unit protects privacy by anonymizing identifiable information such as names and addresses in the data.

[0753] Next, the information processing device analyzes emotional information in the data using an emotion analysis engine. Here, natural language processing technology is utilized to identify the user's emotional state. The identified emotional information and anonymized data are input into a generation algorithm. This algorithm uses a generative AI model such as OpenAI's GPT series to automatically generate an appropriate response that matches the user's emotions.

[0754] The generated response is sent back to the terminal via the server. The terminal displays the generated response to the user, who can then decide on their next action based on it. This entire process makes it possible to quickly provide more humane and empathetic responses while maintaining the protection of personal information.

[0755] A concrete example is a customer support scenario. For instance, if a user enters "I'm very upset because the product didn't arrive on schedule," the server passes this information to the information processing unit. The information processing unit then uses an analysis engine to identify the emotion of "confusion," and based on this, a generative AI model generates a response such as "We apologize, we will address this immediately." This response is sent to the user's terminal via the server and displayed on the screen.

[0756] Examples of prompt statements include the following:

[0757] User comment: "I'm very upset because the product didn't arrive on schedule."

[0758] Sentiment analysis result: "confused"

[0759] Response generation prompt: "Create a response that expresses empathy for the customer's distress, apologizes, and proposes a solution."

[0760] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0761] Step 1:

[0762] The user inputs text data into a data entry device. The user uses a keyboard or touch panel to input opinions, questions, and other information about the product. This process includes input data containing personal information and emotional information. The output is the user's raw text data.

[0763] Step 2:

[0764] The terminal sends user input data to the server. Here, the terminal uses a secure communication protocol to encrypt and securely transfer the data. The input is the raw text data generated in step 1, and the output is the encrypted data received by the server.

[0765] Step 3:

[0766] The server passes the received data to the information processing unit. The server checks the data for integrity, and if there are no problems, it passes the data to the information processing unit for processing. It receives encrypted text data as input and sends the integrity-checked text data as output to the information processing unit.

[0767] Step 4:

[0768] The information processing device detects, removes, or anonymizes personal information within the data using machine learning algorithms. In this step, for example, TensorFlow is used to identify names, addresses, etc., and anonymize them, such as by displaying only the "name." The input is consistent text data, and the output is text data with the personal information anonymized.

[0769] Step 5:

[0770] An information processing device analyzes the emotions of input data using an emotion analysis engine. Natural language processing techniques are used to identify emotions such as "joy" and "anger" from the input text. The input is anonymized text data, and the output is data with identified emotions.

[0771] Step 6:

[0772] An information processing device generates an emotion-appropriate response using a generation algorithm. The generation AI model constructs prompt sentences that create an appropriate response based on identified emotion information. The input consists of emotion-identified data and prompt sentences, and the output is the generated response.

[0773] Step 7:

[0774] The server sends the generated response to the terminal. The server encrypts the output text and sends it securely to the terminal. The input is the generated response, and the output is the response data received by the terminal.

[0775] Step 8:

[0776] The user confirms the response on their device. The device displays the received response on the screen in a format that is easy for the user to understand. The input is the response sent from the server, and the output is the response that the user visually confirms.

[0777] (Application Example 2)

[0778] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0779] In recent years, services using text data have required the protection of personal information and the provision of appropriate responses that respond to user emotions. To prevent the leakage of personal information and improve the user experience, it is necessary to integrate technologies that address each of these aspects, but such technologies are still under development. This invention aims to address this challenge by providing a system that handles personal information securely while providing prompt responses that are appropriate to the user's emotions.

[0780] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0781] In this invention, the server includes means for receiving text data containing personal information from a data input device, means for using information processing means to detect personal information contained in the text data, data processing means for removing or anonymizing and transforming the detected personal information, and means for inputting the transformed data and sentiment information into a generation model and performing a predetermined task. This makes it possible to automatically generate appropriate responses according to the user's sentiment while securely protecting personal information.

[0782] A "data input device" is a hardware or software means for a user to input text data, and includes smartphones, computers, and other similar devices.

[0783] "Personal information" refers to information used to identify a specific individual, and includes names, addresses, contact information, etc.

[0784] "Information processing means" refers to a device or program for identifying and processing personal information from received data, and may utilize machine learning algorithms.

[0785] "Data processing means" refers to a device or program for removing or anonymizing detected personal information and, if necessary, transforming the data.

[0786] "Converted data" refers to text data from which personal information has been removed or anonymized and converted into a secure format.

[0787] A "generative model" is a machine learning model that uses natural language processing techniques to generate output based on input data.

[0788] An "emotion engine" is software or an algorithm used to analyze and identify a user's emotions from text data.

[0789] A "response generation means" is a device or program that automatically creates an appropriate response based on the user's emotions using a generation model.

[0790] This invention begins with a user inputting text data using a data input device. This input data may contain personal information and the user's emotions. The terminal sends this data to a server for processing.

[0791] The server passes the received data to an information processing system. This information processing system uses machine learning models to detect and remove or anonymize personal information within the text data. Common data anonymization algorithms and libraries are used as tools. This process removes information that identifies specific individuals.

[0792] Next, the emotion engine analyzes the anonymized data and extracts emotions from the text entered by the user. This emotion engine uses natural language processing techniques to identify emotional states (e.g., joy, anxiety, anger) contained in the text and quantifies or categorizes the emotions.

[0793] Anonymized data and sentiment information are input into a generative model. This generative model automatically generates responses that correspond to the user's emotions. The generated responses take the user's emotions into consideration; for example, for data indicating anxiety, it selects words that provide reassurance to construct a response.

[0794] Finally, the generated response is sent to the terminal via the server. The user can then review the response on the terminal and obtain problem resolution or an answer to their inquiry. This process ensures both the protection of personal information and the provision of emotionally responsive responses.

[0795] For example, if a user enters the text, "I've encountered a payment error. I'm worried because I don't know how to fix it," the system will identify the user's anxiety and respond with something like, "We apologize for the inconvenience. We will guide you through the steps to resolve the issue."

[0796] An example of a prompt message is: "User input: A payment error occurred. I am worried because I don't know how to resolve it. Please provide sentiment analysis and an appropriate response."

[0797] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0798] Step 1:

[0799] Users input text data using a data entry device. This text data may include personal information and emotions. The data entered by the user is received by the terminal and immediately sent to the server.

[0800] Step 2:

[0801] The server passes the received text data to an information processing system. The information processing system uses a machine learning model to identify personal information within the text and removes or anonymizes it. Specifically, it extracts identifying information such as names and addresses and deletes or replaces it from the data to generate anonymized data. Through this step, the input becomes anonymized data.

[0802] Step 3:

[0803] The server passes anonymized data to the emotion engine. The emotion engine analyzes the sentiment of the text using natural language processing techniques. In this step, it identifies emotional states (e.g., anxiety, joy) from the text and outputs the sentiment information as a number or category. The sentiment information obtained here is used in the next step.

[0804] Step 4:

[0805] The server inputs anonymized data and sentiment information into a generative model. The generative model generates an appropriate response, taking the sentiment information into account. Specifically, it selects reassuring language structures to correspond to the user's emotions (e.g., anxiety), constructs a response in natural language, and outputs it. This response is tailored to the user's emotions.

[0806] Step 5:

[0807] The server sends the generated response to the terminal. The terminal presents the received response to the user. The user can then review the response and take appropriate action based on it.

[0808] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0809] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0810] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0811] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0812] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0813] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0814] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0815] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0816] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0817] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0818] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0819] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0820] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0821] 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.

[0822] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0823] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0824] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0825] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0826] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0827] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0828] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0829] The following is further disclosed regarding the embodiments described above.

[0830] (Claim 1)

[0831] A means of receiving text data containing personal information at a data input device,

[0832] A means of using an information processing device that detects personal information contained in the aforementioned text data,

[0833] A data processing means for removing or anonymizing and transforming the detected personal information,

[0834] A means for inputting the converted data into a generation model and executing a predetermined task,

[0835] Means for transmitting the output of the generation model to a data input device,

[0836] A system that includes this.

[0837] (Claim 2)

[0838] The system according to claim 1, wherein the information processing device includes means for identifying personal information using a machine learning model.

[0839] (Claim 3)

[0840] The system according to claim 1, wherein the generative model includes means for performing a natural language processing task.

[0841] "Example 1"

[0842] (Claim 1)

[0843] A means for receiving document information containing personally identifiable information at a data input mechanism,

[0844] Means for using an information processing device to detect personal identification information contained in the document information,

[0845] A data processing means for removing or anonymizing and transforming the detected personal identification information,

[0846] A means for inputting the converted information into a generation algorithm and executing a predetermined process,

[0847] Means for transmitting the output of the generation algorithm to a data input mechanism,

[0848] The aforementioned data input mechanism uses a secure communication protocol when sending and receiving data,

[0849] A system that includes this.

[0850] (Claim 2)

[0851] The system according to claim 1, wherein the information processing device includes means for identifying personal identification information using a machine learning algorithm.

[0852] (Claim 3)

[0853] The system according to claim 1, wherein the generation algorithm includes means for performing processing including natural language processing.

[0854] "Application Example 1"

[0855] (Claim 1)

[0856] The data acquisition unit includes means for receiving document information including identification information,

[0857] Means for detecting identification information contained in the document information using an analysis device,

[0858] A data processing means for removing or anonymizing and transforming the detected identification information,

[0859] A means for inputting the converted data into a generation model and executing an information processing task,

[0860] Means for transmitting the output of the generation model to the data acquisition unit,

[0861] A means for making the aforementioned data viewable on a user terminal,

[0862] A system that includes this.

[0863] (Claim 2)

[0864] The system according to claim 1, wherein the analysis device includes means for identifying identification information using a learning model.

[0865] (Claim 3)

[0866] The system according to claim 1, wherein the generative model includes means for performing a language processing task.

[0867] "Example 2 of combining an emotion engine"

[0868] (Claim 1)

[0869] A means for receiving data including personal information and emotional information at a data input device,

[0870] Means for transmitting the aforementioned data to the server,

[0871] A means of transferring the data to an information processing device using a server,

[0872] The aforementioned information processing device includes means for identifying, removing, or anonymizing personal information using a machine learning algorithm,

[0873] The aforementioned information processing device includes means for analyzing emotional information using an emotion analysis engine,

[0874] A means for inputting the anonymized data and sentiment information into a generation algorithm and generating an appropriate response,

[0875] Means for transmitting the generated response to a data input device via a server,

[0876] A system that includes this.

[0877] (Claim 2)

[0878] The system according to claim 1, wherein the information processing device includes means for identifying emotional information using an emotion analysis engine.

[0879] (Claim 3)

[0880] The system according to claim 1, wherein the generation algorithm includes means for performing a natural language processing task that generates a response based on emotional information.

[0881] "Application example 2 when combining with an emotional engine"

[0882] (Claim 1)

[0883] A means of receiving text data containing personal information at a data input device,

[0884] A means of using information processing means to detect personal information contained in the aforementioned text data,

[0885] A data processing means for removing or anonymizing and transforming the detected personal information,

[0886] A means for inputting the converted data and sentiment information into a generative model and performing a predetermined task,

[0887] Means for transmitting the output of the generation model to a data input device,

[0888] A means of using an emotion engine to analyze the user's emotions from the converted data,

[0889] A means for automatically generating a predetermined response based on the user's emotions,

[0890] A system that includes this.

[0891] (Claim 2)

[0892] The system according to claim 1, wherein the information processing means includes means for identifying personal information using a machine learning model.

[0893] (Claim 3)

[0894] The system according to claim 1, wherein the generative model includes means for performing a natural language processing task and generating a response in accordance with an emotional state. [Explanation of Symbols]

[0895] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving text data containing personal information at a data input device, A means of using an information processing device that detects personal information contained in the aforementioned text data, A data processing means for removing or anonymizing and transforming the detected personal information, A means for inputting the converted data into a generation model and executing a predetermined task, Means for transmitting the output of the generation model to a data input device, A system that includes this.

2. The system according to claim 1, wherein the information processing device includes means for identifying personal information using a machine learning model.

3. The system according to claim 1, wherein the generative model includes means for performing a natural language processing task.

Images