System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting location information from a plurality of terminals; means for concealing the collected location information; means for transmitting the concealed location information to a server; means for integrating the plurality of transmitted location information and updating a AI model; means for returning the updated AI model to the plurality of terminals; means for causing the terminals to further learn the returned AI model; and means for analyzing the location information based on the AI model learned by the terminals.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional data analysis methods require direct data sharing between companies, raising privacy concerns. While each company has a need to utilize other companies' data to improve the accuracy of their analysis, direct data sharing is difficult from the perspective of security and privacy. Furthermore, when handling sensitive data such as location information, a method is needed to effectively utilize the data while ensuring its security. The present invention aims to solve these problems and simultaneously achieve advanced data utilization and privacy protection. [Means for solving the problem]

[0005] The present invention provides a system that collects location information data from multiple devices, anonymizes the data, and transmits it to a server. The server then integrates the data and updates an AI model using a machine learning algorithm. The updated AI model is then returned to the device, where further learning takes place. This allows each company to evolve the model using its own data. Specifically, as described in claim 1, the system includes a location information data collection means, anonymization means, data transmission means, integration means, model update means, and analysis means. This eliminates the need for direct data sharing between companies, enabling effective use of location information data while protecting data privacy.

[0006] A "terminal" is an information processing device used by a user, and has the functions of collecting and concealing location information data and communicating with a server.

[0007] "Location data" is data that indicates a user's specific location and movement history based on information from GPS, Wi-Fi towers, etc.

[0008] "Anonymization" is the process of anonymizing or encrypting data to make it impossible to identify individuals.

[0009] A "server" is an information processing device that receives location information data sent from multiple devices, integrates it, and processes it to update the AI ​​model.

[0010] An "AI model" is a mathematical model that uses artificial intelligence technology to learn from data and make predictions and analyses.

[0011] A "machine learning algorithm" is an algorithm that automatically learns from data and makes predictions and classifications.

[0012] "Data transmission" is the process by which the terminal transmits collected data to the server.

[0013] "Integration" is the process of combining multiple data sets into a single, cohesive data set.

[0014] "Model updating" is the process of using new data to train an existing AI model and improve its performance.

[0015] "Analysis" is the process of using AI models to extract useful information and patterns from data. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention is a system that conceals location information data collected from multiple devices and transmits the data to a server to update an AI model. Here, we will explain in detail the program processing based on the roles of the server, device, and user.

[0038] Server-side processing

[0039] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[0040] The trained AI model is returned to each device. At this time, the returned model is encrypted and sent securely.

[0041] Terminal side processing

[0042] The device collects location data from smartphones. The collected data is anonymized or encrypted to prevent eavesdropping or tampering. When transmitting this anonymized data to a server, a secure communication protocol (e.g., HTTPS) is used to prevent data eavesdropping or tampering.

[0043] Once the device receives the AI ​​model returned from the server, it uses the model to further train its own location data. This training improves the accuracy of the model, allowing the device to use it to perform highly accurate location data analysis. Specifically, this is effective for area marketing and analyzing people flow during disasters, for example.

[0044] User-side processing

[0045] The user operates the device to collect location data from the smartphone. At this time, the data is appropriately anonymized and set up to be securely transmitted to the server. The AI ​​model returned from the server is then applied to the device to analyze the location data.

[0046] Users can check the analysis results and request that the AI ​​model be updated again if necessary. For example, a marketing department user may frequently retrain the model to simulate campaign effectiveness in a specific region in advance.

[0047] Specific examples

[0048] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. The server then returns the trained AI model to companies A and B.

[0049] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[0050] In this way, the present invention enables multiple companies to safely utilize their own data while conducting advanced location data analysis. It is also a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[0051] The processing flow will be explained below.

[0052] Step 1:

[0053] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[0054] Step 2:

[0055] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[0056] Step 3:

[0057] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[0058] Step 4:

[0059] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[0060] Step 5:

[0061] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[0062] Step 6:

[0063] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[0064] Step 7:

[0065] The server returns the trained AI model to each device. The model file is encrypted before transmission to prevent data tampering or eavesdropping.

[0066] Step 8:

[0067] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[0068] Step 9:

[0069] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[0070] Step 10:

[0071] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[0072] Step 11:

[0073] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[0074] Step 12:

[0075] Users can send data to the server again as needed, and the federated learning process can be repeated, incorporating additional data and the latest trends to further improve the accuracy of the analysis results.

[0076] Example 1

[0077] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0078] Conventional location data analysis systems face the challenge of performing highly accurate data analysis while ensuring the confidentiality and security of location data collected from multiple devices. In particular, secure data transmission and reception, preprocessing, integration, secure transmission of trained models, and highly accurate data analysis on each device are critical issues in practical operation. A system that provides effective solutions to these challenges is needed.

[0079] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0080] In this invention, the server includes means for collecting location information data from multiple terminals, means for concealing the collected location information data, means for transmitting the concealed location information data to the server, means for preprocessing the transmitted multiple location information data, means for updating an AI model based on the preprocessed location information data, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, means for encrypting and securely transmitting the trained model, and means for transmitting the location information data using a secure communication protocol. This enables secure concealment and transmission / reception of location information data collected from multiple terminals, preprocessing, updating of the AI ​​model, secure transmission of the trained model, and highly accurate data analysis.

[0081] A "terminal" is a device operated by a user that collects location data, conceals it, and transmits it to a server.

[0082] The "server" is a computer system that receives anonymized location data sent from multiple devices, preprocesses it, and updates the AI ​​model.

[0083] "Location Data" means data collected from a device that indicates its geographic location and is obtained through GPS or location services APIs.

[0084] "Redaction" is the process of anonymizing or encrypting location data to protect its privacy.

[0085] "Preprocessing" refers to the process of checking the consistency of the received location data, removing incomplete data and outliers, and normalizing the data.

[0086] An "AI model" is a model for making predictions and classifications that is built using machine learning algorithms based on location data.

[0087] "Updating" is the process of recalibrating the weights and parameters of an existing AI model using new location data.

[0088] A "trained model" is an AI model that has been trained on location data and is ready to be used for prediction or classification tasks.

[0089] "Encryption" is the process of converting data using a specific algorithm to make it unintelligible to third parties in order to protect the privacy of trained models and transmitted data.

[0090] A "secure communication protocol" is a set of communication rules for securely sending and receiving data, and uses technologies such as HTTPS and TLS.

[0091] The present invention provides a system for anonymizing location information data collected from multiple devices and transmitting the data to a server to update an AI model. Specific embodiments of the system are described below.

[0092] Server-side processing

[0093] The server receives location data sent from multiple devices. The received data is secure because it is encrypted using the Secured HTTPS protocol. The server first preprocesses the received data. For preprocessing, it uses Python's Pandas library to detect and remove incomplete data and outliers. It also normalizes the data using NumPy. The server updates the AI ​​model based on the preprocessed data. Machine learning frameworks such as TensorFlow and PyTorch are used to update the model. Once the training is complete, the model is encrypted and then securely sent to each device. The AES algorithm is used for encryption, and the HTTPS protocol is used for transmission.

[0094] Terminal side processing

[0095] The terminal collects location data from devices such as smartphones operated by the user. This is done using GPS functions or location service APIs. The collected data is first anonymized within the device. This anonymization is done by using Python's GeoPy library or the AES encryption algorithm. The anonymized data is then sent to the server using the HTTPS protocol. Once the trained AI model is returned from the server, the terminal uses the model to further train the location data locally. This training is done using machine learning libraries such as Scikit-learn and Onnx. Using the trained model enables highly accurate location data analysis, such as area marketing and people flow analysis during disasters.

[0096] User-side processing

[0097] The user operates the device to collect location data from their smartphone. At this time, the collected data is set to be appropriately anonymized and securely sent to the server. The AI ​​model returned from the server is applied to the device, and analysis is performed based on the location data. For example, a user on a business trip can find an appropriate route by analyzing movement patterns in a new area. If, based on the analysis results, it is determined that new data collection or improvement in model accuracy is necessary, the user can request an update of the AI ​​model again. An example prompt is, "I would like to update the AI ​​model to improve the efficiency of location data analysis in area marketing. Please explain in detail the process for retraining the model based on new location data and obtaining highly accurate predictions."

[0098] Specific examples

[0099] Suppose a company collects location data to analyze customer behavior. When a user configures their smartphone to collect location information, the device periodically acquires location information, anonymizes the data using the GeoPy library, and sends it to a server. The server receives the data, preprocesses it using Pandas and NumPy, and updates the model using TensorFlow. The new model is returned to each device via AES encryption, where it is trained again using Scikit-learn. This allows the company to perform highly accurate people flow analysis and optimize its marketing strategy. By repeating this cycle, operations that utilize data while protecting privacy can be achieved.

[0100] In this way, the system of the present invention enables highly accurate analysis of location data through data collection from multiple devices, secure data transmission, and updating and application of AI models.

[0101] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0102] Server-side processing

[0103] Step 1: Receiving location data

[0104] Input: Anonymized location data sent from multiple devices

[0105] Processing: The server receives the data in an encrypted form using the HTTPS protocol.

[0106] Output: Received anonymized location data

[0107] Specific operation: The server's receiving module monitors HTTPS communications, decrypts and stores the incoming data packets.

[0108] Step 2: Preprocessing the data

[0109] Input: Received anonymized location data

[0110] Processing: The server uses Python's Pandas library to detect and remove incomplete data and outliers, and also normalizes the data using NumPy.

[0111] Output: Preprocessed location data

[0112] Specific operation: A script is executed to fill in missing values ​​in the data, remove outliers, and scale numerical data.

[0113] Step 3: Update the AI ​​model

[0114] Input: Preprocessed geolocation data

[0115] Processing: The server uses TensorFlow to train the AI ​​model, learning a new model using all the data and optimizing the model weights.

[0116] Output: Updated AI model

[0117] What it does: The Python script calls the TensorFlow library and runs the process of training a model using the data as input.

[0118] Step 4: Encrypt the trained model

[0119] Input: Updated AI model

[0120] Processing: The server encrypts the model using the AES encryption algorithm.

[0121] Output: Encrypted AI model

[0122] Specific operation: A script is executed to encrypt files containing the weights and biases of the trained model using the AES algorithm.

[0123] Step 5: Send the encrypted model

[0124] Input: Encrypted AI model

[0125] Processing: The server sends the encrypted model to multiple devices using the HTTPS protocol.

[0126] Output: Trained model sent to multiple devices

[0127] Specific operation: HTTPS communication is initiated from the server to the device, and an encrypted model file is sent.

[0128] Terminal side processing

[0129] Step 1: Collect location data

[0130] Input: Smartphone GPS function and location information service API

[0131] Processing: The device periodically acquires location data using these functions.

[0132] Output: Collected location data

[0133] Specific operation: The device application calls the location information service API, periodically obtains location data, and stores it in an internal database.

[0134] Step 2: Data redaction

[0135] Input: Collected location data

[0136] Processing: The device anonymizes the data using the GeoPy library or uses the AES encryption algorithm.

[0137] Output: Anonymized location data

[0138] What it does: The collected data is processed by a Python script, anonymized or encrypted, and then stored as ready data.

[0139] Step 3: Sending data

[0140] Input: anonymized location data

[0141] Processing: The terminal sends data to the server using the HTTPS protocol.

[0142] Output: Location data sent to the server

[0143] Specific operation: The terminal's transmission module initiates HTTPS communication and sends the anonymized data to the server.

[0144] Step 4: Receive the trained model

[0145] Input: The encrypted AI model sent from the server

[0146] Processing: The terminal receives the model from the server using the HTTPS protocol and decrypts it using the AES algorithm.

[0147] Output: Decoded AI model

[0148] Specific operation: The terminal's receiving module monitors HTTPS communications and decrypts the encrypted files that arrive using the AES algorithm.

[0149] Step 5: Retraining the local data

[0150] Input: Decoded AI model and local location data

[0151] Processing: The device retrains the model on the local data using the Scikit-learn or Onnx library.

[0152] Output: Retrained AI model

[0153] What happens: A script on the device loads the decoded model and retrains it with locally collected data.

[0154] User-side processing

[0155] Step 1: Configure location data collection

[0156] Input: User settings information (permission of location services, frequency of collection, etc.)

[0157] Processing: The user operates the application on the device to set up location information collection.

[0158] Output: Start of collection process based on configuration information

[0159] Specific behavior: Collects settings through the user interface and activates the application's location collection function.

[0160] Step 2: Applying AI models to analyze data

[0161] Input: Updated AI model and collected location data

[0162] Processing: The device analyzes the location data using the received AI model. The user can then review the analysis results and use them for marketing strategies, etc.

[0163] Output: Analyzed data and results

[0164] What it does: Loads the updated model, runs analytical algorithms on the collected data, and displays the results.

[0165] Step 3: Request an update to the AI ​​model

[0166] Input: User feedback on analysis results and model accuracy

[0167] Processing: The user requests the server to update the AI ​​model again if necessary.

[0168] Output: Update request sent to the server

[0169] What it does: Collect feedback through the user interface and submit retraining requests as needed.

[0170] (Application example 1)

[0171] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0172] Current online shopping sites lack mechanisms for effectively utilizing users' location information to provide individually customized product recommendations and promotions. Furthermore, technologies for providing highly accurate recommendations while protecting user privacy are limited. This makes it difficult to provide users with the most appropriate product information and promotions in a timely manner. Furthermore, there is a need for the development of technologies that can effectively analyze and utilize location data while ensuring the security of user data.

[0173] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0174] In this invention, the server includes a means for integrating the transmitted multiple pieces of location information data and updating the AI ​​model, a means for returning the updated AI model to the multiple terminals, and a means for generating product recommendations and promotions based on the user's location based on the collected location information data, thereby enabling the provision of highly accurate product recommendations and promotions using location information data.

[0175] "Multiple terminals" are electronic devices used to collect and transmit location information data, such as multiple smartphones and personal computers.

[0176] "Location information data" refers to information about geographical locations collected from terminals, and is made up of numerical data such as latitude and longitude.

[0177] "Redaction" is the process of protecting the privacy of collected location data by anonymizing or encrypting it.

[0178] The "Server" is a central processing system for receiving collected location data, integrating the data, and updating and distributing the AI ​​model.

[0179] An "AI model" is a model that learns location data and is automatically adjusted using machine learning algorithms to make predictions and classifications.

[0180] An "updated AI model" is an AI model that has been trained and adjusted based on the latest location data, resulting in improved accuracy.

[0181] "Product recommendation" is the act of suggesting suitable products or services to a user based on the user's location information.

[0182] A "promotion" is a marketing method that notifies users of discounts and special offers on specific products and services.

[0183] "Notification" refers to the act of informing a user of generated product recommendations or promotions by displaying them on the user's terminal.

[0184] MODE FOR CARRYING OUT THE INVENTION

[0185] This invention is a system that anonymizes location information data collected from devices and sends it to a server to update an AI model. Specifically, the server receives location information data collected from multiple devices, preprocesses it, and updates the AI ​​model using a machine learning algorithm. The updated AI model is then sent back to multiple devices, where it is used for analyzing location information data, recommending products, and generating promotions.

[0186] Program generation and natural language explanation

[0187] 1. Hardware and Software:

[0188] Hardware:

[0189] Smartphones: Location information collection and user notification

[0190] Server: Data reception, preprocessing, AI model update and distribution

[0191] software:

[0192] Language: Python

[0193] Libraries: requests (HTTP requests), cryptography.fernet (data encryption)

[0194] Communication protocol: HTTPS (secure transmission of data)

[0195] Algorithm: Machine learning algorithm (e.g. TensorFlow, Scikit-learn)

[0196] 2. Data processing and calculation:

[0197] Terminal processing:

[0198] Location data is collected from smartphones, encrypted and anonymized, and then securely transmitted to a server using the HTTPS protocol.

[0199] Server side:

[0200] The server preprocesses the received location data, removing incomplete data and outliers and normalizing it. It then updates the AI ​​model using a machine learning algorithm based on the integrated location data. It then encrypts the updated AI model and sends it to each device.

[0201] Train the model:

[0202] Each device uses the updated AI model received from the server to retrain its own location data, enabling the device to analyze location data with high accuracy, recommend products, and generate promotions.

[0203] 3. Example:

[0204] For example, location information of users in Tokyo is collected and periodically sent to a server in an anonymous format. The server then uses this data to update the learning model and provide sales information and recommended products based on the user's location to the smartphone.

[0205] Prompt Sentence Examples

[0206] "We would like to generate an AI model that will enable users to receive personalized promotions and product recommendations based on their current location. Anonymizing and secure transmission of location data is essential. Please provide an example based on location data within Tokyo."

[0207] According to the above-described embodiment, the present invention makes it possible to effectively utilize user location information data while protecting privacy, and to provide highly accurate product recommendations and promotions.

[0208] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0209] Program processing flow

[0210] Step 1:

[0211] The user collects location data using a smartphone. The device uses the GPS function to obtain current latitude and longitude data and sends that data to an application. The input of this process is the user's current location (location data), and the output is the collected location data.

[0212] Step 2:

[0213] The device conceals the collected location data by encrypting it using an encryption algorithm (e.g., Fernet). The input is raw location data, and the output is encrypted data.

[0214] Step 3:

[0215] The device securely transmits the anonymized location data to the server using the HTTPS protocol. The input is the encrypted location data, and the output is a transmission confirmation message.

[0216] Step 4:

[0217] The server preprocesses the received location data, which includes removing incomplete data and outliers, and normalizing the data. The input is the encrypted location data, and the output is the preprocessed data.

[0218] Step 5:

[0219] The server integrates the preprocessed data and updates the AI ​​model using machine learning algorithms (e.g., TensorFlow, Scikit-learn). The input is the preprocessed location data, and the output is the updated AI model.

[0220] Step 6:

[0221] The server encrypts the updated AI model and returns it to each device. The input is the updated AI model, and the output is the encrypted AI model data.

[0222] Step 7:

[0223] The device decrypts the updated AI model received from the server and uses it to retrain its own location data. The input is the encrypted AI model data, and the output is the trained AI model.

[0224] Step 8:

[0225] The device uses the trained AI model to analyze the location data and generate product recommendations and promotions based on the user's location. The input is the trained AI model and location data, and the output is the generated product recommendations and promotion information.

[0226] Step 9:

[0227] The terminal notifies the user of the generated product recommendations and promotions. The input is the generated product recommendation and promotion information, and the output is a notification displayed on the user's smartphone.

[0228] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0229] This invention combines a system that conceals location information data collected from multiple devices, sends that data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail the program processing based on the roles of the server, device, and user.

[0230] Server-side processing

[0231] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[0232] The server also receives emotion recognition data sent from the device and uses it to optimize the AI ​​model. This emotion recognition data includes the user's emotional state as a parameter and is used to improve the accuracy of the model. The server then encrypts the trained AI model and returns it to each device.

[0233] Terminal side processing

[0234] The device collects location data from the smartphone. It uses GPS data and Wi-Fi tower information to obtain the user's current location and movement history. The collected data is anonymized or encrypted to prevent eavesdropping or tampering with the data when it is sent to the server.

[0235] Once the AI ​​model is returned from the server, the device uses it to further learn its own location data. In addition to this learning, the device also has a built-in emotion engine that recognizes the user's emotions, analyzing the user's voice data and facial expressions to recognize emotions. This emotion recognition data is then used to further refine the AI ​​model.

[0236] User-side processing

[0237] Users operate the device to collect location data from their smartphone. At this time, the data is appropriately anonymized and set up to be securely sent to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also sent to the server and used to update the AI ​​model.

[0238] The AI ​​model returned from the server is applied to the device to analyze the location data. The user can review the analysis results that incorporate emotional information and request that the AI ​​model be updated again if necessary. This system is extremely useful for marketing department users who want to simulate the effectiveness of campaigns in specific areas in advance, or for taking a personalized approach based on the user's emotional state.

[0239] Specific examples

[0240] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is then returned to companies A and B.

[0241] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[0242] In this way, the present invention enables multiple companies to safely utilize their own data while performing advanced location data analysis that incorporates emotional information. It also provides a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[0243] The processing flow will be explained below.

[0244] Step 1:

[0245] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[0246] Step 2:

[0247] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[0248] Step 3:

[0249] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[0250] Step 4:

[0251] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[0252] Step 5:

[0253] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[0254] Step 6:

[0255] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[0256] Step 7:

[0257] The device uses an emotion engine that recognizes the user's emotions and analyzes voice data and facial expressions to recognize the user's emotional state.

[0258] Step 8:

[0259] The device sends the user's emotion data recognized by the emotion engine to the server using a secure communication protocol.

[0260] Step 9:

[0261] The server uses the received emotion data to optimize the AI ​​model, using the emotion data as parameters to further improve the accuracy of the model.

[0262] Step 10:

[0263] The server encrypts the trained AI model and returns it to each device. The return process is performed using a secure communication protocol (e.g., HTTPS).

[0264] Step 11:

[0265] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[0266] Step 12:

[0267] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[0268] Step 13:

[0269] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[0270] Step 14:

[0271] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[0272] Step 15:

[0273] Users can send data to the server again as needed, and the federated learning process can be repeated to further improve the accuracy of the analysis results by incorporating additional data and the latest trends.

[0274] Example 2

[0275] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0276] Conventional location data collection systems lacked a method for effectively analyzing data while maintaining data anonymity and security. Furthermore, there was no way to integrate and analyze users' emotional states with location data, making it difficult to achieve more advanced personalization and evaluate marketing effectiveness. Furthermore, retraining models on each device and applying them efficiently was a challenge.

[0277] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, and means for analyzing the location information data based on the AI ​​model trained at the terminals. This enables advanced data analysis that takes user emotion information into account while maintaining the anonymity and security of the data. Furthermore, it also enables each terminal to retrain the model and efficiently apply it.

[0278] "Multiple terminals" refers to multiple user devices connected to the Internet, including smartphones, tablets, etc.

[0279] "Location data" refers to data about a user's current location and movement history obtained using GPS, Wi-Fi, etc.

[0280] "Redaction" is the process of anonymizing or encrypting data to protect it so that third parties cannot easily identify its contents.

[0281] A "server" is a computer system that receives, processes, and transmits data over a network, and refers to infrastructure capable of processing large amounts of data.

[0282] An "AI model" is a predictive model built using machine learning algorithms, and is a technology for data analysis and classification.

[0283] "Emotion recognition data" refers to data about a user's emotional state obtained by analyzing their vocal tone and facial expressions.

[0284] A "communication protocol" refers to the rules and methods for securely sending and receiving data, and includes HTTPS and TLS.

[0285] "Machine learning algorithm" is a general term for mathematical models and methods used to make predictions and classifications based on collected data.

[0286] "AES encryption" refers to a method of encrypting data using the Advanced Encryption Standard, providing a high level of security.

[0287] "Anonymization" is a general term for techniques that protect data privacy by removing personally identifiable information.

[0288] A "secure communication protocol" is a protocol that prevents eavesdropping or tampering with data when it is being sent or received.

[0289] This invention is a system that collects location information data from multiple devices, anonymizes the data, transmits it to a server, and updates an AI model. Furthermore, by combining it with user emotion recognition data, the accuracy of the model can be improved. The specific system configuration and operation procedure are described below.

[0290] Server-side processing

[0291] The server receives anonymized location data sent by multiple companies and individuals. The communication protocol used is HTTPS, which prevents data eavesdropping and tampering. The received data is first preprocessed to remove missing values ​​and outliers and normalize the data using the Pandas library. The specific data cleansing method used is the IQR (Interquartile Range) method.

[0292] Next, the AI ​​model is updated using the integrated location data. TensorFlow or PyTorch are typically used as machine learning frameworks. This allows for efficient learning of large amounts of data and improves the accuracy of the predictive model. For example, cluster analysis can be performed to identify user behavior patterns in specific areas.

[0293] The server also receives emotion recognition data sent from the device and integrates this data into the AI ​​model. Emotion data is obtained by analyzing voice tone and facial expressions and is added to the model as features. This process uses tools such as the Google Cloud Speech-to-Text API and OpenCV. Finally, the trained AI model is AES encrypted and securely returned to each device. The encryption is performed using the Python cryptography library.

[0294] Terminal side processing

[0295] The device uses a smartphone to collect location data. Specifically, it periodically obtains location information using Android or iOS APIs. The collected data is anonymized using RSA encryption technology. The anonymized data is sent to the server via the HTTPS protocol. At this time, a TLS certificate is used to ensure the security of the communication.

[0296] The device receives the AI ​​model returned from the server and further trains it on its own location data. Additional training is performed in a local environment using TensorFlow and other tools. The device also has a built-in emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This is done using voice recognition APIs and face recognition APIs. The analyzed emotion data is sent to the server and used to update the AI ​​model.

[0297] User-side processing

[0298] The user operates the device to collect location data from the smartphone. They confirm that the data collection is being carried out appropriately and that it is confidential. They then launch the location data collection app, turn on the GPS function, and begin data collection. The user then provides emotional data, such as voice and facial expressions, through the emotion engine. The device automatically recognizes the emotion and sends the data to the server.

[0299] The device receives the AI ​​model returned from the server and checks the results of the location data analysis. The analysis results, which take into account emotional information, can be viewed on a dashboard or app, feedback can be provided as needed, and an update of the AI ​​model can be requested.

[0300] Specific examples

[0301] For example, companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is returned to companies A and B. Company A's device uses the returned AI model to further train its own location data. This learning allows company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services.

[0302] Example prompts to input to the generative AI model

[0303] What are the specific steps to integrate location data from Company A and Company B to update the AI ​​model?

[0304] Please elaborate on how you integrate user emotion recognition data into your AI models.

[0305] Please provide specific steps for data redaction and secure transmission.

[0306] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0307] Step 1:

[0308] The server receives anonymized location data sent by companies or individuals. The input data is encrypted data sent via the HTTPS protocol. The server receives this data and saves it in storage. Specifically, the server takes the data into a receiving buffer and stores it in preparation for the next step of processing.

[0309] Step 2:

[0310] The server preprocesses the received location data. The input is the stored location data. The server uses the Pandas library to remove missing values ​​and outliers and normalize the data. The IQR (Interquartile Range) method is used to detect and eliminate outliers. The output is the preprocessed location data.

[0311] Step 3:

[0312] The server updates the AI ​​model using the preprocessed location data. The input is the combined preprocessed location data. The server applies machine learning algorithms using TensorFlow or PyTorch to train the model based on the new data. The output is an updated AI model.

[0313] Step 4:

[0314] The server receives emotion recognition data sent from the device and integrates it into the AI ​​model. The input is the emotion recognition data and the updated AI model, including voice tone and facial expression analysis data. The output is an optimized model that integrates the emotion data. This optimization is performed using the Google Cloud Speech-to-Text API and OpenCV.

[0315] Step 5:

[0316] The server encrypts the trained AI model using AES and returns it to each device. The input is an optimized model integrated with emotion data. The encryption is performed using Python's cryptography library. The output is an encrypted AI model file. The server sends this file to each device via the HTTPS protocol.

[0317] Step 6:

[0318] The device receives the encrypted AI model from the server. The input is the encrypted AI model file. The device receives this file and stores it in local storage. Specifically, the device first confirms receipt of the file and stores it in the appropriate directory.

[0319] Step 7:

[0320] The device deserializes and loads the received AI model to further train it on its own location data. The input is the encrypted AI model file and local location data. The device uses code such as TensorFlow to unpack the model and retrain it with the new data. The output is a locally updated AI model.

[0321] Step 8:

[0322] The device uses an emotion recognition engine to collect user emotional data. Inputs include the user's voice data and facial expression video. The device analyzes this to recognize emotions and saves them as data. Specifically, it captures data in real time using a camera or microphone and applies an analysis algorithm.

[0323] Step 9:

[0324] The device encrypts the emotion data it recognizes and sends it to the server. The input is the emotion recognition data obtained. The device encrypts the data using RSA encryption technology and sends it to the server via the HTTPS protocol. The output is the encrypted emotion data.

[0325] Step 10:

[0326] The user operates the device to collect location data in real time. The input is the user's settings and operations. The user launches a location collection app and enables GPS and WiFi data. The output is the collected location data.

[0327] Step 11:

[0328] The user provides emotional data through the emotion engine. The input is the user's voice and facial expressions. The device analyzes this to recognize the emotion and stores it for transmission in the next step. The output is the analyzed emotional data.

[0329] Step 12:

[0330] The user checks the analysis results on a dashboard or app. The input is the analysis results of the AI ​​model returned from the server. The user visually checks the analysis results of the location data that take into account emotional information and provides feedback as needed. The output is the user's understanding of the results and feedback.

[0331] (Application example 2)

[0332] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0333] Conventional autonomous vehicle systems have been unable to adjust vehicle settings in real time based on passengers' emotional states, and lack a means to effectively combine location information data and emotion recognition data collected from multiple devices to improve safety and comfort.

[0334] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data transmitted from the terminals and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, and means for adjusting vehicle settings in accordance with passenger emotions. This makes it possible to adjust vehicle settings in real time in accordance with the emotional state of passengers and achieve highly safe and comfortable autonomous driving.

[0335] "Multiple terminals" refers to multiple electronic devices, and in particular to portable information processing devices such as smartphones and tablets.

[0336] "Location data" refers to data including location coordinates and movement history obtained from GPS and Wi-Fi towers.

[0337] "Redaction" refers to the process of making personally identifiable information unidentifiable by anonymizing or encrypting data.

[0338] "Server" refers to a computer system that stores, processes, and distributes data over a network.

[0339] "Emotion recognition data" is data that indicates the emotional state of the user, obtained using voice data and facial expression analysis.

[0340] "AI model" refers to an artificial intelligence model created using machine learning algorithms that learns the characteristics of data and makes predictions and classifications.

[0341] "Updating" refers to the process of retraining an existing model based on new data to improve its accuracy and performance.

[0342] "Optimizing" means adjusting various parameters and factors to maximize the performance of a model.

[0343] "Returning" is the operation of sending data or models processed by the server back to the original terminal.

[0344] "Training" refers to the process of using newly collected data on the device to further train the AI ​​model and improve its performance.

[0345] "Analyzing" means applying statistical methods and algorithms to interpret collected data and extract useful information.

[0346] "Adjusting vehicle settings based on passenger emotions" means taking actions such as changing the route or playing relaxing music if a passenger is feeling stressed.

[0347] The present invention combines a system that anonymizes location information data collected from multiple devices, transmits the data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail an embodiment based on the roles of the server, device, and user.

[0348] Server-side processing

[0349] The server receives anonymized location data sent by the company. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. Next, this location data is integrated with emotion recognition data to update the AI ​​model. To update the model, a machine learning algorithm is used to learn from the data and improve the accuracy of predictions and classifications. The updated AI model is then encrypted and returned to each device.

[0350] Specific software used on the server side includes machine learning frameworks such as TensorFlow and PyTorch, and SSL / TLS protocols are used to keep data confidential.

[0351] Terminal side processing

[0352] The device collects location data from smartphones and tablets, including using GPS and Wi-Fi tower information to obtain a user's current location and movement history, and includes measures to anonymize or encrypt the data. The anonymized data is then sent to a server using a secure communications protocol (e.g., HTTPS).

[0353] The device that receives the returned AI model uses it to further train its own location data. The device also has an emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This emotion recognition data is used to adjust the vehicle's settings in real time according to the user's condition.

[0354] The specific hardware used in the device includes a GPS module, camera, and microphone, and the emotion engine uses Azure Cognitive Services and Google Cloud Vision.

[0355] User-side processing

[0356] Users operate their devices to collect location data from their smartphones or tablets. At this time, the data is appropriately anonymized and securely transmitted to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also transmitted to the server and used to update the AI ​​model.

[0357] For example, if a user is riding in a self-driving vehicle, the vehicle's settings can be adjusted in real time based on the user's emotional state: for example, if the passenger is feeling stressed, the vehicle can change to a route with less traffic or play relaxing music.

[0358] Prompt Sentence Examples

[0359] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[0360] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[0361] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0362] Step 1:

[0363] The server receives location data and emotion recognition data from multiple devices. It receives anonymized location data and emotion recognition data as input and stores them in a database. Specifically, it checks the data structure when receiving the data and removes outliers and incomplete data.

[0364] Step 2:

[0365] The server preprocesses the received data. It uses location data and emotion recognition data as input and normalizes them. It standardizes the data format and performs processing to impute missing values. Specifically, it scales the range of numerical data and encodes categorical data.

[0366] Step 3:

[0367] The server updates the AI ​​model based on the preprocessed data. Using the normalized location data and emotion recognition data as input, it applies machine learning algorithms to retrain the model, improving the prediction and classification accuracy of the AI ​​model. Specifically, it tunes the model's hyperparameters and trains it until convergence is achieved.

[0368] Step 4:

[0369] The server encrypts the updated AI model and returns it to each device. It uses the latest AI model and each device's identification information as input and distributes the model using a secure communication protocol (e.g., HTTPS). Specifically, it generates an encryption key and encodes the model data.

[0370] Step 5:

[0371] The device receives the AI ​​model returned from the server and further trains it using its own data. It uses the received AI model and the location and emotion recognition data stored on the device as input. Specifically, it performs incremental training to improve the adaptability of the model.

[0372] Step 6:

[0373] The device analyzes real-time location and emotion recognition data using the latest AI models to adjust vehicle settings. It uses the passenger's current location and emotional state as input to optimize the route and environment. Specifically, it will select a route with less traffic and play relaxing music if it detects stress.

[0374] Step 7:

[0375] Users operate their devices to collect location data in real time. GPS modules and Wi-Fi information are used as input to obtain current locations and movement histories. The collected data is then anonymized or encrypted and stored in the device's memory.

[0376] Step 8:

[0377] The user recognizes their own emotional state using the device's emotion engine. Voice data and facial expressions are acquired as input from a camera or microphone, and an emotion recognition algorithm is applied. Specifically, the emotional state is extracted as a parameter and stored in a database.

[0378] Prompt Sentence Examples

[0379] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[0380] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[0381] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0382] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0383] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0384] [Second embodiment]

[0385] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0386] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0387] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0388] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0389] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0390] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0391] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0392] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0393] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0394] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0395] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0396] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0397] This invention is a system that conceals location information data collected from multiple devices and transmits the data to a server to update an AI model. Here, we will explain in detail the program processing based on the roles of the server, device, and user.

[0398] Server-side processing

[0399] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[0400] The trained AI model is returned to each device. At this time, the returned model is encrypted and sent securely.

[0401] Terminal side processing

[0402] The device collects location data from smartphones. The collected data is anonymized or encrypted to prevent eavesdropping or tampering. When transmitting this anonymized data to a server, a secure communication protocol (e.g., HTTPS) is used to prevent data eavesdropping or tampering.

[0403] Once the device receives the AI ​​model returned from the server, it uses the model to further train its own location data. This training improves the accuracy of the model, allowing the device to use it to perform highly accurate location data analysis. Specifically, this is effective for area marketing and analyzing people flow during disasters, for example.

[0404] User-side processing

[0405] The user operates the device to collect location data from the smartphone. At this time, the data is appropriately anonymized and set up to be securely transmitted to the server. The AI ​​model returned from the server is then applied to the device to analyze the location data.

[0406] Users can check the analysis results and request that the AI ​​model be updated again if necessary. For example, a marketing department user may frequently retrain the model to simulate campaign effectiveness in a specific region in advance.

[0407] Specific examples

[0408] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. The server then returns the trained AI model to companies A and B.

[0409] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[0410] In this way, the present invention enables multiple companies to safely utilize their own data while conducting advanced location data analysis. It is also a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[0411] The processing flow will be explained below.

[0412] Step 1:

[0413] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[0414] Step 2:

[0415] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[0416] Step 3:

[0417] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[0418] Step 4:

[0419] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[0420] Step 5:

[0421] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[0422] Step 6:

[0423] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[0424] Step 7:

[0425] The server returns the trained AI model to each device. The model file is encrypted before transmission to prevent data tampering or eavesdropping.

[0426] Step 8:

[0427] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[0428] Step 9:

[0429] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[0430] Step 10:

[0431] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[0432] Step 11:

[0433] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[0434] Step 12:

[0435] Users can send data to the server again as needed, and the federated learning process can be repeated, incorporating additional data and the latest trends to further improve the accuracy of the analysis results.

[0436] Example 1

[0437] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0438] Conventional location data analysis systems face the challenge of performing highly accurate data analysis while ensuring the confidentiality and security of location data collected from multiple devices. In particular, secure data transmission and reception, preprocessing, integration, secure transmission of trained models, and highly accurate data analysis on each device are critical issues in practical operation. A system that provides effective solutions to these challenges is needed.

[0439] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0440] In this invention, the server includes means for collecting location information data from multiple terminals, means for concealing the collected location information data, means for transmitting the concealed location information data to the server, means for preprocessing the transmitted multiple location information data, means for updating an AI model based on the preprocessed location information data, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, means for encrypting and securely transmitting the trained model, and means for transmitting the location information data using a secure communication protocol. This enables secure concealment and transmission / reception of location information data collected from multiple terminals, preprocessing, updating of the AI ​​model, secure transmission of the trained model, and highly accurate data analysis.

[0441] A "terminal" is a device operated by a user that collects location data, conceals it, and transmits it to a server.

[0442] The "server" is a computer system that receives anonymized location data sent from multiple devices, preprocesses it, and updates the AI ​​model.

[0443] "Location Data" means data collected from a device that indicates its geographic location and is obtained through GPS or location services APIs.

[0444] "Redaction" is the process of anonymizing or encrypting location data to protect its privacy.

[0445] "Preprocessing" refers to the process of checking the consistency of the received location data, removing incomplete data and outliers, and normalizing the data.

[0446] An "AI model" is a model for making predictions and classifications that is built using machine learning algorithms based on location data.

[0447] "Updating" is the process of recalibrating the weights and parameters of an existing AI model using new location data.

[0448] A "trained model" is an AI model that has been trained on location data and is ready to be used for prediction or classification tasks.

[0449] "Encryption" is the process of converting data using a specific algorithm to make it unintelligible to third parties in order to protect the privacy of trained models and transmitted data.

[0450] A "secure communication protocol" is a set of communication rules for securely sending and receiving data, and uses technologies such as HTTPS and TLS.

[0451] The present invention provides a system for anonymizing location information data collected from multiple devices and transmitting the data to a server to update an AI model. Specific embodiments of the system are described below.

[0452] Server-side processing

[0453] The server receives location data sent from multiple devices. The received data is secure because it is encrypted using the Secured HTTPS protocol. The server first preprocesses the received data. For preprocessing, it uses Python's Pandas library to detect and remove incomplete data and outliers. It also normalizes the data using NumPy. The server updates the AI ​​model based on the preprocessed data. Machine learning frameworks such as TensorFlow and PyTorch are used to update the model. Once the training is complete, the model is encrypted and then securely sent to each device. The AES algorithm is used for encryption, and the HTTPS protocol is used for transmission.

[0454] Terminal side processing

[0455] The terminal collects location data from devices such as smartphones operated by the user. This is done using GPS functions or location service APIs. The collected data is first anonymized within the device. This anonymization is done by using Python's GeoPy library or the AES encryption algorithm. The anonymized data is then sent to the server using the HTTPS protocol. Once the trained AI model is returned from the server, the terminal uses the model to further train the location data locally. This training is done using machine learning libraries such as Scikit-learn and Onnx. Using the trained model enables highly accurate location data analysis, such as area marketing and people flow analysis during disasters.

[0456] User-side processing

[0457] The user operates the device to collect location data from their smartphone. At this time, the collected data is set to be appropriately anonymized and securely sent to the server. The AI ​​model returned from the server is applied to the device, and analysis is performed based on the location data. For example, a user on a business trip can find an appropriate route by analyzing movement patterns in a new area. If, based on the analysis results, it is determined that new data collection or improvement in model accuracy is necessary, the user can request an update of the AI ​​model again. An example prompt is, "I would like to update the AI ​​model to improve the efficiency of location data analysis in area marketing. Please explain in detail the process for retraining the model based on new location data and obtaining highly accurate predictions."

[0458] Specific examples

[0459] Suppose a company collects location data to analyze customer behavior. When a user configures their smartphone to collect location information, the device periodically acquires location information, anonymizes the data using the GeoPy library, and sends it to a server. The server receives the data, preprocesses it using Pandas and NumPy, and updates the model using TensorFlow. The new model is returned to each device via AES encryption, where it is trained again using Scikit-learn. This allows the company to perform highly accurate people flow analysis and optimize its marketing strategy. By repeating this cycle, operations that utilize data while protecting privacy can be achieved.

[0460] In this way, the system of the present invention enables highly accurate analysis of location data through data collection from multiple devices, secure data transmission, and updating and application of AI models.

[0461] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0462] Server-side processing

[0463] Step 1: Receiving location data

[0464] Input: Anonymized location data sent from multiple devices

[0465] Processing: The server receives the data in an encrypted form using the HTTPS protocol.

[0466] Output: Received anonymized location data

[0467] Specific operation: The server's receiving module monitors HTTPS communications, decrypts and stores the incoming data packets.

[0468] Step 2: Preprocessing the data

[0469] Input: Received anonymized location data

[0470] Processing: The server uses Python's Pandas library to detect and remove incomplete data and outliers, and also normalizes the data using NumPy.

[0471] Output: Preprocessed location data

[0472] Specific operation: A script is executed to fill in missing values ​​in the data, remove outliers, and scale numerical data.

[0473] Step 3: Update the AI ​​model

[0474] Input: Preprocessed geolocation data

[0475] Processing: The server uses TensorFlow to train the AI ​​model, learning a new model using all the data and optimizing the model weights.

[0476] Output: Updated AI model

[0477] What it does: The Python script calls the TensorFlow library and runs the process of training a model using the data as input.

[0478] Step 4: Encrypt the trained model

[0479] Input: Updated AI model

[0480] Processing: The server encrypts the model using the AES encryption algorithm.

[0481] Output: Encrypted AI model

[0482] Specific operation: A script is executed to encrypt files containing the weights and biases of the trained model using the AES algorithm.

[0483] Step 5: Send the encrypted model

[0484] Input: Encrypted AI model

[0485] Processing: The server sends the encrypted model to multiple devices using the HTTPS protocol.

[0486] Output: Trained model sent to multiple devices

[0487] Specific operation: HTTPS communication is initiated from the server to the device, and an encrypted model file is sent.

[0488] Terminal side processing

[0489] Step 1: Collect location data

[0490] Input: Smartphone GPS function and location information service API

[0491] Processing: The device periodically acquires location data using these functions.

[0492] Output: Collected location data

[0493] Specific operation: The device application calls the location information service API, periodically obtains location data, and stores it in an internal database.

[0494] Step 2: Data redaction

[0495] Input: Collected location data

[0496] Processing: The device anonymizes the data using the GeoPy library or uses the AES encryption algorithm.

[0497] Output: Anonymized location data

[0498] What it does: The collected data is processed by a Python script, anonymized or encrypted, and then stored as ready data.

[0499] Step 3: Sending data

[0500] Input: anonymized location data

[0501] Processing: The terminal sends data to the server using the HTTPS protocol.

[0502] Output: Location data sent to the server

[0503] Specific operation: The terminal's transmission module initiates HTTPS communication and sends the anonymized data to the server.

[0504] Step 4: Receive the trained model

[0505] Input: The encrypted AI model sent from the server

[0506] Processing: The terminal receives the model from the server using the HTTPS protocol and decrypts it using the AES algorithm.

[0507] Output: Decoded AI model

[0508] Specific operation: The terminal's receiving module monitors HTTPS communications and decrypts the encrypted files that arrive using the AES algorithm.

[0509] Step 5: Retraining the local data

[0510] Input: Decoded AI model and local location data

[0511] Processing: The device retrains the model on the local data using the Scikit-learn or Onnx library.

[0512] Output: Retrained AI model

[0513] What happens: A script on the device loads the decoded model and retrains it with locally collected data.

[0514] User-side processing

[0515] Step 1: Configure location data collection

[0516] Input: User settings information (permission of location services, frequency of collection, etc.)

[0517] Processing: The user operates the application on the device to set up location information collection.

[0518] Output: Start of collection process based on configuration information

[0519] Specific behavior: Collects settings through the user interface and activates the application's location collection function.

[0520] Step 2: Applying AI models to analyze data

[0521] Input: Updated AI model and collected location data

[0522] Processing: The device analyzes the location data using the received AI model. The user can then review the analysis results and use them for marketing strategies, etc.

[0523] Output: Analyzed data and results

[0524] What it does: Loads the updated model, runs analytical algorithms on the collected data, and displays the results.

[0525] Step 3: Request an update to the AI ​​model

[0526] Input: User feedback on analysis results and model accuracy

[0527] Processing: The user requests the server to update the AI ​​model again if necessary.

[0528] Output: Update request sent to the server

[0529] What it does: Collect feedback through the user interface and submit retraining requests as needed.

[0530] (Application example 1)

[0531] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0532] Current online shopping sites lack mechanisms for effectively utilizing users' location information to provide individually customized product recommendations and promotions. Furthermore, technologies for providing highly accurate recommendations while protecting user privacy are limited. This makes it difficult to provide users with the most appropriate product information and promotions in a timely manner. Furthermore, there is a need for the development of technologies that can effectively analyze and utilize location data while ensuring the security of user data.

[0533] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0534] In this invention, the server includes a means for integrating the transmitted multiple pieces of location information data and updating the AI ​​model, a means for returning the updated AI model to the multiple terminals, and a means for generating product recommendations and promotions based on the user's location based on the collected location information data, thereby enabling the provision of highly accurate product recommendations and promotions using location information data.

[0535] "Multiple terminals" are electronic devices used to collect and transmit location information data, such as multiple smartphones and personal computers.

[0536] "Location information data" refers to information about geographical locations collected from terminals, and is made up of numerical data such as latitude and longitude.

[0537] "Redaction" is the process of protecting the privacy of collected location data by anonymizing or encrypting it.

[0538] The "Server" is a central processing system for receiving collected location data, integrating the data, and updating and distributing the AI ​​model.

[0539] An "AI model" is a model that learns location data and is automatically adjusted using machine learning algorithms to make predictions and classifications.

[0540] An "updated AI model" is an AI model that has been trained and adjusted based on the latest location data, resulting in improved accuracy.

[0541] "Product recommendation" is the act of suggesting suitable products or services to a user based on the user's location information.

[0542] A "promotion" is a marketing method that notifies users of discounts and special offers on specific products and services.

[0543] "Notification" refers to the act of informing a user of generated product recommendations or promotions by displaying them on the user's terminal.

[0544] MODE FOR CARRYING OUT THE INVENTION

[0545] This invention is a system that anonymizes location information data collected from devices and sends it to a server to update an AI model. Specifically, the server receives location information data collected from multiple devices, preprocesses it, and updates the AI ​​model using a machine learning algorithm. The updated AI model is then sent back to multiple devices, where it is used for analyzing location information data, recommending products, and generating promotions.

[0546] Program generation and natural language explanation

[0547] 1. Hardware and Software:

[0548] Hardware:

[0549] Smartphones: Location information collection and user notification

[0550] Server: Data reception, preprocessing, AI model update and distribution

[0551] software:

[0552] Language: Python

[0553] Libraries: requests (HTTP requests), cryptography.fernet (data encryption)

[0554] Communication protocol: HTTPS (secure transmission of data)

[0555] Algorithm: Machine learning algorithm (e.g. TensorFlow, Scikit-learn)

[0556] 2. Data processing and calculation:

[0557] Terminal processing:

[0558] Location data is collected from smartphones, encrypted and anonymized, and then securely transmitted to a server using the HTTPS protocol.

[0559] Server side:

[0560] The server preprocesses the received location data, removing incomplete data and outliers and normalizing it. It then updates the AI ​​model using a machine learning algorithm based on the integrated location data. It then encrypts the updated AI model and sends it to each device.

[0561] Train the model:

[0562] Each device uses the updated AI model received from the server to retrain its own location data, enabling the device to analyze location data with high accuracy, recommend products, and generate promotions.

[0563] 3. Example:

[0564] For example, location information of users in Tokyo is collected and periodically sent to a server in an anonymous format. The server then uses this data to update the learning model and provide sales information and recommended products based on the user's location to the smartphone.

[0565] Prompt Sentence Examples

[0566] "We would like to generate an AI model that will enable users to receive personalized promotions and product recommendations based on their current location. Anonymizing and secure transmission of location data is essential. Please provide an example based on location data within Tokyo."

[0567] According to the above-described embodiment, the present invention makes it possible to effectively utilize user location information data while protecting privacy, and to provide highly accurate product recommendations and promotions.

[0568] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0569] Program processing flow

[0570] Step 1:

[0571] The user collects location data using a smartphone. The device uses the GPS function to obtain current latitude and longitude data and sends that data to an application. The input of this process is the user's current location (location data), and the output is the collected location data.

[0572] Step 2:

[0573] The device conceals the collected location data by encrypting it using an encryption algorithm (e.g., Fernet). The input is raw location data, and the output is encrypted data.

[0574] Step 3:

[0575] The device securely transmits the anonymized location data to the server using the HTTPS protocol. The input is the encrypted location data, and the output is a transmission confirmation message.

[0576] Step 4:

[0577] The server preprocesses the received location data, which includes removing incomplete data and outliers, and normalizing the data. The input is the encrypted location data, and the output is the preprocessed data.

[0578] Step 5:

[0579] The server integrates the preprocessed data and updates the AI ​​model using machine learning algorithms (e.g., TensorFlow, Scikit-learn). The input is the preprocessed location data, and the output is the updated AI model.

[0580] Step 6:

[0581] The server encrypts the updated AI model and returns it to each device. The input is the updated AI model, and the output is the encrypted AI model data.

[0582] Step 7:

[0583] The device decrypts the updated AI model received from the server and uses it to retrain its own location data. The input is the encrypted AI model data, and the output is the trained AI model.

[0584] Step 8:

[0585] The device uses the trained AI model to analyze the location data and generate product recommendations and promotions based on the user's location. The input is the trained AI model and location data, and the output is the generated product recommendations and promotion information.

[0586] Step 9:

[0587] The terminal notifies the user of the generated product recommendations and promotions. The input is the generated product recommendation and promotion information, and the output is a notification displayed on the user's smartphone.

[0588] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0589] This invention combines a system that conceals location information data collected from multiple devices, sends that data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail the program processing based on the roles of the server, device, and user.

[0590] Server-side processing

[0591] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[0592] The server also receives emotion recognition data sent from the device and uses it to optimize the AI ​​model. This emotion recognition data includes the user's emotional state as a parameter and is used to improve the accuracy of the model. The server then encrypts the trained AI model and returns it to each device.

[0593] Terminal side processing

[0594] The device collects location data from the smartphone. It uses GPS data and Wi-Fi tower information to obtain the user's current location and movement history. The collected data is anonymized or encrypted to prevent eavesdropping or tampering with the data when it is sent to the server.

[0595] Once the AI ​​model is returned from the server, the device uses it to further learn its own location data. In addition to this learning, the device also has a built-in emotion engine that recognizes the user's emotions, analyzing the user's voice data and facial expressions to recognize emotions. This emotion recognition data is then used to further refine the AI ​​model.

[0596] User-side processing

[0597] Users operate the device to collect location data from their smartphone. At this time, the data is appropriately anonymized and set up to be securely sent to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also sent to the server and used to update the AI ​​model.

[0598] The AI ​​model returned from the server is applied to the device to analyze the location data. The user can review the analysis results that incorporate emotional information and request that the AI ​​model be updated again if necessary. This system is extremely useful for marketing department users who want to simulate the effectiveness of campaigns in specific areas in advance, or for taking a personalized approach based on the user's emotional state.

[0599] Specific examples

[0600] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is then returned to companies A and B.

[0601] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[0602] In this way, the present invention enables multiple companies to safely utilize their own data while performing advanced location data analysis that incorporates emotional information. It also provides a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[0603] The processing flow will be explained below.

[0604] Step 1:

[0605] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[0606] Step 2:

[0607] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[0608] Step 3:

[0609] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[0610] Step 4:

[0611] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[0612] Step 5:

[0613] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[0614] Step 6:

[0615] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[0616] Step 7:

[0617] The device uses an emotion engine that recognizes the user's emotions and analyzes voice data and facial expressions to recognize the user's emotional state.

[0618] Step 8:

[0619] The device sends the user's emotion data recognized by the emotion engine to the server using a secure communication protocol.

[0620] Step 9:

[0621] The server uses the received emotion data to optimize the AI ​​model, using the emotion data as parameters to further improve the accuracy of the model.

[0622] Step 10:

[0623] The server encrypts the trained AI model and returns it to each device. The return process is performed using a secure communication protocol (e.g., HTTPS).

[0624] Step 11:

[0625] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[0626] Step 12:

[0627] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[0628] Step 13:

[0629] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[0630] Step 14:

[0631] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[0632] Step 15:

[0633] Users can send data to the server again as needed, and the federated learning process can be repeated to further improve the accuracy of the analysis results by incorporating additional data and the latest trends.

[0634] Example 2

[0635] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0636] Conventional location data collection systems lacked a method for effectively analyzing data while maintaining data anonymity and security. Furthermore, there was no way to integrate and analyze users' emotional states with location data, making it difficult to achieve more advanced personalization and evaluate marketing effectiveness. Furthermore, retraining models on each device and applying them efficiently was a challenge.

[0637] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, and means for analyzing the location information data based on the AI ​​model trained at the terminals. This enables advanced data analysis that takes user emotion information into account while maintaining the anonymity and security of the data. Furthermore, it also enables each terminal to retrain the model and efficiently apply it.

[0638] "Multiple terminals" refers to multiple user devices connected to the Internet, including smartphones, tablets, etc.

[0639] "Location data" refers to data about a user's current location and movement history obtained using GPS, Wi-Fi, etc.

[0640] "Redaction" is the process of anonymizing or encrypting data to protect it so that third parties cannot easily identify its contents.

[0641] A "server" is a computer system that receives, processes, and transmits data over a network, and refers to infrastructure capable of processing large amounts of data.

[0642] An "AI model" is a predictive model built using machine learning algorithms, and is a technology for data analysis and classification.

[0643] "Emotion recognition data" refers to data about a user's emotional state obtained by analyzing their vocal tone and facial expressions.

[0644] A "communication protocol" refers to the rules and methods for securely sending and receiving data, and includes HTTPS and TLS.

[0645] "Machine learning algorithm" is a general term for mathematical models and methods used to make predictions and classifications based on collected data.

[0646] "AES encryption" refers to a method of encrypting data using the Advanced Encryption Standard, providing a high level of security.

[0647] "Anonymization" is a general term for techniques that protect data privacy by removing personally identifiable information.

[0648] A "secure communication protocol" is a protocol that prevents eavesdropping or tampering with data when it is being sent or received.

[0649] This invention is a system that collects location information data from multiple devices, anonymizes the data, transmits it to a server, and updates an AI model. Furthermore, by combining it with user emotion recognition data, the accuracy of the model can be improved. The specific system configuration and operation procedure are described below.

[0650] Server-side processing

[0651] The server receives anonymized location data sent by multiple companies and individuals. The communication protocol used is HTTPS, which prevents data eavesdropping and tampering. The received data is first preprocessed to remove missing values ​​and outliers and normalize the data using the Pandas library. The specific data cleansing method used is the IQR (Interquartile Range) method.

[0652] Next, the AI ​​model is updated using the integrated location data. TensorFlow or PyTorch are typically used as machine learning frameworks. This allows for efficient learning of large amounts of data and improves the accuracy of the predictive model. For example, cluster analysis can be performed to identify user behavior patterns in specific areas.

[0653] The server also receives emotion recognition data sent from the device and integrates this data into the AI ​​model. Emotion data is obtained by analyzing voice tone and facial expressions and is added to the model as features. This process uses tools such as the Google Cloud Speech-to-Text API and OpenCV. Finally, the trained AI model is AES encrypted and securely returned to each device. The encryption is performed using the Python cryptography library.

[0654] Terminal side processing

[0655] The device uses a smartphone to collect location data. Specifically, it periodically obtains location information using Android or iOS APIs. The collected data is anonymized using RSA encryption technology. The anonymized data is sent to the server via the HTTPS protocol. At this time, a TLS certificate is used to ensure the security of the communication.

[0656] The device receives the AI ​​model returned from the server and further trains it on its own location data. Additional training is performed in a local environment using TensorFlow and other tools. The device also has a built-in emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This is done using voice recognition APIs and face recognition APIs. The analyzed emotion data is sent to the server and used to update the AI ​​model.

[0657] User-side processing

[0658] The user operates the device to collect location data from the smartphone. They confirm that the data collection is being carried out appropriately and that it is confidential. They then launch the location data collection app, turn on the GPS function, and begin data collection. The user then provides emotional data, such as voice and facial expressions, through the emotion engine. The device automatically recognizes the emotion and sends the data to the server.

[0659] The device receives the AI ​​model returned from the server and checks the results of the location data analysis. The analysis results, which take into account emotional information, can be viewed on a dashboard or app, feedback can be provided as needed, and an update of the AI ​​model can be requested.

[0660] Specific examples

[0661] For example, companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is returned to companies A and B. Company A's device uses the returned AI model to further train its own location data. This learning allows company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services.

[0662] Example prompts to input to the generative AI model

[0663] What are the specific steps to integrate location data from Company A and Company B to update the AI ​​model?

[0664] Please elaborate on how you integrate user emotion recognition data into your AI models.

[0665] Please provide specific steps for data redaction and secure transmission.

[0666] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0667] Step 1:

[0668] The server receives anonymized location data sent by companies or individuals. The input data is encrypted data sent via the HTTPS protocol. The server receives this data and saves it in storage. Specifically, the server takes the data into a receiving buffer and stores it in preparation for the next step of processing.

[0669] Step 2:

[0670] The server preprocesses the received location data. The input is the stored location data. The server uses the Pandas library to remove missing values ​​and outliers and normalize the data. The IQR (Interquartile Range) method is used to detect and eliminate outliers. The output is the preprocessed location data.

[0671] Step 3:

[0672] The server updates the AI ​​model using the preprocessed location data. The input is the combined preprocessed location data. The server applies machine learning algorithms using TensorFlow or PyTorch to train the model based on the new data. The output is an updated AI model.

[0673] Step 4:

[0674] The server receives emotion recognition data sent from the device and integrates it into the AI ​​model. The input is the emotion recognition data and the updated AI model, including voice tone and facial expression analysis data. The output is an optimized model that integrates the emotion data. This optimization is performed using the Google Cloud Speech-to-Text API and OpenCV.

[0675] Step 5:

[0676] The server encrypts the trained AI model using AES and returns it to each device. The input is an optimized model integrated with emotion data. The encryption is performed using Python's cryptography library. The output is an encrypted AI model file. The server sends this file to each device via the HTTPS protocol.

[0677] Step 6:

[0678] The device receives the encrypted AI model from the server. The input is the encrypted AI model file. The device receives this file and stores it in local storage. Specifically, the device first confirms receipt of the file and stores it in the appropriate directory.

[0679] Step 7:

[0680] The device deserializes and loads the received AI model to further train it on its own location data. The input is the encrypted AI model file and local location data. The device uses code such as TensorFlow to unpack the model and retrain it with the new data. The output is a locally updated AI model.

[0681] Step 8:

[0682] The device uses an emotion recognition engine to collect user emotional data. Inputs include the user's voice data and facial expression video. The device analyzes this to recognize emotions and saves them as data. Specifically, it captures data in real time using a camera or microphone and applies an analysis algorithm.

[0683] Step 9:

[0684] The device encrypts the emotion data it recognizes and sends it to the server. The input is the emotion recognition data obtained. The device encrypts the data using RSA encryption technology and sends it to the server via the HTTPS protocol. The output is the encrypted emotion data.

[0685] Step 10:

[0686] The user operates the device to collect location data in real time. The input is the user's settings and operations. The user launches a location collection app and enables GPS and WiFi data. The output is the collected location data.

[0687] Step 11:

[0688] The user provides emotional data through the emotion engine. The input is the user's voice and facial expressions. The device analyzes this to recognize the emotion and stores it for transmission in the next step. The output is the analyzed emotional data.

[0689] Step 12:

[0690] The user checks the analysis results on a dashboard or app. The input is the analysis results of the AI ​​model returned from the server. The user visually checks the analysis results of the location data that take into account emotional information and provides feedback as needed. The output is the user's understanding of the results and feedback.

[0691] (Application example 2)

[0692] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0693] Conventional autonomous vehicle systems have been unable to adjust vehicle settings in real time based on passengers' emotional states, and lack a means to effectively combine location information data and emotion recognition data collected from multiple devices to improve safety and comfort.

[0694] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data transmitted from the terminals and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, and means for adjusting vehicle settings in accordance with passenger emotions. This makes it possible to adjust vehicle settings in real time in accordance with the emotional state of passengers and achieve highly safe and comfortable autonomous driving.

[0695] "Multiple terminals" refers to multiple electronic devices, and in particular to portable information processing devices such as smartphones and tablets.

[0696] "Location data" refers to data including location coordinates and movement history obtained from GPS and Wi-Fi towers.

[0697] "Redaction" refers to the process of making personally identifiable information unidentifiable by anonymizing or encrypting data.

[0698] "Server" refers to a computer system that stores, processes, and distributes data over a network.

[0699] "Emotion recognition data" is data that indicates the emotional state of the user, obtained using voice data and facial expression analysis.

[0700] "AI model" refers to an artificial intelligence model created using machine learning algorithms that learns the characteristics of data and makes predictions and classifications.

[0701] "Updating" refers to the process of retraining an existing model based on new data to improve its accuracy and performance.

[0702] "Optimizing" means adjusting various parameters and factors to maximize the performance of a model.

[0703] "Returning" is the operation of sending data or models processed by the server back to the original terminal.

[0704] "Training" refers to the process of using newly collected data on the device to further train the AI ​​model and improve its performance.

[0705] "Analyzing" means applying statistical methods and algorithms to interpret collected data and extract useful information.

[0706] "Adjusting vehicle settings based on passenger emotions" means taking actions such as changing the route or playing relaxing music if a passenger is feeling stressed.

[0707] The present invention combines a system that anonymizes location information data collected from multiple devices, transmits the data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail an embodiment based on the roles of the server, device, and user.

[0708] Server-side processing

[0709] The server receives anonymized location data sent by the company. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. Next, this location data is integrated with emotion recognition data to update the AI ​​model. To update the model, a machine learning algorithm is used to learn from the data and improve the accuracy of predictions and classifications. The updated AI model is then encrypted and returned to each device.

[0710] Specific software used on the server side includes machine learning frameworks such as TensorFlow and PyTorch, and SSL / TLS protocols are used to keep data confidential.

[0711] Terminal side processing

[0712] The device collects location data from smartphones and tablets, including using GPS and Wi-Fi tower information to obtain a user's current location and movement history, and includes measures to anonymize or encrypt the data. The anonymized data is then sent to a server using a secure communications protocol (e.g., HTTPS).

[0713] The device that receives the returned AI model uses it to further train its own location data. The device also has an emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This emotion recognition data is used to adjust the vehicle's settings in real time according to the user's condition.

[0714] The specific hardware used in the device includes a GPS module, camera, and microphone, and the emotion engine uses Azure Cognitive Services and Google Cloud Vision.

[0715] User-side processing

[0716] Users operate their devices to collect location data from their smartphones or tablets. At this time, the data is appropriately anonymized and securely transmitted to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also transmitted to the server and used to update the AI ​​model.

[0717] For example, if a user is riding in a self-driving vehicle, the vehicle's settings can be adjusted in real time based on the user's emotional state: for example, if the passenger is feeling stressed, the vehicle can change to a route with less traffic or play relaxing music.

[0718] Prompt Sentence Examples

[0719] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[0720] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[0721] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0722] Step 1:

[0723] The server receives location data and emotion recognition data from multiple devices. It receives anonymized location data and emotion recognition data as input and stores them in a database. Specifically, it checks the data structure when receiving the data and removes outliers and incomplete data.

[0724] Step 2:

[0725] The server preprocesses the received data. It uses location data and emotion recognition data as input and normalizes them. It standardizes the data format and performs processing to impute missing values. Specifically, it scales the range of numerical data and encodes categorical data.

[0726] Step 3:

[0727] The server updates the AI ​​model based on the preprocessed data. Using the normalized location data and emotion recognition data as input, it applies machine learning algorithms to retrain the model, improving the prediction and classification accuracy of the AI ​​model. Specifically, it tunes the model's hyperparameters and trains it until convergence is achieved.

[0728] Step 4:

[0729] The server encrypts the updated AI model and returns it to each device. It uses the latest AI model and each device's identification information as input and distributes the model using a secure communication protocol (e.g., HTTPS). Specifically, it generates an encryption key and encodes the model data.

[0730] Step 5:

[0731] The device receives the AI ​​model returned from the server and further trains it using its own data. It uses the received AI model and the location and emotion recognition data stored on the device as input. Specifically, it performs incremental training to improve the adaptability of the model.

[0732] Step 6:

[0733] The device analyzes real-time location and emotion recognition data using the latest AI models to adjust vehicle settings. It uses the passenger's current location and emotional state as input to optimize the route and environment. Specifically, it will select a route with less traffic and play relaxing music if it detects stress.

[0734] Step 7:

[0735] Users operate their devices to collect location data in real time. GPS modules and Wi-Fi information are used as input to obtain current locations and movement histories. The collected data is then anonymized or encrypted and stored in the device's memory.

[0736] Step 8:

[0737] The user recognizes their own emotional state using the device's emotion engine. Voice data and facial expressions are acquired as input from a camera or microphone, and an emotion recognition algorithm is applied. Specifically, the emotional state is extracted as a parameter and stored in a database.

[0738] Prompt Sentence Examples

[0739] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[0740] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[0741] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0742] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0743] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0744] [Third embodiment]

[0745] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0746] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0747] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0748] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0749] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0750] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0751] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0752] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0753] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0754] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0755] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0756] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0757] This invention is a system that conceals location information data collected from multiple devices and transmits the data to a server to update an AI model. Here, we will explain in detail the program processing based on the roles of the server, device, and user.

[0758] Server-side processing

[0759] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[0760] The trained AI model is returned to each device. At this time, the returned model is encrypted and sent securely.

[0761] Terminal side processing

[0762] The device collects location data from smartphones. The collected data is anonymized or encrypted to prevent eavesdropping or tampering. When transmitting this anonymized data to a server, a secure communication protocol (e.g., HTTPS) is used to prevent data eavesdropping or tampering.

[0763] Once the device receives the AI ​​model returned from the server, it uses the model to further train its own location data. This training improves the accuracy of the model, allowing the device to use it to perform highly accurate location data analysis. Specifically, this is effective for area marketing and analyzing people flow during disasters, for example.

[0764] User-side processing

[0765] The user operates the device to collect location data from the smartphone. At this time, the data is appropriately anonymized and set up to be securely transmitted to the server. The AI ​​model returned from the server is then applied to the device to analyze the location data.

[0766] Users can check the analysis results and request that the AI ​​model be updated again if necessary. For example, a marketing department user may frequently retrain the model to simulate campaign effectiveness in a specific region in advance.

[0767] Specific examples

[0768] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. The server then returns the trained AI model to companies A and B.

[0769] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[0770] In this way, the present invention enables multiple companies to safely utilize their own data while conducting advanced location data analysis. It is also a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[0771] The processing flow will be explained below.

[0772] Step 1:

[0773] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[0774] Step 2:

[0775] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[0776] Step 3:

[0777] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[0778] Step 4:

[0779] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[0780] Step 5:

[0781] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[0782] Step 6:

[0783] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[0784] Step 7:

[0785] The server returns the trained AI model to each device. The model file is encrypted before transmission to prevent data tampering or eavesdropping.

[0786] Step 8:

[0787] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[0788] Step 9:

[0789] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[0790] Step 10:

[0791] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[0792] Step 11:

[0793] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[0794] Step 12:

[0795] Users can send data to the server again as needed, and the federated learning process can be repeated, incorporating additional data and the latest trends to further improve the accuracy of the analysis results.

[0796] Example 1

[0797] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0798] Conventional location data analysis systems face the challenge of performing highly accurate data analysis while ensuring the confidentiality and security of location data collected from multiple devices. In particular, secure data transmission and reception, preprocessing, integration, secure transmission of trained models, and highly accurate data analysis on each device are critical issues in practical operation. A system that provides effective solutions to these challenges is needed.

[0799] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0800] In this invention, the server includes means for collecting location information data from multiple terminals, means for concealing the collected location information data, means for transmitting the concealed location information data to the server, means for preprocessing the transmitted multiple location information data, means for updating an AI model based on the preprocessed location information data, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, means for encrypting and securely transmitting the trained model, and means for transmitting the location information data using a secure communication protocol. This enables secure concealment and transmission / reception of location information data collected from multiple terminals, preprocessing, updating of the AI ​​model, secure transmission of the trained model, and highly accurate data analysis.

[0801] A "terminal" is a device operated by a user that collects location data, conceals it, and transmits it to a server.

[0802] The "server" is a computer system that receives anonymized location data sent from multiple devices, preprocesses it, and updates the AI ​​model.

[0803] "Location Data" means data collected from a device that indicates its geographic location and is obtained through GPS or location services APIs.

[0804] "Redaction" is the process of anonymizing or encrypting location data to protect its privacy.

[0805] "Preprocessing" refers to the process of checking the consistency of the received location data, removing incomplete data and outliers, and normalizing the data.

[0806] An "AI model" is a model for making predictions and classifications that is built using machine learning algorithms based on location data.

[0807] "Updating" is the process of recalibrating the weights and parameters of an existing AI model using new location data.

[0808] A "trained model" is an AI model that has been trained on location data and is ready to be used for prediction or classification tasks.

[0809] "Encryption" is the process of converting data using a specific algorithm to make it unintelligible to third parties in order to protect the privacy of trained models and transmitted data.

[0810] A "secure communication protocol" is a set of communication rules for securely sending and receiving data, and uses technologies such as HTTPS and TLS.

[0811] The present invention provides a system for anonymizing location information data collected from multiple devices and transmitting the data to a server to update an AI model. Specific embodiments of the system are described below.

[0812] Server-side processing

[0813] The server receives location data sent from multiple devices. The received data is secure because it is encrypted using the Secured HTTPS protocol. The server first preprocesses the received data. For preprocessing, it uses Python's Pandas library to detect and remove incomplete data and outliers. It also normalizes the data using NumPy. The server updates the AI ​​model based on the preprocessed data. Machine learning frameworks such as TensorFlow and PyTorch are used to update the model. Once the training is complete, the model is encrypted and then securely sent to each device. The AES algorithm is used for encryption, and the HTTPS protocol is used for transmission.

[0814] Terminal side processing

[0815] The terminal collects location data from devices such as smartphones operated by the user. This is done using GPS functions or location service APIs. The collected data is first anonymized within the device. This anonymization is done by using Python's GeoPy library or the AES encryption algorithm. The anonymized data is then sent to the server using the HTTPS protocol. Once the trained AI model is returned from the server, the terminal uses the model to further train the location data locally. This training is done using machine learning libraries such as Scikit-learn and Onnx. Using the trained model enables highly accurate location data analysis, such as area marketing and people flow analysis during disasters.

[0816] User-side processing

[0817] The user operates the device to collect location data from their smartphone. At this time, the collected data is set to be appropriately anonymized and securely sent to the server. The AI ​​model returned from the server is applied to the device, and analysis is performed based on the location data. For example, a user on a business trip can find an appropriate route by analyzing movement patterns in a new area. If, based on the analysis results, it is determined that new data collection or improvement in model accuracy is necessary, the user can request an update of the AI ​​model again. An example prompt is, "I would like to update the AI ​​model to improve the efficiency of location data analysis in area marketing. Please explain in detail the process for retraining the model based on new location data and obtaining highly accurate predictions."

[0818] Specific examples

[0819] Suppose a company collects location data to analyze customer behavior. When a user configures their smartphone to collect location information, the device periodically acquires location information, anonymizes the data using the GeoPy library, and sends it to a server. The server receives the data, preprocesses it using Pandas and NumPy, and updates the model using TensorFlow. The new model is returned to each device via AES encryption, where it is trained again using Scikit-learn. This allows the company to perform highly accurate people flow analysis and optimize its marketing strategy. By repeating this cycle, operations that utilize data while protecting privacy can be achieved.

[0820] In this way, the system of the present invention enables highly accurate analysis of location data through data collection from multiple devices, secure data transmission, and updating and application of AI models.

[0821] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0822] Server-side processing

[0823] Step 1: Receiving location data

[0824] Input: Anonymized location data sent from multiple devices

[0825] Processing: The server receives the data in an encrypted form using the HTTPS protocol.

[0826] Output: Received anonymized location data

[0827] Specific operation: The server's receiving module monitors HTTPS communications, decrypts and stores the incoming data packets.

[0828] Step 2: Preprocessing the data

[0829] Input: Received anonymized location data

[0830] Processing: The server uses Python's Pandas library to detect and remove incomplete data and outliers, and also normalizes the data using NumPy.

[0831] Output: Preprocessed location data

[0832] Specific operation: A script is executed to fill in missing values ​​in the data, remove outliers, and scale numerical data.

[0833] Step 3: Update the AI ​​model

[0834] Input: Preprocessed geolocation data

[0835] Processing: The server uses TensorFlow to train the AI ​​model, learning a new model using all the data and optimizing the model weights.

[0836] Output: Updated AI model

[0837] What it does: The Python script calls the TensorFlow library and runs the process of training a model using the data as input.

[0838] Step 4: Encrypt the trained model

[0839] Input: Updated AI model

[0840] Processing: The server encrypts the model using the AES encryption algorithm.

[0841] Output: Encrypted AI model

[0842] Specific operation: A script is executed to encrypt files containing the weights and biases of the trained model using the AES algorithm.

[0843] Step 5: Send the encrypted model

[0844] Input: Encrypted AI model

[0845] Processing: The server sends the encrypted model to multiple devices using the HTTPS protocol.

[0846] Output: Trained model sent to multiple devices

[0847] Specific operation: HTTPS communication is initiated from the server to the device, and an encrypted model file is sent.

[0848] Terminal side processing

[0849] Step 1: Collect location data

[0850] Input: Smartphone GPS function and location information service API

[0851] Processing: The device periodically acquires location data using these functions.

[0852] Output: Collected location data

[0853] Specific operation: The device application calls the location information service API, periodically obtains location data, and stores it in an internal database.

[0854] Step 2: Data redaction

[0855] Input: Collected location data

[0856] Processing: The device anonymizes the data using the GeoPy library or uses the AES encryption algorithm.

[0857] Output: Anonymized location data

[0858] What it does: The collected data is processed by a Python script, anonymized or encrypted, and then stored as ready data.

[0859] Step 3: Sending data

[0860] Input: anonymized location data

[0861] Processing: The terminal sends data to the server using the HTTPS protocol.

[0862] Output: Location data sent to the server

[0863] Specific operation: The terminal's transmission module initiates HTTPS communication and sends the anonymized data to the server.

[0864] Step 4: Receive the trained model

[0865] Input: The encrypted AI model sent from the server

[0866] Processing: The terminal receives the model from the server using the HTTPS protocol and decrypts it using the AES algorithm.

[0867] Output: Decoded AI model

[0868] Specific operation: The terminal's receiving module monitors HTTPS communications and decrypts the encrypted files that arrive using the AES algorithm.

[0869] Step 5: Retraining the local data

[0870] Input: Decoded AI model and local location data

[0871] Processing: The device retrains the model on the local data using the Scikit-learn or Onnx library.

[0872] Output: Retrained AI model

[0873] What happens: A script on the device loads the decoded model and retrains it with locally collected data.

[0874] User-side processing

[0875] Step 1: Configure location data collection

[0876] Input: User settings information (permission of location services, frequency of collection, etc.)

[0877] Processing: The user operates the application on the device to set up location information collection.

[0878] Output: Start of collection process based on configuration information

[0879] Specific behavior: Collects settings through the user interface and activates the application's location collection function.

[0880] Step 2: Applying AI models to analyze data

[0881] Input: Updated AI model and collected location data

[0882] Processing: The device analyzes the location data using the received AI model. The user can then review the analysis results and use them for marketing strategies, etc.

[0883] Output: Analyzed data and results

[0884] What it does: Loads the updated model, runs analytical algorithms on the collected data, and displays the results.

[0885] Step 3: Request an update to the AI ​​model

[0886] Input: User feedback on analysis results and model accuracy

[0887] Processing: The user requests the server to update the AI ​​model again if necessary.

[0888] Output: Update request sent to the server

[0889] What it does: Collect feedback through the user interface and submit retraining requests as needed.

[0890] (Application example 1)

[0891] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0892] Current online shopping sites lack mechanisms for effectively utilizing users' location information to provide individually customized product recommendations and promotions. Furthermore, technologies for providing highly accurate recommendations while protecting user privacy are limited. This makes it difficult to provide users with the most appropriate product information and promotions in a timely manner. Furthermore, there is a need for the development of technologies that can effectively analyze and utilize location data while ensuring the security of user data.

[0893] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0894] In this invention, the server includes a means for integrating the transmitted multiple pieces of location information data and updating the AI ​​model, a means for returning the updated AI model to the multiple terminals, and a means for generating product recommendations and promotions based on the user's location based on the collected location information data, thereby enabling the provision of highly accurate product recommendations and promotions using location information data.

[0895] "Multiple terminals" are electronic devices used to collect and transmit location information data, such as multiple smartphones and personal computers.

[0896] "Location information data" refers to information about geographical locations collected from terminals, and is made up of numerical data such as latitude and longitude.

[0897] "Redaction" is the process of protecting the privacy of collected location data by anonymizing or encrypting it.

[0898] The "Server" is a central processing system for receiving collected location data, integrating the data, and updating and distributing the AI ​​model.

[0899] An "AI model" is a model that learns location data and is automatically adjusted using machine learning algorithms to make predictions and classifications.

[0900] An "updated AI model" is an AI model that has been trained and adjusted based on the latest location data, resulting in improved accuracy.

[0901] "Product recommendation" is the act of suggesting suitable products or services to a user based on the user's location information.

[0902] A "promotion" is a marketing method that notifies users of discounts and special offers on specific products and services.

[0903] "Notification" refers to the act of informing a user of generated product recommendations or promotions by displaying them on the user's terminal.

[0904] MODE FOR CARRYING OUT THE INVENTION

[0905] This invention is a system that anonymizes location information data collected from devices and sends it to a server to update an AI model. Specifically, the server receives location information data collected from multiple devices, preprocesses it, and updates the AI ​​model using a machine learning algorithm. The updated AI model is then sent back to multiple devices, where it is used for analyzing location information data, recommending products, and generating promotions.

[0906] Program generation and natural language explanation

[0907] 1. Hardware and Software:

[0908] Hardware:

[0909] Smartphones: Location information collection and user notification

[0910] Server: Data reception, preprocessing, AI model update and distribution

[0911] software:

[0912] Language: Python

[0913] Libraries: requests (HTTP requests), cryptography.fernet (data encryption)

[0914] Communication protocol: HTTPS (secure transmission of data)

[0915] Algorithm: Machine learning algorithm (e.g. TensorFlow, Scikit-learn)

[0916] 2. Data processing and calculation:

[0917] Terminal processing:

[0918] Location data is collected from smartphones, encrypted and anonymized, and then securely transmitted to a server using the HTTPS protocol.

[0919] Server side:

[0920] The server preprocesses the received location data, removing incomplete data and outliers and normalizing it. It then updates the AI ​​model using a machine learning algorithm based on the integrated location data. It then encrypts the updated AI model and sends it to each device.

[0921] Train the model:

[0922] Each device uses the updated AI model received from the server to retrain its own location data, enabling the device to analyze location data with high accuracy, recommend products, and generate promotions.

[0923] 3. Example:

[0924] For example, location information of users in Tokyo is collected and periodically sent to a server in an anonymous format. The server then uses this data to update the learning model and provide sales information and recommended products based on the user's location to the smartphone.

[0925] Prompt Sentence Examples

[0926] "We would like to generate an AI model that will enable users to receive personalized promotions and product recommendations based on their current location. Anonymizing and secure transmission of location data is essential. Please provide an example based on location data within Tokyo."

[0927] According to the above-described embodiment, the present invention makes it possible to effectively utilize user location information data while protecting privacy, and to provide highly accurate product recommendations and promotions.

[0928] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0929] Program processing flow

[0930] Step 1:

[0931] The user collects location data using a smartphone. The device uses the GPS function to obtain current latitude and longitude data and sends that data to an application. The input of this process is the user's current location (location data), and the output is the collected location data.

[0932] Step 2:

[0933] The device conceals the collected location data by encrypting it using an encryption algorithm (e.g., Fernet). The input is raw location data, and the output is encrypted data.

[0934] Step 3:

[0935] The device securely transmits the anonymized location data to the server using the HTTPS protocol. The input is the encrypted location data, and the output is a transmission confirmation message.

[0936] Step 4:

[0937] The server preprocesses the received location data, which includes removing incomplete data and outliers, and normalizing the data. The input is the encrypted location data, and the output is the preprocessed data.

[0938] Step 5:

[0939] The server integrates the preprocessed data and updates the AI ​​model using machine learning algorithms (e.g., TensorFlow, Scikit-learn). The input is the preprocessed location data, and the output is the updated AI model.

[0940] Step 6:

[0941] The server encrypts the updated AI model and returns it to each device. The input is the updated AI model, and the output is the encrypted AI model data.

[0942] Step 7:

[0943] The device decrypts the updated AI model received from the server and uses it to retrain its own location data. The input is the encrypted AI model data, and the output is the trained AI model.

[0944] Step 8:

[0945] The device uses the trained AI model to analyze the location data and generate product recommendations and promotions based on the user's location. The input is the trained AI model and location data, and the output is the generated product recommendations and promotion information.

[0946] Step 9:

[0947] The terminal notifies the user of the generated product recommendations and promotions. The input is the generated product recommendation and promotion information, and the output is a notification displayed on the user's smartphone.

[0948] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0949] This invention combines a system that conceals location information data collected from multiple devices, sends that data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail the program processing based on the roles of the server, device, and user.

[0950] Server-side processing

[0951] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[0952] The server also receives emotion recognition data sent from the device and uses it to optimize the AI ​​model. This emotion recognition data includes the user's emotional state as a parameter and is used to improve the accuracy of the model. The server then encrypts the trained AI model and returns it to each device.

[0953] Terminal side processing

[0954] The device collects location data from the smartphone. It uses GPS data and Wi-Fi tower information to obtain the user's current location and movement history. The collected data is anonymized or encrypted to prevent eavesdropping or tampering with the data when it is sent to the server.

[0955] Once the AI ​​model is returned from the server, the device uses it to further learn its own location data. In addition to this learning, the device also has a built-in emotion engine that recognizes the user's emotions, analyzing the user's voice data and facial expressions to recognize emotions. This emotion recognition data is then used to further refine the AI ​​model.

[0956] User-side processing

[0957] Users operate the device to collect location data from their smartphone. At this time, the data is appropriately anonymized and set up to be securely sent to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also sent to the server and used to update the AI ​​model.

[0958] The AI ​​model returned from the server is applied to the device to analyze the location data. The user can review the analysis results that incorporate emotional information and request that the AI ​​model be updated again if necessary. This system is extremely useful for marketing department users who want to simulate the effectiveness of campaigns in specific areas in advance, or for taking a personalized approach based on the user's emotional state.

[0959] Specific examples

[0960] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is then returned to companies A and B.

[0961] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[0962] In this way, the present invention enables multiple companies to safely utilize their own data while performing advanced location data analysis that incorporates emotional information. It also provides a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[0963] The processing flow will be explained below.

[0964] Step 1:

[0965] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[0966] Step 2:

[0967] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[0968] Step 3:

[0969] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[0970] Step 4:

[0971] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[0972] Step 5:

[0973] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[0974] Step 6:

[0975] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[0976] Step 7:

[0977] The device uses an emotion engine that recognizes the user's emotions and analyzes voice data and facial expressions to recognize the user's emotional state.

[0978] Step 8:

[0979] The device sends the user's emotion data recognized by the emotion engine to the server using a secure communication protocol.

[0980] Step 9:

[0981] The server uses the received emotion data to optimize the AI ​​model, using the emotion data as parameters to further improve the accuracy of the model.

[0982] Step 10:

[0983] The server encrypts the trained AI model and returns it to each device. The return process is performed using a secure communication protocol (e.g., HTTPS).

[0984] Step 11:

[0985] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[0986] Step 12:

[0987] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[0988] Step 13:

[0989] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[0990] Step 14:

[0991] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[0992] Step 15:

[0993] Users can send data to the server again as needed, and the federated learning process can be repeated to further improve the accuracy of the analysis results by incorporating additional data and the latest trends.

[0994] Example 2

[0995] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0996] Conventional location data collection systems lacked a method for effectively analyzing data while maintaining data anonymity and security. Furthermore, there was no way to integrate and analyze users' emotional states with location data, making it difficult to achieve more advanced personalization and evaluate marketing effectiveness. Furthermore, retraining models on each device and applying them efficiently was a challenge.

[0997] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, and means for analyzing the location information data based on the AI ​​model trained at the terminals. This enables advanced data analysis that takes user emotion information into account while maintaining the anonymity and security of the data. Furthermore, it also enables each terminal to retrain the model and efficiently apply it.

[0998] "Multiple terminals" refers to multiple user devices connected to the Internet, including smartphones, tablets, etc.

[0999] "Location data" refers to data about a user's current location and movement history obtained using GPS, Wi-Fi, etc.

[1000] "Redaction" is the process of anonymizing or encrypting data to protect it so that third parties cannot easily identify its contents.

[1001] A "server" is a computer system that receives, processes, and transmits data over a network, and refers to infrastructure capable of processing large amounts of data.

[1002] An "AI model" is a predictive model built using machine learning algorithms, and is a technology for data analysis and classification.

[1003] "Emotion recognition data" refers to data about a user's emotional state obtained by analyzing their vocal tone and facial expressions.

[1004] A "communication protocol" refers to the rules and methods for securely sending and receiving data, and includes HTTPS and TLS.

[1005] "Machine learning algorithm" is a general term for mathematical models and methods used to make predictions and classifications based on collected data.

[1006] "AES encryption" refers to a method of encrypting data using the Advanced Encryption Standard, providing a high level of security.

[1007] "Anonymization" is a general term for techniques that protect data privacy by removing personally identifiable information.

[1008] A "secure communication protocol" is a protocol that prevents eavesdropping or tampering with data when it is being sent or received.

[1009] This invention is a system that collects location information data from multiple devices, anonymizes the data, transmits it to a server, and updates an AI model. Furthermore, by combining it with user emotion recognition data, the accuracy of the model can be improved. The specific system configuration and operation procedure are described below.

[1010] Server-side processing

[1011] The server receives anonymized location data sent by multiple companies and individuals. The communication protocol used is HTTPS, which prevents data eavesdropping and tampering. The received data is first preprocessed to remove missing values ​​and outliers and normalize the data using the Pandas library. The specific data cleansing method used is the IQR (Interquartile Range) method.

[1012] Next, the AI ​​model is updated using the integrated location data. TensorFlow or PyTorch are typically used as machine learning frameworks. This allows for efficient learning of large amounts of data and improves the accuracy of the predictive model. For example, cluster analysis can be performed to identify user behavior patterns in specific areas.

[1013] The server also receives emotion recognition data sent from the device and integrates this data into the AI ​​model. Emotion data is obtained by analyzing voice tone and facial expressions and is added to the model as features. This process uses tools such as the Google Cloud Speech-to-Text API and OpenCV. Finally, the trained AI model is AES encrypted and securely returned to each device. The encryption is performed using the Python cryptography library.

[1014] Terminal side processing

[1015] The device uses a smartphone to collect location data. Specifically, it periodically obtains location information using Android or iOS APIs. The collected data is anonymized using RSA encryption technology. The anonymized data is sent to the server via the HTTPS protocol. At this time, a TLS certificate is used to ensure the security of the communication.

[1016] The device receives the AI ​​model returned from the server and further trains it on its own location data. Additional training is performed in a local environment using TensorFlow and other tools. The device also has a built-in emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This is done using voice recognition APIs and face recognition APIs. The analyzed emotion data is sent to the server and used to update the AI ​​model.

[1017] User-side processing

[1018] The user operates the device to collect location data from the smartphone. They confirm that the data collection is being carried out appropriately and that it is confidential. They then launch the location data collection app, turn on the GPS function, and begin data collection. The user then provides emotional data, such as voice and facial expressions, through the emotion engine. The device automatically recognizes the emotion and sends the data to the server.

[1019] The device receives the AI ​​model returned from the server and checks the results of the location data analysis. The analysis results, which take into account emotional information, can be viewed on a dashboard or app, feedback can be provided as needed, and an update of the AI ​​model can be requested.

[1020] Specific examples

[1021] For example, companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is returned to companies A and B. Company A's device uses the returned AI model to further train its own location data. This learning allows company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services.

[1022] Example prompts to input to the generative AI model

[1023] What are the specific steps to integrate location data from Company A and Company B to update the AI ​​model?

[1024] Please elaborate on how you integrate user emotion recognition data into your AI models.

[1025] Please provide specific steps for data redaction and secure transmission.

[1026] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1027] Step 1:

[1028] The server receives anonymized location data sent by companies or individuals. The input data is encrypted data sent via the HTTPS protocol. The server receives this data and saves it in storage. Specifically, the server takes the data into a receiving buffer and stores it in preparation for the next step of processing.

[1029] Step 2:

[1030] The server preprocesses the received location data. The input is the stored location data. The server uses the Pandas library to remove missing values ​​and outliers and normalize the data. The IQR (Interquartile Range) method is used to detect and eliminate outliers. The output is the preprocessed location data.

[1031] Step 3:

[1032] The server updates the AI ​​model using the preprocessed location data. The input is the combined preprocessed location data. The server applies machine learning algorithms using TensorFlow or PyTorch to train the model based on the new data. The output is an updated AI model.

[1033] Step 4:

[1034] The server receives emotion recognition data sent from the device and integrates it into the AI ​​model. The input is the emotion recognition data and the updated AI model, including voice tone and facial expression analysis data. The output is an optimized model that integrates the emotion data. This optimization is performed using the Google Cloud Speech-to-Text API and OpenCV.

[1035] Step 5:

[1036] The server encrypts the trained AI model using AES and returns it to each device. The input is an optimized model integrated with emotion data. The encryption is performed using Python's cryptography library. The output is an encrypted AI model file. The server sends this file to each device via the HTTPS protocol.

[1037] Step 6:

[1038] The device receives the encrypted AI model from the server. The input is the encrypted AI model file. The device receives this file and stores it in local storage. Specifically, the device first confirms receipt of the file and stores it in the appropriate directory.

[1039] Step 7:

[1040] The device deserializes and loads the received AI model to further train it on its own location data. The input is the encrypted AI model file and local location data. The device uses code such as TensorFlow to unpack the model and retrain it with the new data. The output is a locally updated AI model.

[1041] Step 8:

[1042] The device uses an emotion recognition engine to collect user emotional data. Inputs include the user's voice data and facial expression video. The device analyzes this to recognize emotions and saves them as data. Specifically, it captures data in real time using a camera or microphone and applies an analysis algorithm.

[1043] Step 9:

[1044] The device encrypts the emotion data it recognizes and sends it to the server. The input is the emotion recognition data obtained. The device encrypts the data using RSA encryption technology and sends it to the server via the HTTPS protocol. The output is the encrypted emotion data.

[1045] Step 10:

[1046] The user operates the device to collect location data in real time. The input is the user's settings and operations. The user launches a location collection app and enables GPS and WiFi data. The output is the collected location data.

[1047] Step 11:

[1048] The user provides emotional data through the emotion engine. The input is the user's voice and facial expressions. The device analyzes this to recognize the emotion and stores it for transmission in the next step. The output is the analyzed emotional data.

[1049] Step 12:

[1050] The user checks the analysis results on a dashboard or app. The input is the analysis results of the AI ​​model returned from the server. The user visually checks the analysis results of the location data that take into account emotional information and provides feedback as needed. The output is the user's understanding of the results and feedback.

[1051] (Application example 2)

[1052] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1053] Conventional autonomous vehicle systems have been unable to adjust vehicle settings in real time based on passengers' emotional states, and lack a means to effectively combine location information data and emotion recognition data collected from multiple devices to improve safety and comfort.

[1054] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data transmitted from the terminals and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, and means for adjusting vehicle settings in accordance with passenger emotions. This makes it possible to adjust vehicle settings in real time in accordance with the emotional state of passengers and achieve highly safe and comfortable autonomous driving.

[1055] "Multiple terminals" refers to multiple electronic devices, and in particular to portable information processing devices such as smartphones and tablets.

[1056] "Location data" refers to data including location coordinates and movement history obtained from GPS and Wi-Fi towers.

[1057] "Redaction" refers to the process of making personally identifiable information unidentifiable by anonymizing or encrypting data.

[1058] "Server" refers to a computer system that stores, processes, and distributes data over a network.

[1059] "Emotion recognition data" is data that indicates the emotional state of the user, obtained using voice data and facial expression analysis.

[1060] "AI model" refers to an artificial intelligence model created using machine learning algorithms that learns the characteristics of data and makes predictions and classifications.

[1061] "Updating" refers to the process of retraining an existing model based on new data to improve its accuracy and performance.

[1062] "Optimizing" means adjusting various parameters and factors to maximize the performance of a model.

[1063] "Returning" is the operation of sending data or models processed by the server back to the original terminal.

[1064] "Training" refers to the process of using newly collected data on the device to further train the AI ​​model and improve its performance.

[1065] "Analyzing" means applying statistical methods and algorithms to interpret collected data and extract useful information.

[1066] "Adjusting vehicle settings based on passenger emotions" means taking actions such as changing the route or playing relaxing music if a passenger is feeling stressed.

[1067] The present invention combines a system that anonymizes location information data collected from multiple devices, transmits the data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail an embodiment based on the roles of the server, device, and user.

[1068] Server-side processing

[1069] The server receives anonymized location data sent by the company. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. Next, this location data is integrated with emotion recognition data to update the AI ​​model. To update the model, a machine learning algorithm is used to learn from the data and improve the accuracy of predictions and classifications. The updated AI model is then encrypted and returned to each device.

[1070] Specific software used on the server side includes machine learning frameworks such as TensorFlow and PyTorch, and SSL / TLS protocols are used to keep data confidential.

[1071] Terminal side processing

[1072] The device collects location data from smartphones and tablets, including using GPS and Wi-Fi tower information to obtain a user's current location and movement history, and includes measures to anonymize or encrypt the data. The anonymized data is then sent to a server using a secure communications protocol (e.g., HTTPS).

[1073] The device that receives the returned AI model uses it to further train its own location data. The device also has an emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This emotion recognition data is used to adjust the vehicle's settings in real time according to the user's condition.

[1074] The specific hardware used in the device includes a GPS module, camera, and microphone, and the emotion engine uses Azure Cognitive Services and Google Cloud Vision.

[1075] User-side processing

[1076] Users operate their devices to collect location data from their smartphones or tablets. At this time, the data is appropriately anonymized and securely transmitted to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also transmitted to the server and used to update the AI ​​model.

[1077] For example, if a user is riding in a self-driving vehicle, the vehicle's settings can be adjusted in real time based on the user's emotional state: for example, if the passenger is feeling stressed, the vehicle can change to a route with less traffic or play relaxing music.

[1078] Prompt Sentence Examples

[1079] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[1080] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[1081] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1082] Step 1:

[1083] The server receives location data and emotion recognition data from multiple devices. It receives anonymized location data and emotion recognition data as input and stores them in a database. Specifically, it checks the data structure when receiving the data and removes outliers and incomplete data.

[1084] Step 2:

[1085] The server preprocesses the received data. It uses location data and emotion recognition data as input and normalizes them. It standardizes the data format and performs processing to impute missing values. Specifically, it scales the range of numerical data and encodes categorical data.

[1086] Step 3:

[1087] The server updates the AI ​​model based on the preprocessed data. Using the normalized location data and emotion recognition data as input, it applies machine learning algorithms to retrain the model, improving the prediction and classification accuracy of the AI ​​model. Specifically, it tunes the model's hyperparameters and trains it until convergence is achieved.

[1088] Step 4:

[1089] The server encrypts the updated AI model and returns it to each device. It uses the latest AI model and each device's identification information as input and distributes the model using a secure communication protocol (e.g., HTTPS). Specifically, it generates an encryption key and encodes the model data.

[1090] Step 5:

[1091] The device receives the AI ​​model returned from the server and further trains it using its own data. It uses the received AI model and the location and emotion recognition data stored on the device as input. Specifically, it performs incremental training to improve the adaptability of the model.

[1092] Step 6:

[1093] The device analyzes real-time location and emotion recognition data using the latest AI models to adjust vehicle settings. It uses the passenger's current location and emotional state as input to optimize the route and environment. Specifically, it will select a route with less traffic and play relaxing music if it detects stress.

[1094] Step 7:

[1095] Users operate their devices to collect location data in real time. GPS modules and Wi-Fi information are used as input to obtain current locations and movement histories. The collected data is then anonymized or encrypted and stored in the device's memory.

[1096] Step 8:

[1097] The user recognizes their own emotional state using the device's emotion engine. Voice data and facial expressions are acquired as input from a camera or microphone, and an emotion recognition algorithm is applied. Specifically, the emotional state is extracted as a parameter and stored in a database.

[1098] Prompt Sentence Examples

[1099] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[1100] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[1101] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1102] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1103] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1104] [Fourth embodiment]

[1105] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1106] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1107] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1108] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1109] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1110] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1111] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1112] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1113] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1114] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1115] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1116] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1117] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1118] This invention is a system that conceals location information data collected from multiple devices and transmits the data to a server to update an AI model. Here, we will explain in detail the program processing based on the roles of the server, device, and user.

[1119] Server-side processing

[1120] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[1121] The trained AI model is returned to each device. At this time, the returned model is encrypted and sent securely.

[1122] Terminal side processing

[1123] The device collects location data from smartphones. The collected data is anonymized or encrypted to prevent eavesdropping or tampering. When transmitting this anonymized data to a server, a secure communication protocol (e.g., HTTPS) is used to prevent data eavesdropping or tampering.

[1124] Once the device receives the AI ​​model returned from the server, it uses the model to further train its own location data. This training improves the accuracy of the model, allowing the device to use it to perform highly accurate location data analysis. Specifically, this is effective for area marketing and analyzing people flow during disasters, for example.

[1125] User-side processing

[1126] The user operates the device to collect location data from the smartphone. At this time, the data is appropriately anonymized and set up to be securely transmitted to the server. The AI ​​model returned from the server is then applied to the device to analyze the location data.

[1127] Users can check the analysis results and request that the AI ​​model be updated again if necessary. For example, a marketing department user may frequently retrain the model to simulate campaign effectiveness in a specific region in advance.

[1128] Specific examples

[1129] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. The server then returns the trained AI model to companies A and B.

[1130] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[1131] In this way, the present invention enables multiple companies to safely utilize their own data while conducting advanced location data analysis. It is also a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[1132] The processing flow will be explained below.

[1133] Step 1:

[1134] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[1135] Step 2:

[1136] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[1137] Step 3:

[1138] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[1139] Step 4:

[1140] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[1141] Step 5:

[1142] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[1143] Step 6:

[1144] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[1145] Step 7:

[1146] The server returns the trained AI model to each device. The model file is encrypted before transmission to prevent data tampering or eavesdropping.

[1147] Step 8:

[1148] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[1149] Step 9:

[1150] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[1151] Step 10:

[1152] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[1153] Step 11:

[1154] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[1155] Step 12:

[1156] Users can send data to the server again as needed, and the federated learning process can be repeated, incorporating additional data and the latest trends to further improve the accuracy of the analysis results.

[1157] Example 1

[1158] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1159] Conventional location data analysis systems face the challenge of performing highly accurate data analysis while ensuring the confidentiality and security of location data collected from multiple devices. In particular, secure data transmission and reception, preprocessing, integration, secure transmission of trained models, and highly accurate data analysis on each device are critical issues in practical operation. A system that provides effective solutions to these challenges is needed.

[1160] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1161] In this invention, the server includes means for collecting location information data from multiple terminals, means for concealing the collected location information data, means for transmitting the concealed location information data to the server, means for preprocessing the transmitted multiple location information data, means for updating an AI model based on the preprocessed location information data, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, means for encrypting and securely transmitting the trained model, and means for transmitting the location information data using a secure communication protocol. This enables secure concealment and transmission / reception of location information data collected from multiple terminals, preprocessing, updating of the AI ​​model, secure transmission of the trained model, and highly accurate data analysis.

[1162] A "terminal" is a device operated by a user that collects location data, conceals it, and transmits it to a server.

[1163] The "server" is a computer system that receives anonymized location data sent from multiple devices, preprocesses it, and updates the AI ​​model.

[1164] "Location Data" means data collected from a device that indicates its geographic location and is obtained through GPS or location services APIs.

[1165] "Redaction" is the process of anonymizing or encrypting location data to protect its privacy.

[1166] "Preprocessing" refers to the process of checking the consistency of the received location data, removing incomplete data and outliers, and normalizing the data.

[1167] An "AI model" is a model for making predictions and classifications that is built using machine learning algorithms based on location data.

[1168] "Updating" is the process of recalibrating the weights and parameters of an existing AI model using new location data.

[1169] A "trained model" is an AI model that has been trained on location data and is ready to be used for prediction or classification tasks.

[1170] "Encryption" is the process of converting data using a specific algorithm to make it unintelligible to third parties in order to protect the privacy of trained models and transmitted data.

[1171] A "secure communication protocol" is a set of communication rules for securely sending and receiving data, and uses technologies such as HTTPS and TLS.

[1172] The present invention provides a system for anonymizing location information data collected from multiple devices and transmitting the data to a server to update an AI model. Specific embodiments of the system are described below.

[1173] Server-side processing

[1174] The server receives location data sent from multiple devices. The received data is secure because it is encrypted using the Secured HTTPS protocol. The server first preprocesses the received data. For preprocessing, it uses Python's Pandas library to detect and remove incomplete data and outliers. It also normalizes the data using NumPy. The server updates the AI ​​model based on the preprocessed data. Machine learning frameworks such as TensorFlow and PyTorch are used to update the model. Once the training is complete, the model is encrypted and then securely sent to each device. The AES algorithm is used for encryption, and the HTTPS protocol is used for transmission.

[1175] Terminal side processing

[1176] The terminal collects location data from devices such as smartphones operated by the user. This is done using GPS functions or location service APIs. The collected data is first anonymized within the device. This anonymization is done by using Python's GeoPy library or the AES encryption algorithm. The anonymized data is then sent to the server using the HTTPS protocol. Once the trained AI model is returned from the server, the terminal uses the model to further train the location data locally. This training is done using machine learning libraries such as Scikit-learn and Onnx. Using the trained model enables highly accurate location data analysis, such as area marketing and people flow analysis during disasters.

[1177] User-side processing

[1178] The user operates the device to collect location data from their smartphone. At this time, the collected data is set to be appropriately anonymized and securely sent to the server. The AI ​​model returned from the server is applied to the device, and analysis is performed based on the location data. For example, a user on a business trip can find an appropriate route by analyzing movement patterns in a new area. If, based on the analysis results, it is determined that new data collection or improvement in model accuracy is necessary, the user can request an update of the AI ​​model again. An example prompt is, "I would like to update the AI ​​model to improve the efficiency of location data analysis in area marketing. Please explain in detail the process for retraining the model based on new location data and obtaining highly accurate predictions."

[1179] Specific examples

[1180] Suppose a company collects location data to analyze customer behavior. When a user configures their smartphone to collect location information, the device periodically acquires location information, anonymizes the data using the GeoPy library, and sends it to a server. The server receives the data, preprocesses it using Pandas and NumPy, and updates the model using TensorFlow. The new model is returned to each device via AES encryption, where it is trained again using Scikit-learn. This allows the company to perform highly accurate people flow analysis and optimize its marketing strategy. By repeating this cycle, operations that utilize data while protecting privacy can be achieved.

[1181] In this way, the system of the present invention enables highly accurate analysis of location data through data collection from multiple devices, secure data transmission, and updating and application of AI models.

[1182] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1183] Server-side processing

[1184] Step 1: Receiving location data

[1185] Input: Anonymized location data sent from multiple devices

[1186] Processing: The server receives the data in an encrypted form using the HTTPS protocol.

[1187] Output: Received anonymized location data

[1188] Specific operation: The server's receiving module monitors HTTPS communications, decrypts and stores the incoming data packets.

[1189] Step 2: Preprocessing the data

[1190] Input: Received anonymized location data

[1191] Processing: The server uses Python's Pandas library to detect and remove incomplete data and outliers, and also normalizes the data using NumPy.

[1192] Output: Preprocessed location data

[1193] Specific operation: A script is executed to fill in missing values ​​in the data, remove outliers, and scale numerical data.

[1194] Step 3: Update the AI ​​model

[1195] Input: Preprocessed geolocation data

[1196] Processing: The server uses TensorFlow to train the AI ​​model, learning a new model using all the data and optimizing the model weights.

[1197] Output: Updated AI model

[1198] What it does: The Python script calls the TensorFlow library and runs the process of training a model using the data as input.

[1199] Step 4: Encrypt the trained model

[1200] Input: Updated AI model

[1201] Processing: The server encrypts the model using the AES encryption algorithm.

[1202] Output: Encrypted AI model

[1203] Specific operation: A script is executed to encrypt files containing the weights and biases of the trained model using the AES algorithm.

[1204] Step 5: Send the encrypted model

[1205] Input: Encrypted AI model

[1206] Processing: The server sends the encrypted model to multiple devices using the HTTPS protocol.

[1207] Output: Trained model sent to multiple devices

[1208] Specific operation: HTTPS communication is initiated from the server to the device, and an encrypted model file is sent.

[1209] Terminal side processing

[1210] Step 1: Collect location data

[1211] Input: Smartphone GPS function and location information service API

[1212] Processing: The device periodically acquires location data using these functions.

[1213] Output: Collected location data

[1214] Specific operation: The device application calls the location information service API, periodically obtains location data, and stores it in an internal database.

[1215] Step 2: Data redaction

[1216] Input: Collected location data

[1217] Processing: The device anonymizes the data using the GeoPy library or uses the AES encryption algorithm.

[1218] Output: Anonymized location data

[1219] What it does: The collected data is processed by a Python script, anonymized or encrypted, and then stored as ready data.

[1220] Step 3: Sending data

[1221] Input: anonymized location data

[1222] Processing: The terminal sends data to the server using the HTTPS protocol.

[1223] Output: Location data sent to the server

[1224] Specific operation: The terminal's transmission module initiates HTTPS communication and sends the anonymized data to the server.

[1225] Step 4: Receive the trained model

[1226] Input: The encrypted AI model sent from the server

[1227] Processing: The terminal receives the model from the server using the HTTPS protocol and decrypts it using the AES algorithm.

[1228] Output: Decoded AI model

[1229] Specific operation: The terminal's receiving module monitors HTTPS communications and decrypts the encrypted files that arrive using the AES algorithm.

[1230] Step 5: Retraining the local data

[1231] Input: Decoded AI model and local location data

[1232] Processing: The device retrains the model on the local data using the Scikit-learn or Onnx library.

[1233] Output: Retrained AI model

[1234] What happens: A script on the device loads the decoded model and retrains it with locally collected data.

[1235] User-side processing

[1236] Step 1: Configure location data collection

[1237] Input: User settings information (permission of location services, frequency of collection, etc.)

[1238] Processing: The user operates the application on the device to set up location information collection.

[1239] Output: Start of collection process based on configuration information

[1240] Specific behavior: Collects settings through the user interface and activates the application's location collection function.

[1241] Step 2: Applying AI models to analyze data

[1242] Input: Updated AI model and collected location data

[1243] Processing: The device analyzes the location data using the received AI model. The user can then review the analysis results and use them for marketing strategies, etc.

[1244] Output: Analyzed data and results

[1245] What it does: Loads the updated model, runs analytical algorithms on the collected data, and displays the results.

[1246] Step 3: Request an update to the AI ​​model

[1247] Input: User feedback on analysis results and model accuracy

[1248] Processing: The user requests the server to update the AI ​​model again if necessary.

[1249] Output: Update request sent to the server

[1250] What it does: Collect feedback through the user interface and submit retraining requests as needed.

[1251] (Application example 1)

[1252] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1253] Current online shopping sites lack mechanisms for effectively utilizing users' location information to provide individually customized product recommendations and promotions. Furthermore, technologies for providing highly accurate recommendations while protecting user privacy are limited. This makes it difficult to provide users with the most appropriate product information and promotions in a timely manner. Furthermore, there is a need for the development of technologies that can effectively analyze and utilize location data while ensuring the security of user data.

[1254] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1255] In this invention, the server includes a means for integrating the transmitted multiple pieces of location information data and updating the AI ​​model, a means for returning the updated AI model to the multiple terminals, and a means for generating product recommendations and promotions based on the user's location based on the collected location information data, thereby enabling the provision of highly accurate product recommendations and promotions using location information data.

[1256] "Multiple terminals" are electronic devices used to collect and transmit location information data, such as multiple smartphones and personal computers.

[1257] "Location information data" refers to information about geographical locations collected from terminals, and is made up of numerical data such as latitude and longitude.

[1258] "Redaction" is the process of protecting the privacy of collected location data by anonymizing or encrypting it.

[1259] The "Server" is a central processing system for receiving collected location data, integrating the data, and updating and distributing the AI ​​model.

[1260] An "AI model" is a model that learns location data and is automatically adjusted using machine learning algorithms to make predictions and classifications.

[1261] An "updated AI model" is an AI model that has been trained and adjusted based on the latest location data, resulting in improved accuracy.

[1262] "Product recommendation" is the act of suggesting suitable products or services to a user based on the user's location information.

[1263] A "promotion" is a marketing method that notifies users of discounts and special offers on specific products and services.

[1264] "Notification" refers to the act of informing a user of generated product recommendations or promotions by displaying them on the user's terminal.

[1265] MODE FOR CARRYING OUT THE INVENTION

[1266] This invention is a system that anonymizes location information data collected from devices and sends it to a server to update an AI model. Specifically, the server receives location information data collected from multiple devices, preprocesses it, and updates the AI ​​model using a machine learning algorithm. The updated AI model is then sent back to multiple devices, where it is used for analyzing location information data, recommending products, and generating promotions.

[1267] Program generation and natural language explanation

[1268] 1. Hardware and Software:

[1269] Hardware:

[1270] Smartphones: Location information collection and user notification

[1271] Server: Data reception, preprocessing, AI model update and distribution

[1272] software:

[1273] Language: Python

[1274] Libraries: requests (HTTP requests), cryptography.fernet (data encryption)

[1275] Communication protocol: HTTPS (secure transmission of data)

[1276] Algorithm: Machine learning algorithm (e.g. TensorFlow, Scikit-learn)

[1277] 2. Data processing and calculation:

[1278] Terminal processing:

[1279] Location data is collected from smartphones, encrypted and anonymized, and then securely transmitted to a server using the HTTPS protocol.

[1280] Server side:

[1281] The server preprocesses the received location data, removing incomplete data and outliers and normalizing it. It then updates the AI ​​model using a machine learning algorithm based on the integrated location data. It then encrypts the updated AI model and sends it to each device.

[1282] Train the model:

[1283] Each device uses the updated AI model received from the server to retrain its own location data, enabling the device to analyze location data with high accuracy, recommend products, and generate promotions.

[1284] 3. Example:

[1285] For example, location information of users in Tokyo is collected and periodically sent to a server in an anonymous format. The server then uses this data to update the learning model and provide sales information and recommended products based on the user's location to the smartphone.

[1286] Prompt Sentence Examples

[1287] "We would like to generate an AI model that will enable users to receive personalized promotions and product recommendations based on their current location. Anonymizing and secure transmission of location data is essential. Please provide an example based on location data within Tokyo."

[1288] According to the above-described embodiment, the present invention makes it possible to effectively utilize user location information data while protecting privacy, and to provide highly accurate product recommendations and promotions.

[1289] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1290] Program processing flow

[1291] Step 1:

[1292] The user collects location data using a smartphone. The device uses the GPS function to obtain current latitude and longitude data and sends that data to an application. The input of this process is the user's current location (location data), and the output is the collected location data.

[1293] Step 2:

[1294] The device conceals the collected location data by encrypting it using an encryption algorithm (e.g., Fernet). The input is raw location data, and the output is encrypted data.

[1295] Step 3:

[1296] The device securely transmits the anonymized location data to the server using the HTTPS protocol. The input is the encrypted location data, and the output is a transmission confirmation message.

[1297] Step 4:

[1298] The server preprocesses the received location data, which includes removing incomplete data and outliers, and normalizing the data. The input is the encrypted location data, and the output is the preprocessed data.

[1299] Step 5:

[1300] The server integrates the preprocessed data and updates the AI ​​model using machine learning algorithms (e.g., TensorFlow, Scikit-learn). The input is the preprocessed location data, and the output is the updated AI model.

[1301] Step 6:

[1302] The server encrypts the updated AI model and returns it to each device. The input is the updated AI model, and the output is the encrypted AI model data.

[1303] Step 7:

[1304] The device decrypts the updated AI model received from the server and uses it to retrain its own location data. The input is the encrypted AI model data, and the output is the trained AI model.

[1305] Step 8:

[1306] The device uses the trained AI model to analyze the location data and generate product recommendations and promotions based on the user's location. The input is the trained AI model and location data, and the output is the generated product recommendations and promotion information.

[1307] Step 9:

[1308] The terminal notifies the user of the generated product recommendations and promotions. The input is the generated product recommendation and promotion information, and the output is a notification displayed on the user's smartphone.

[1309] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1310] This invention combines a system that conceals location information data collected from multiple devices, sends that data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail the program processing based on the roles of the server, device, and user.

[1311] Server-side processing

[1312] The server receives anonymized location data sent by companies. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. The AI ​​model is then updated based on the integrated location data. This model update uses machine learning algorithms to learn from the data and improve the accuracy of predictions and classifications.

[1313] The server also receives emotion recognition data sent from the device and uses it to optimize the AI ​​model. This emotion recognition data includes the user's emotional state as a parameter and is used to improve the accuracy of the model. The server then encrypts the trained AI model and returns it to each device.

[1314] Terminal side processing

[1315] The device collects location data from the smartphone. It uses GPS data and Wi-Fi tower information to obtain the user's current location and movement history. The collected data is anonymized or encrypted to prevent eavesdropping or tampering with the data when it is sent to the server.

[1316] Once the AI ​​model is returned from the server, the device uses it to further learn its own location data. In addition to this learning, the device also has a built-in emotion engine that recognizes the user's emotions, analyzing the user's voice data and facial expressions to recognize emotions. This emotion recognition data is then used to further refine the AI ​​model.

[1317] User-side processing

[1318] Users operate the device to collect location data from their smartphone. At this time, the data is appropriately anonymized and set up to be securely sent to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also sent to the server and used to update the AI ​​model.

[1319] The AI ​​model returned from the server is applied to the device to analyze the location data. The user can review the analysis results that incorporate emotional information and request that the AI ​​model be updated again if necessary. This system is extremely useful for marketing department users who want to simulate the effectiveness of campaigns in specific areas in advance, or for taking a personalized approach based on the user's emotional state.

[1320] Specific examples

[1321] Companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is then returned to companies A and B.

[1322] Company A's device uses the returned AI model to further train its own location data. This training allows Company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services. Company A's data analysts (users) analyze the results and use them in their marketing strategies.

[1323] In this way, the present invention enables multiple companies to safely utilize their own data while performing advanced location data analysis that incorporates emotional information. It also provides a system that simultaneously achieves advanced data utilization and privacy protection while avoiding direct data sharing between companies.

[1324] The processing flow will be explained below.

[1325] Step 1:

[1326] The device collects location data from smartphones, using GPS data and Wi-Fi tower information to determine the user's current location and movement history.

[1327] Step 2:

[1328] The device will anonymize or encrypt the location data it collects, and process it in a form that cannot be used to identify individuals.

[1329] Step 3:

[1330] The device sends the anonymized location data to the server using a secure communication protocol (e.g., HTTPS) to prevent data eavesdropping or tampering.

[1331] Step 4:

[1332] The server receives the anonymized location information data sent from multiple devices and centrally collects the data sent by each company's device.

[1333] Step 5:

[1334] The server preprocesses the received data, removing incomplete data and outliers and normalizing the data. This process prepares the data in a form suitable for learning.

[1335] Step 6:

[1336] The server integrates the pre-processed data and updates the AI ​​model, using machine learning algorithms to train the data and improve the accuracy of predictions and classifications.

[1337] Step 7:

[1338] The device uses an emotion engine that recognizes the user's emotions and analyzes voice data and facial expressions to recognize the user's emotional state.

[1339] Step 8:

[1340] The device sends the user's emotion data recognized by the emotion engine to the server using a secure communication protocol.

[1341] Step 9:

[1342] The server uses the received emotion data to optimize the AI ​​model, using the emotion data as parameters to further improve the accuracy of the model.

[1343] Step 10:

[1344] The server encrypts the trained AI model and returns it to each device. The return process is performed using a secure communication protocol (e.g., HTTPS).

[1345] Step 11:

[1346] The device receives the trained AI model returned from the server and saves it in the appropriate directory.

[1347] Step 12:

[1348] The device uses the received AI model to further train its own location data, and further training is performed in the local data environment to improve the accuracy of the model.

[1349] Step 13:

[1350] The updated device will analyze location data for specific purposes, such as area marketing and pedestrian flow analysis during disasters.

[1351] Step 14:

[1352] Users can operate the device to check the analysis results, and use them to develop marketing strategies and disaster prevention measures.

[1353] Step 15:

[1354] Users can send data to the server again as needed, and the federated learning process can be repeated to further improve the accuracy of the analysis results by incorporating additional data and the latest trends.

[1355] Example 2

[1356] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1357] Conventional location data collection systems lacked a method for effectively analyzing data while maintaining data anonymity and security. Furthermore, there was no way to integrate and analyze users' emotional states with location data, making it difficult to achieve more advanced personalization and evaluate marketing effectiveness. Furthermore, retraining models on each device and applying them efficiently was a challenge.

[1358] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, and means for analyzing the location information data based on the AI ​​model trained at the terminals. This enables advanced data analysis that takes user emotion information into account while maintaining the anonymity and security of the data. Furthermore, it also enables each terminal to retrain the model and efficiently apply it.

[1359] "Multiple terminals" refers to multiple user devices connected to the Internet, including smartphones, tablets, etc.

[1360] "Location data" refers to data about a user's current location and movement history obtained using GPS, Wi-Fi, etc.

[1361] "Redaction" is the process of anonymizing or encrypting data to protect it so that third parties cannot easily identify its contents.

[1362] A "server" is a computer system that receives, processes, and transmits data over a network, and refers to infrastructure capable of processing large amounts of data.

[1363] An "AI model" is a predictive model built using machine learning algorithms, and is a technology for data analysis and classification.

[1364] "Emotion recognition data" refers to data about a user's emotional state obtained by analyzing their vocal tone and facial expressions.

[1365] A "communication protocol" refers to the rules and methods for securely sending and receiving data, and includes HTTPS and TLS.

[1366] "Machine learning algorithm" is a general term for mathematical models and methods used to make predictions and classifications based on collected data.

[1367] "AES encryption" refers to a method of encrypting data using the Advanced Encryption Standard, providing a high level of security.

[1368] "Anonymization" is a general term for techniques that protect data privacy by removing personally identifiable information.

[1369] A "secure communication protocol" is a protocol that prevents eavesdropping or tampering with data when it is being sent or received.

[1370] This invention is a system that collects location information data from multiple devices, anonymizes the data, transmits it to a server, and updates an AI model. Furthermore, by combining it with user emotion recognition data, the accuracy of the model can be improved. The specific system configuration and operation procedure are described below.

[1371] Server-side processing

[1372] The server receives anonymized location data sent by multiple companies and individuals. The communication protocol used is HTTPS, which prevents data eavesdropping and tampering. The received data is first preprocessed to remove missing values ​​and outliers and normalize the data using the Pandas library. The specific data cleansing method used is the IQR (Interquartile Range) method.

[1373] Next, the AI ​​model is updated using the integrated location data. TensorFlow or PyTorch are typically used as machine learning frameworks. This allows for efficient learning of large amounts of data and improves the accuracy of the predictive model. For example, cluster analysis can be performed to identify user behavior patterns in specific areas.

[1374] The server also receives emotion recognition data sent from the device and integrates this data into the AI ​​model. Emotion data is obtained by analyzing voice tone and facial expressions and is added to the model as features. This process uses tools such as the Google Cloud Speech-to-Text API and OpenCV. Finally, the trained AI model is AES encrypted and securely returned to each device. The encryption is performed using the Python cryptography library.

[1375] Terminal side processing

[1376] The device uses a smartphone to collect location data. Specifically, it periodically obtains location information using Android or iOS APIs. The collected data is anonymized using RSA encryption technology. The anonymized data is sent to the server via the HTTPS protocol. At this time, a TLS certificate is used to ensure the security of the communication.

[1377] The device receives the AI ​​model returned from the server and further trains it on its own location data. Additional training is performed in a local environment using TensorFlow and other tools. The device also has a built-in emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This is done using voice recognition APIs and face recognition APIs. The analyzed emotion data is sent to the server and used to update the AI ​​model.

[1378] User-side processing

[1379] The user operates the device to collect location data from the smartphone. They confirm that the data collection is being carried out appropriately and that it is confidential. They then launch the location data collection app, turn on the GPS function, and begin data collection. The user then provides emotional data, such as voice and facial expressions, through the emotion engine. The device automatically recognizes the emotion and sends the data to the server.

[1380] The device receives the AI ​​model returned from the server and checks the results of the location data analysis. The analysis results, which take into account emotional information, can be viewed on a dashboard or app, feedback can be provided as needed, and an update of the AI ​​model can be requested.

[1381] Specific examples

[1382] For example, companies A and B use their respective devices to collect location data, anonymize the data, and send it to a server. The server receives and preprocesses the data from companies A and B, and updates the AI ​​model based on the integrated data. It also receives data from the emotion engine and optimizes the model based on this. The updated model is returned to companies A and B. Company A's device uses the returned AI model to further train its own location data. This learning allows company A's device to improve the accuracy of people flow data analysis and area marketing during disasters. By incorporating the user's emotional state into the analysis, it becomes possible to provide more personalized services.

[1383] Example prompts to input to the generative AI model

[1384] What are the specific steps to integrate location data from Company A and Company B to update the AI ​​model?

[1385] Please elaborate on how you integrate user emotion recognition data into your AI models.

[1386] Please provide specific steps for data redaction and secure transmission.

[1387] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1388] Step 1:

[1389] The server receives anonymized location data sent by companies or individuals. The input data is encrypted data sent via the HTTPS protocol. The server receives this data and saves it in storage. Specifically, the server takes the data into a receiving buffer and stores it in preparation for the next step of processing.

[1390] Step 2:

[1391] The server preprocesses the received location data. The input is the stored location data. The server uses the Pandas library to remove missing values ​​and outliers and normalize the data. The IQR (Interquartile Range) method is used to detect and eliminate outliers. The output is the preprocessed location data.

[1392] Step 3:

[1393] The server updates the AI ​​model using the preprocessed location data. The input is the combined preprocessed location data. The server applies machine learning algorithms using TensorFlow or PyTorch to train the model based on the new data. The output is an updated AI model.

[1394] Step 4:

[1395] The server receives emotion recognition data sent from the device and integrates it into the AI ​​model. The input is the emotion recognition data and the updated AI model, including voice tone and facial expression analysis data. The output is an optimized model that integrates the emotion data. This optimization is performed using the Google Cloud Speech-to-Text API and OpenCV.

[1396] Step 5:

[1397] The server encrypts the trained AI model using AES and returns it to each device. The input is an optimized model integrated with emotion data. The encryption is performed using Python's cryptography library. The output is an encrypted AI model file. The server sends this file to each device via the HTTPS protocol.

[1398] Step 6:

[1399] The device receives the encrypted AI model from the server. The input is the encrypted AI model file. The device receives this file and stores it in local storage. Specifically, the device first confirms receipt of the file and stores it in the appropriate directory.

[1400] Step 7:

[1401] The device deserializes and loads the received AI model to further train it on its own location data. The input is the encrypted AI model file and local location data. The device uses code such as TensorFlow to unpack the model and retrain it with the new data. The output is a locally updated AI model.

[1402] Step 8:

[1403] The device uses an emotion recognition engine to collect user emotional data. Inputs include the user's voice data and facial expression video. The device analyzes this to recognize emotions and saves them as data. Specifically, it captures data in real time using a camera or microphone and applies an analysis algorithm.

[1404] Step 9:

[1405] The device encrypts the emotion data it recognizes and sends it to the server. The input is the emotion recognition data obtained. The device encrypts the data using RSA encryption technology and sends it to the server via the HTTPS protocol. The output is the encrypted emotion data.

[1406] Step 10:

[1407] The user operates the device to collect location data in real time. The input is the user's settings and operations. The user launches a location collection app and enables GPS and WiFi data. The output is the collected location data.

[1408] Step 11:

[1409] The user provides emotional data through the emotion engine. The input is the user's voice and facial expressions. The device analyzes this to recognize the emotion and stores it for transmission in the next step. The output is the analyzed emotional data.

[1410] Step 12:

[1411] The user checks the analysis results on a dashboard or app. The input is the analysis results of the AI ​​model returned from the server. The user visually checks the analysis results of the location data that take into account emotional information and provides feedback as needed. The output is the user's understanding of the results and feedback.

[1412] (Application example 2)

[1413] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1414] Conventional autonomous vehicle systems have been unable to adjust vehicle settings in real time based on passengers' emotional states, and lack a means to effectively combine location information data and emotion recognition data collected from multiple devices to improve safety and comfort.

[1415] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting location information data from multiple terminals, means for anonymizing the collected location information data, means for transmitting the anonymized location information data to the server, means for integrating the transmitted multiple location information data and updating the AI ​​model, means for receiving emotion recognition data transmitted from the terminals and optimizing the AI ​​model, means for returning the updated AI model to the multiple terminals, means for further training the returned AI model at the terminals, means for analyzing the location information data based on the AI ​​model trained at the terminals, and means for adjusting vehicle settings in accordance with passenger emotions. This makes it possible to adjust vehicle settings in real time in accordance with the emotional state of passengers and achieve highly safe and comfortable autonomous driving.

[1416] "Multiple terminals" refers to multiple electronic devices, and in particular to portable information processing devices such as smartphones and tablets.

[1417] "Location data" refers to data including location coordinates and movement history obtained from GPS and Wi-Fi towers.

[1418] "Redaction" refers to the process of making personally identifiable information unidentifiable by anonymizing or encrypting data.

[1419] "Server" refers to a computer system that stores, processes, and distributes data over a network.

[1420] "Emotion recognition data" is data that indicates the emotional state of the user, obtained using voice data and facial expression analysis.

[1421] "AI model" refers to an artificial intelligence model created using machine learning algorithms that learns the characteristics of data and makes predictions and classifications.

[1422] "Updating" refers to the process of retraining an existing model based on new data to improve its accuracy and performance.

[1423] "Optimizing" means adjusting various parameters and factors to maximize the performance of a model.

[1424] "Returning" is the operation of sending data or models processed by the server back to the original terminal.

[1425] "Training" refers to the process of using newly collected data on the device to further train the AI ​​model and improve its performance.

[1426] "Analyzing" means applying statistical methods and algorithms to interpret collected data and extract useful information.

[1427] "Adjusting vehicle settings based on passenger emotions" means taking actions such as changing the route or playing relaxing music if a passenger is feeling stressed.

[1428] The present invention combines a system that anonymizes location information data collected from multiple devices, transmits the data to a server, and updates an AI model with an emotion engine that recognizes user emotions. Below, we will explain in detail an embodiment based on the roles of the server, device, and user.

[1429] Server-side processing

[1430] The server receives anonymized location data sent by the company. The received data is first preprocessed to remove incomplete data and outliers and normalize the data. Next, this location data is integrated with emotion recognition data to update the AI ​​model. To update the model, a machine learning algorithm is used to learn from the data and improve the accuracy of predictions and classifications. The updated AI model is then encrypted and returned to each device.

[1431] Specific software used on the server side includes machine learning frameworks such as TensorFlow and PyTorch, and SSL / TLS protocols are used to keep data confidential.

[1432] Terminal side processing

[1433] The device collects location data from smartphones and tablets, including using GPS and Wi-Fi tower information to obtain a user's current location and movement history, and includes measures to anonymize or encrypt the data. The anonymized data is then sent to a server using a secure communications protocol (e.g., HTTPS).

[1434] The device that receives the returned AI model uses it to further train its own location data. The device also has an emotion engine that analyzes the user's voice data and facial expressions to recognize emotions. This emotion recognition data is used to adjust the vehicle's settings in real time according to the user's condition.

[1435] The specific hardware used in the device includes a GPS module, camera, and microphone, and the emotion engine uses Azure Cognitive Services and Google Cloud Vision.

[1436] User-side processing

[1437] Users operate their devices to collect location data from their smartphones or tablets. At this time, the data is appropriately anonymized and securely transmitted to the server. The device is also equipped with an emotion engine that recognizes emotions based on the user's voice and facial expressions. This emotion recognition data is also transmitted to the server and used to update the AI ​​model.

[1438] For example, if a user is riding in a self-driving vehicle, the vehicle's settings can be adjusted in real time based on the user's emotional state: for example, if the passenger is feeling stressed, the vehicle can change to a route with less traffic or play relaxing music.

[1439] Prompt Sentence Examples

[1440] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[1441] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[1442] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1443] Step 1:

[1444] The server receives location data and emotion recognition data from multiple devices. It receives anonymized location data and emotion recognition data as input and stores them in a database. Specifically, it checks the data structure when receiving the data and removes outliers and incomplete data.

[1445] Step 2:

[1446] The server preprocesses the received data. It uses location data and emotion recognition data as input and normalizes them. It standardizes the data format and performs processing to impute missing values. Specifically, it scales the range of numerical data and encodes categorical data.

[1447] Step 3:

[1448] The server updates the AI ​​model based on the preprocessed data. Using the normalized location data and emotion recognition data as input, it applies machine learning algorithms to retrain the model, improving the prediction and classification accuracy of the AI ​​model. Specifically, it tunes the model's hyperparameters and trains it until convergence is achieved.

[1449] Step 4:

[1450] The server encrypts the updated AI model and returns it to each device. It uses the latest AI model and each device's identification information as input and distributes the model using a secure communication protocol (e.g., HTTPS). Specifically, it generates an encryption key and encodes the model data.

[1451] Step 5:

[1452] The device receives the AI ​​model returned from the server and further trains it using its own data. It uses the received AI model and the location and emotion recognition data stored on the device as input. Specifically, it performs incremental training to improve the adaptability of the model.

[1453] Step 6:

[1454] The device analyzes real-time location and emotion recognition data using the latest AI models to adjust vehicle settings. It uses the passenger's current location and emotional state as input to optimize the route and environment. Specifically, it will select a route with less traffic and play relaxing music if it detects stress.

[1455] Step 7:

[1456] Users operate their devices to collect location data in real time. GPS modules and Wi-Fi information are used as input to obtain current locations and movement histories. The collected data is then anonymized or encrypted and stored in the device's memory.

[1457] Step 8:

[1458] The user recognizes their own emotional state using the device's emotion engine. Voice data and facial expressions are acquired as input from a camera or microphone, and an emotion recognition algorithm is applied. Specifically, the emotional state is extracted as a parameter and stored in a database.

[1459] Prompt Sentence Examples

[1460] The following prompts can be fed into the generative AI model to provide further insight into system design and implementation:

[1461] Please provide detailed design details for the systems installed in the autonomous vehicle. We would like to know the specific methods for collecting and analyzing location data and passenger emotion recognition data, the hardware and software used, and data confidentiality and security measures. Additionally, please provide details on the vehicle's automatic adjustment function in the event of passenger stress.

[1462] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1463] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1464] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1465] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1466] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1467] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1468] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1469] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1470] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1471] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1472] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1473] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1474] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1475] 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.

[1476] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1477] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1478] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1479] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1480] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1481] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1482] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1483] The following is further disclosed regarding the above embodiment.

[1484] (Claim 1)

[1485] means for collecting location information data from a plurality of devices;

[1486] a means for anonymizing the collected location data; and

[1487] means for transmitting the anonymized location information data to a server;

[1488] A means for integrating the multiple location data sent and updating the AI ​​model;

[1489] A means of returning the updated AI model to multiple devices;

[1490] A means to further train the returned AI model on the device, and

[1491] A system including a means for analyzing location data based on an AI model trained on the device.

[1492] (Claim 2)

[1493] 10. The system of claim 1,

[1494] The system uses an algorithm to anonymize or encrypt the location information data collected by the terminal as a means for concealing the data.

[1495] (Claim 3)

[1496] 10. The system of claim 1,

[1497] The server integrates location data and updates the AI ​​model using machine learning algorithms to improve the accuracy of the model while protecting external data.

[1498] "Example 1"

[1499] (Claim 1)

[1500] means for collecting location information data from a plurality of devices;

[1501] a means for anonymizing the collected location data; and

[1502] means for transmitting the anonymized location information data to a server;

[1503] means for pre-processing the transmitted plurality of location information data;

[1504] A means to update the AI ​​model based on the preprocessed location data; and

[1505] A means of returning the updated AI model to multiple devices;

[1506] A means to further train the returned AI model on the device, and

[1507] A means of analyzing location data based on the AI ​​model trained on the device; and

[1508] A means to encrypt and securely transmit the trained model;

[1509] A system including means for transmitting location data using a secure communication protocol.

[1510] (Claim 2)

[1511] 2. The system according to claim 1, wherein the means for concealing the location information data collected by the terminal uses an algorithm for anonymizing or encrypting the data.

[1512] (Claim 3)

[1513] The system of claim 1, wherein the server's means for integrating location data and updating the AI ​​model uses machine learning algorithms to improve the accuracy of the model while protecting external data.

[1514] "Application Example 1"

[1515] (Claim 1)

[1516] means for collecting location information data from a plurality of devices;

[1517] a means for anonymizing the collected location data; and

[1518] means for transmitting the anonymized location information data to a server;

[1519] A means for integrating the multiple location data sent and updating the AI ​​model;

[1520] A means of returning the updated AI model to multiple devices;

[1521] A means to further train the returned AI model on the device, and

[1522] A means for analyzing location data based on an AI model trained on the device to generate product recommendations and promotions based on the user's location;

[1523] The system includes a means for notifying users of generated recommendations and promotions.

[1524] (Claim 2)

[1525] 2. The system according to claim 1, wherein the means for concealing the location information data collected by the terminal uses an algorithm for anonymizing or encrypting the data.

[1526] (Claim 3)

[1527] The system of claim 1, wherein the server's means for integrating location data and updating the AI ​​model uses machine learning algorithms to improve the accuracy of the model while protecting external data.

[1528] "Example 2: Combining Emotion Engines"

[1529] (Claim 1)

[1530] means for collecting location information data from a plurality of devices;

[1531] a means for anonymizing the collected location data; and

[1532] means for transmitting the anonymized location information data to a server;

[1533] A means for integrating the multiple location data sent and updating the AI ​​model;

[1534] A means to receive emotion recognition data and optimize the AI ​​model;

[1535] A means of returning the updated AI model to multiple devices;

[1536] A means to further train the returned AI model on the device, and

[1537] A system including a means for analyzing location data based on an AI model trained on the device.

[1538] (Claim 2)

[1539] The system uses an algorithm to anonymize or encrypt the location information data collected by the terminal as a means for concealing the data.

[1540] (Claim 3)

[1541] The server integrates location data and updates the AI ​​model using machine learning algorithms to improve the accuracy of the model while protecting external data.

[1542] "Application example 2 when combining emotion engines"

[1543] (Claim 1)

[1544] means for collecting location information data from a plurality of devices;

[1545] a means for anonymizing the collected location data; and

[1546] means for transmitting the anonymized location information data to a server;

[1547] A means for integrating the multiple location data sent and updating the AI ​​model;

[1548] A means for receiving emotion recognition data sent from the device and optimizing the AI ​​model;

[1549] A means of returning the updated AI model to multiple devices;

[1550] A means to further train the returned AI model on the device, and

[1551] A means of analyzing location data based on the AI ​​model trained on the device; and

[1552] means for adjusting vehicle settings in response to passenger emotions;

[1553] A system including:

[1554] (Claim 2)

[1555] 2. The system according to claim 1, wherein the means for concealing the location information data collected by the terminal uses an algorithm for anonymizing or encrypting the data.

[1556] (Claim 3)

[1557] The system of claim 1, wherein the server integrates location data and emotion recognition data to update the AI ​​model using a machine learning algorithm to improve the accuracy of the model while protecting external data. [Explanation of symbols]

[1558] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for collecting location information data from a plurality of devices; a means for anonymizing the collected location data; and means for transmitting the anonymized location information data to a server; A means for integrating the multiple location data sent and updating the AI ​​model; A means of returning the updated AI model to multiple devices; A means to further train the returned AI model on the device, and A system including a means for analyzing location data based on an AI model trained on the device.

2. 10. The system of claim 1, The system uses an algorithm to anonymize or encrypt the location information data collected by the terminal as a means for concealing the data.

3. 10. The system of claim 1, The server integrates location data and updates the AI ​​model using machine learning algorithms to improve the accuracy of the model while protecting external data.

Images