system

Abstract

Provide a system. A method for collecting life log data; A means for preprocessing the collected life log data; means for analyzing the pre-processed data to extract behavioral patterns and preferences; means for generating personalized suggestions based on the extracted behavioral patterns and preferences; means for presenting personalized suggestions to the user; A system including a means for collecting user feedback.

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] In today's information society, Generation Z in particular is exposed to a vast amount of information, which causes them to expend a lot of energy sifting through it. This can lead to a lack of motivation to try new options, and as a result, they often end up repeating the same patterns of behavior. This problem can hinder personal growth and reduce the quality of life.

[0005] Furthermore, existing systems lack personalization based on user behavior and preferences, making it difficult to provide diverse options. Furthermore, making accurate suggestions while preserving user privacy is also a challenge. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, the present invention provides the following means: a system including means for collecting life log data, means for preprocessing the collected life log data, means for analyzing the preprocessed data and extracting behavioral patterns and preferences, means for generating personalized suggestions based on the extracted behavioral patterns and preferences, means for presenting the personalized suggestions to a user, and means for collecting user feedback.

[0007] Specifically, the system collects various life log data from users' daily lives, formats it into a specific format, and preprocesses it. During preprocessing, personal information is anonymized to protect privacy. The preprocessed data is then sent to a server and analyzed using a large-scale language model (LLM), which extracts the user's behavioral patterns and preferences with high accuracy. Furthermore, personalized suggestions are generated and presented to the user based on the extracted information, broadening the range of interests and preferences and resolving the problem of information overload.

[0008] By using this system, users can regain their motivation to try new options and live a richer life. In addition, by continuously collecting user feedback, the system can be further optimized and the accuracy of its next suggestions can be improved.

[0009] "Life log data" is data that records a user's daily activities, such as behavior, devices used, applications used, location information, calendar information, and social media posts.

[0010] "Preprocessing" refers to the process of converting raw data into an analyzable format, and includes data cleansing, anonymization, format conversion, etc.

[0011] A "behavioral pattern" refers to a tendency or habit of a user to repeatedly perform an action at a particular time or place.

[0012] "Preferences" means the preferences or interests that a user has for particular items, places, or activities.

[0013] A "large-scale language model (LLM)" is an artificial intelligence model designed to learn from massive amounts of text data and perform highly accurate human language understanding and generation.

[0014] "Personalized suggestions" are recommendations of activities, events, content, etc. that are individually optimized based on a user's behavioral patterns and preferences.

[0015] "Feedback" is data that records the actions, evaluations, impressions, and other reactions of users to suggestions.

[0016] "Anonymization" is a process that processes data so that personal information cannot be identified, thereby protecting privacy. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] The system of the present invention collects and analyzes users' daily activities and preferences, and provides personalized suggestions to resolve the problem of information overload and improve the quality of their lives. This system is mainly composed of the following elements:

[0039] 1. Collection of life log data

[0040] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0041] 2. Data Preprocessing

[0042] The data collected by the device is cleansed and anonymized to prevent personal information from being identified. Specifically, location information is rounded to within a specific range and personal information such as names is removed. The data is also converted to a specified format (e.g., JSON), encrypted, and securely sent to the server.

[0043] 3. Data Analysis

[0044] The server decodes the received data and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM) to extract user behavior patterns and preferences with high accuracy. This analysis utilizes data mining and machine learning technologies.

[0045] 4. Generating personalized suggestions

[0046] The server generates optimal suggestions for users based on their behavioral patterns and preferences. For example, for a user who enjoys visiting art museums, it generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. It also ensures diversity in the suggestions by adding information about other events that may pique a user's interest.

[0047] 5. Receiving and Displaying Proposals

[0048] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[0049] 6. Collecting User Feedback

[0050] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0051] 7. Repeat the cycle

[0052] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[0053] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[0054] The processing flow will be explained below.

[0055] Step 1:

[0056] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (such as JSON).

[0057] Step 2:

[0058] Preprocessing the lifelog data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0059] Step 3:

[0060] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0061] Step 4:

[0062] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[0063] Step 5:

[0064] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns and preferences are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or that they prefer a particular music genre.

[0065] Step 6:

[0066] Based on the analysis, the server generates personalized suggestions for the user, including relevant new activities and events based on the user's behavior and preferences, such as literary events or newly opened cafes.

[0067] Step 7:

[0068] The server generates a proposal and sends it to the terminal.

[0069] Step 8:

[0070] The device receives the suggestions from the server and notifies the user through a user interface (UI). The suggestions are displayed in a visually easy-to-understand format (e.g., a card format) and provide a link to more information.

[0071] Step 9:

[0072] The user reviews the suggestions and takes action, such as attending a suggested event or visiting a suggested location.

[0073] Step 10:

[0074] The device collects user actions and feedback, such as "I accepted the suggestion" or "The suggested location was crowded" and other information.

[0075] Step 11:

[0076] The feedback collected by the device is encrypted and sent to a server.

[0077] Step 12:

[0078] The server analyzes the feedback and refines the system's large-scale language model (LLM) and suggestion generation algorithms, improving the accuracy of suggestions based on user preferences and behavior.

[0079] Step 13:

[0080] The server and device continuously repeat this cycle to optimize the user experience.

[0081] Example 1

[0082] 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."

[0083] In modern society, users are overwhelmed with information, making it difficult to effectively obtain the information they need and make efficient decisions in their daily lives. In particular, information overload makes it difficult to find the best option for them, resulting in a decline in quality of life.

[0084] 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.

[0085] In this invention, the server includes means for collecting life log data related to the user's daily activities and preferences, means for cleansing the collected life log data to anonymize personal information, converting it into a specified data format, and then encrypting it, and means for decrypting the data and analyzing the behavioral patterns and preferences with high accuracy using a large-scale language model, thereby making it possible to provide personalized suggestions to the user.

[0086] "Life log data" is data that includes information about the user's daily activities and preferences.

[0087] "Cleansing" is a process of removing noise and missing values ​​from data and normalizing the data.

[0088] "Anonymization" is the process of removing information that can identify a specific individual and de-identifying personal information.

[0089] A "specified data format" is a format (e.g., JSON) that allows data to be easily processed and exchanged.

[0090] "Encryption" is the process of encoding data using a specific algorithm to protect it from unauthorized access.

[0091] "Decryption" is the process of restoring encrypted data to its original form.

[0092] A "large-scale language model" is a machine learning model with natural language processing capabilities that is trained using massive amounts of data.

[0093] "Behavior patterns" refer to the user's daily behavioral tendencies and recurring patterns of behavior.

[0094] "Preferences" are data that indicate specific tendencies and interests that a user has.

[0095] "Personalized suggestions" refer to information and suggestions that are customized based on a user's individual behavioral patterns and preferences.

[0096] "Visually easy to understand" means using a user interface to present the suggestions in a format that is easy for the user to understand.

[0097] "Feedback" refers to the actions taken by the user in response to the proposal, as well as their subsequent impressions and evaluations.

[0098] "Used for cyclical improvement" means continually optimizing the system's algorithms and models based on collected feedback.

[0099] MODE FOR CARRYING OUT THE INVENTION

[0100] The system of the present invention collects and analyzes a user's daily activities and preferences to provide personalized suggestions, thereby eliminating the problem of information overload and improving the user's quality of life. This system is implemented mainly using the following hardware and software.

[0101] System Configuration

[0102] The system consists of the following components:

[0103] 1. Device: User's smartphone or tablet

[0104] 2. Server: Cloud-based data analysis system

[0105] 3. Software: Smartphone sensor API, large-scale language model (LLM), machine learning framework (e.g., PyTorch, TENSORFLOW (registered trademark))

[0106] Specific processing

[0107] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it obtains location information using GPS, usage history from app usage history APIs, music playback history from music app APIs, post content from SNS APIs, and schedule information from calendar APIs.

[0108] The collected data is cleansed on the device and processed to anonymize personal information. For example, location information is rounded to a specific precision and fields such as name and phone number are removed. The data is then converted to a specified format (e.g., JSON) and encrypted using an encryption algorithm (e.g., AES). The encrypted data is then sent to a server using a secure communication protocol (e.g., HTTPS).

[0109] The server decrypts the received data and inputs it into a machine learning model. Specifically, the data is read using a Python data processing library (e.g., Pandas), and large-scale language models (LLMs) and data mining techniques are used to analyze user behavior patterns and preferences with high accuracy.

[0110] Based on the analysis results, the server generates personalized suggestions for the user. For example, for a user who likes to visit art museums, it provides information on new exhibitions and nearby cafes. This ensures the diversity and accuracy of suggestions and efficiently provides information that catches the user's interest.

[0111] The generated suggestions are sent to the device and displayed visually to the user through a user interface (UI), often in the form of cards that allow easy access to detailed information.

[0112] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and subsequent feedback (e.g., whether they enjoyed the event), and then sends the formatted data back to the server.

[0113] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next suggestions.Continual optimization based on user feedback leads to sustained improvements in the system's suggestion accuracy and user satisfaction.

[0114] Prompt Sentence Examples

[0115] Examples of prompts include:

[0116] "Consider that the user has a habit of jogging on Saturdays, suggest jogging events they can join on sunny Saturdays, and nearby healthy restaurants."

[0117] In this way, the system of the present invention improves the quality of a user's life by providing personalized suggestions based on the user's daily activities and preferences.

[0118] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0119] Step 1: Collecting Lifelogging Data

[0120] The device collects life log data from various sensors and apps on the smartphone. The device obtains location information from the GPS sensor and app usage history from the app usage history API. It also obtains music playback history from the music app API and schedule information from the calendar API. This data is saved along with time information and used as input data.

[0121] Specific behavior:

[0122] Input: GPS sensor location information, app usage history API data, music playback history API data, calendar API data.

[0123] What it does: The device periodically reads location information from the GPS sensor, saves the current location data, and simultaneously sends a request to the music app's play history API to retrieve a list of recently played songs.

[0124] Output: Life log data such as location information, app usage history, music playback history, and calendar information.

[0125] Step 2: Preprocessing the data

[0126] The lifelog data collected by the device is cleansed and personal information is anonymized. For example, location information is rounded to a specific accuracy and fields such as name and phone number are removed. The data is then converted to JSON format and encrypted using the AES encryption algorithm. The encrypted data is then sent to a server using a secure communication protocol (HTTPS).

[0127] Specific behavior:

[0128] Input: Collected life log data (location information, app usage history, music playback history, calendar information).

[0129] What it does: It rounds the longitude and latitude of the location data to two decimal places, replaces all name and phone number fields with blanks or "", then converts the data to JSON format and encrypts it with the AES encryption algorithm.

[0130] Output: Encrypted data in JSON format.

[0131] Step 3: Analyze the data

[0132] The server decrypts the encrypted data it receives and distributes it to various systems. Next, the received data is input into a machine learning model (e.g., a large-scale language model trained using PyTorch) to analyze user behavior patterns and preferences. This analysis involves data mining techniques.

[0133] Specific behavior:

[0134] Input: Encrypted JSON formatted data.

[0135] How it works: The server decrypts the AES-encrypted data, reads the obtained JSON data using a Python data processing library (e.g., Pandas), and inputs the data into a machine learning model. After that, data mining techniques are used to analyze user behavior patterns and preferences.

[0136] Output: Analysis results of user behavior patterns and preferences.

[0137] Step 4: Generate personalized suggestions

[0138] The server generates optimal suggestions for the user based on the analysis results. These suggestions include new events and suitable spots. For example, for a user who likes to visit art museums, the server can provide information on special exhibitions at new museums and nearby cafes.

[0139] Specific behavior:

[0140] Input: Analysis results of user behavior patterns and preferences.

[0141] Specific operation: Based on the results of behavioral pattern analysis, queries are sent to event databases that are likely to be of interest to the user (e.g., museum APIs, event calendar APIs) to obtain appropriate event information.

[0142] Output: Personalized suggestions for the user.

[0143] Step 5: Receive and view proposals

[0144] The device receives the proposal data from the server and displays it on the user interface (UI). The proposals are presented to the user in a visually easy-to-understand format, such as a card, allowing for easy access to detailed information.

[0145] Specific behavior:

[0146] Input: Personalized suggestions for the user.

[0147] Specific operation: The device analyzes the proposal data received from the server and displays it in a card-style UI within the mobile app. Tapping the card opens the event details page.

[0148] Output: A visually friendly suggestion to the user.

[0149] Step 6: Gather user feedback

[0150] The device collects the user's actions and feedback regarding the proposal, such as whether the proposal was accepted or rejected, and subsequent impressions and evaluations, and sends the collected data to the server.

[0151] Specific behavior:

[0152] Input: User actions and feedback on suggestions.

[0153] Specific behavior: Records the user's reaction when clicking on an event suggestion and accepts feedback using rating buttons (e.g., star ratings). Feedback is saved in JSON format, AES encrypted, and sent to the server.

[0154] Output: Feedback data.

[0155] Step 7: Repeat the cycle

[0156] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. Based on the feedback, the system's algorithms and models are continuously optimized.

[0157] Specific behavior:

[0158] Input: Feedback data.

[0159] Specific operation: The feedback data is input back into the machine learning model and the model weights are updated, allowing the model to learn from the new data and improve the accuracy of the next prediction.

[0160] Output: Optimized large-scale language models and systems.

[0161] (Application example 1)

[0162] 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."

[0163] Conventional autonomous vehicle systems are inconvenient for users because they cannot provide individually optimized route guidance based on the user's life log data and preferences. Therefore, to improve the user experience, it is necessary to provide individually optimized route guidance through the collection and analysis of life log data.

[0164] 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.

[0165] In this invention, the server includes means for collecting life log data, means for preprocessing the collected life log data, means for analyzing the preprocessed data and extracting behavioral patterns and preferences, means for generating optimal route guidance based on the extracted behavioral patterns and preferences, and means for providing the optimal route guidance to an autonomous vehicle, thereby enabling the provision of individually optimized route guidance based on the user's behavioral patterns and preferences.

[0166] "Life log data" refers to information about a user's daily activities and preferences.

[0167] "Pre-processing" refers to the process of anonymizing and cleansing collected data.

[0168] "Behavioral patterns" refer to regularities regarding a user's specific behavior and its frequency, time, location, etc.

[0169] "Preferences" refer to the preferences or interests a user has for particular activities, places, or things.

[0170] "Personalized suggestions" refers to suggestions that are specifically tailored based on a user's behavioral patterns and preferences.

[0171] "Optimal route guidance" refers to the most appropriate route and destination for the user, generated based on the user's life log data and preferences.

[0172] "Feedback" refers to the reactions and opinions of users regarding suggestions they receive.

[0173] An "autonomous vehicle" refers to a vehicle that operates autonomously without the need for human operation.

[0174] The system of this invention collects and analyzes the user's daily activities and preferences, and provides optimal route guidance to an autonomous vehicle. To implement the invention, the following hardware and software are used for processing.

[0175] Hardware and software used

[0176] 1. Smartphone: A device that collects life log data from users (location information, app usage history, music history, etc.).

[0177] 2. Server: A device that preprocesses and analyzes lifelog data, generates personalized suggestions, and generates optimal route guidance.

[0178] 3. Self-driving vehicle: A vehicle that receives the generated optimal route guidance and drives autonomously.

[0179] 4. Large-scale language models (e.g., GPT-3 (registered trademark)): Machine learning models for analyzing user behavioral patterns and preferences and generating personalized suggestions.

[0180] Data collection and preprocessing

[0181] Smartphones periodically collect life log data such as user location information, app usage history, and music history. The collected data is pre-processed by anonymization and encryption and then securely transmitted to a server.

[0182] Data analysis and proposal generation

[0183] The server analyzes the received data using a large-scale language model (LLM) to extract the user's behavioral patterns and preferences. Based on the results of this analysis, it generates individually optimized route guidance. For example, it can provide recommended routes via cafes frequently visited by the user or tourist spots for potential customers.

[0184] Providing optimal route guidance

[0185] The server generates optimal route guidance, which is then sent to the autonomous vehicle, which then drives autonomously according to the guidance. The user can visually check the proposed route guidance through a smartphone app.

[0186] Gathering feedback and optimizing the system

[0187] The smartphone collects user feedback, which is then sent to the server, which then uses the feedback to optimize the entire system and improve the accuracy of the next suggestions.

[0188] Examples of concrete examples and prompts

[0189] Example 1:

[0190] Based on the user's behavioral data when commuting each morning, the system suggests a route that avoids traffic jams and allows them to stop at their favorite cafe.

[0191] Example 2:

[0192] While the user is visiting a tourist spot, the system suggests new tourist spots and restaurants based on past visit data.

[0193] Example prompt sentence:

[0194] Based on the user's behavioral data, generate optimal route guidance based on the following information:

[0195] Current location information (latitude: 35.6895, longitude: 139.6917)

[0196] Past visit history (e.g. cafes, restaurants, tourist spots)

[0197] Commuting route history

[0198] Example 1:

[0199] Destination:Office

[0200] Commuting time: 8:00-9:00 AM

[0201] Favorite stop: Cafe

[0202] Example 2:

[0203] Tourist attractions: famous places

[0204] Favorite restaurant: Restaurant

[0205] Past visit history: tourist attractions, shopping malls

[0206] Each suggestion should be returned in JSON format.

[0207] As described above, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without worrying about information overload.

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

[0209] Step 1: Collecting Lifelogging Data

[0210] Smartphone devices periodically collect life log data, such as location information, app usage history, and music history, from users' daily lives. Raw data obtained from smartphone sensors and apps is used as input. This data is collected using location information services, app monitoring services, music playback history services, etc.

[0211] Step 2: Preprocessing the data

[0212] The collected lifelog data is anonymized and cleansed on the smartphone device. This includes rounding location data to a specific range and removing personally identifiable information. The cleansed data is converted into a specified format (e.g., JSON) and securely sent to a server. The input is raw lifelog data, and the output is anonymized and formatted data.

[0213] Step 3: Analyze the data

[0214] The received data is decoded on the server side and analyzed using a large-scale language model (LLM). Machine learning algorithms and data mining techniques are used for the analysis to extract user behavioral patterns and preferences. The input for the analysis is preprocessed data, and the output is information indicating the user's behavioral patterns and preferences.

[0215] Step 4: Generate optimal route directions

[0216] The server generates optimal route guidance for the user based on the extracted behavioral patterns and preferences. This generation utilizes a large-scale language model and takes into account the user's past behavioral data and preferences. The input is the behavioral patterns and preferences resulting from the analysis, and the output is personalized route guidance.

[0217] Step 5: Provide and display directions

[0218] The route guidance generated by the server is sent to the autonomous vehicle. The autonomous vehicle then drives autonomously according to the received route guidance. The user can visually check the proposed route guidance via a smartphone app. The input is the generated route guidance, and the output is the vehicle's driving instructions and the display on the app.

[0219] Step 6: Gather user feedback

[0220] The user provides feedback on the proposed route guidance through a smartphone app. For example, they send information such as whether the route was comfortable or if there are areas for improvement. The input is the user's feedback, and the output is the feedback data. The feedback data is sent to the server and used to improve the accuracy of the next proposal.

[0221] Step 7: Optimize your system

[0222] The server analyzes the collected feedback data and refines the generative AI model and the entire system, improving the accuracy of the next proposal. The input is the feedback data, and the output is an improved system model and proposal accuracy.

[0223] 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.

[0224] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[0225] 1. Collection of life log data

[0226] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0227] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the data as emotion data. This data is also saved in a specific format (e.g., JSON).

[0228] 2. Data Preprocessing

[0229] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (names and contact details) is removed, and the collected data is converted into a unified format.

[0230] 3. Data transmission

[0231] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0232] 4. Data Analysis

[0233] The server decodes the data received from the device and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM). This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[0234] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[0235] 5. Generating personalized suggestions

[0236] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. The server also adds information about other events that may pique a user's interest, ensuring diversity in the suggestions.

[0237] 6. Receiving and Displaying Proposals

[0238] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[0239] 7. Collecting User Feedback

[0240] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0241] 8. Repeat the cycle

[0242] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[0243] As a specific example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time suggestions are made, it will be more likely to recommend a different cafe or new menu that makes the user feel happy. In this way, the system of the present invention supports the user's daily life in various ways, helping them make fun and new choices without worrying about information overload.

[0244] The processing flow will be explained below.

[0245] Step 1:

[0246] The device collects life log data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0247] Step 2:

[0248] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the emotion data. This emotion data is also saved in a specific format (e.g., JSON).

[0249] Step 3:

[0250] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0251] Step 4:

[0252] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0253] Step 5:

[0254] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[0255] Step 6:

[0256] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis results may include "a user tends to frequently visit a particular cafe on weekdays" or "a user prefers a particular music genre."

[0257] Step 7:

[0258] Based on the analysis results, the server generates personalized suggestions suited to the user. These suggestions include new activities and events that are highly relevant based on the user's behavior and preferences. For example, it generates "information about special exhibitions for a person who likes to visit art museums" and "suggestions to visit popular cafes nearby." It also adds information about other events that may pique the user's interest, ensuring diversity in the suggestions.

[0259] Step 8:

[0260] The device receives the suggestion data from the server, which includes detailed information related to the user's behavior, preferences, and emotions.

[0261] Step 9:

[0262] The device will display the suggestions in a user interface (UI) in a visually understandable format, such as a card, allowing the user to easily access the details.

[0263] Step 10:

[0264] The user takes action on the suggested activity or event, for example, attending an event based on the suggestion or visiting a suggested location.

[0265] Step 11:

[0266] The device collects user actions and feedback, such as "I accepted the suggestion," "The suggested location was crowded," and "I enjoyed the event."

[0267] Step 12:

[0268] The device formats the collected feedback, encrypts it, and sends it to a server.

[0269] Step 13:

[0270] The server analyzes the feedback data and refines the system's large-scale language model (LLM) and proposal generation algorithm, which improves the accuracy of the next proposal.

[0271] Step 14:

[0272] The server and device continuously repeat this process to optimize the user experience, and the data acquired at each step is constantly fed back into the model, continuously improving the accuracy of the system's suggestions and user satisfaction.

[0273] Example 2

[0274] 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."

[0275] In modern society, the problem of information overload is serious, making it difficult for users to find the information they truly need in their daily lives. Furthermore, there is a lack of systems that provide personalized suggestions based on users' behavior, preferences, and emotions. Furthermore, there is a need for technology development that can appropriately reflect user feedback to improve the quality of suggestions.

[0276] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0277] In this invention, the server includes a means for collecting life log data, a means for preprocessing the collected life log data, and a means for analyzing the preprocessed data to extract behavioral patterns, preferences, and emotions, thereby providing users with optimized personalized suggestions, solving the problem of information overload, and improving the quality of life of users.

[0278] "Life log data" refers to a variety of data collected from a user's daily life, such as location information, app usage history, music playback history, social media posts, and calendar information.

[0279] "Preprocessing" refers to the process of cleansing collected data, anonymizing it, standardizing its format, and so on.

[0280] "Behavioral patterns" are data that indicate regular tendencies and habits in the user's daily activities.

[0281] "Preferences" is data that indicates the things a user likes and areas of interest.

[0282] "Emotion" is data that indicates the user's emotional state and is collected using techniques such as facial recognition and voice analysis.

[0283] "Personalized suggestions" are suggestions or recommendations that are individually provided based on a user's behavioral patterns, preferences, and emotions.

[0284] "Feedback" refers to data such as the actions taken by users in response to suggestions and the ratings and opinions they provide thereafter.

[0285] "Anonymization" is the process of protecting personal information by removing any information that could identify an individual from the data.

[0286] A "proposal generator" is a mechanism for creating user-specific proposals based on collected and analyzed data.

[0287] "Optimizing the whole system" is the process of adjusting the system's algorithms and parameters based on collected feedback to improve the accuracy of suggestions.

[0288] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[0289] 1. Collection of life log data

[0290] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0291] The device uses an emotion engine (e.g., Affectiva, IBM Watson®) to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the emotion data. This data is saved in JSON format.

[0292] 2. Data Preprocessing

[0293] The device preprocesses the collected lifelog data and emotion data. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0294] 3. Data transmission

[0295] The terminal encrypts the preprocessed data and transmits it to the server using a secure communication protocol (e.g., HTTPS).

[0296] 4. Data Analysis

[0297] The server decodes the data received from the device and distributes it to each system. Next, it analyzes the data using a large-scale language model (e.g., OpenAI® GPT, Google® BERT). This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[0298] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[0299] 5. Generating personalized suggestions

[0300] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at a museum and a popular cafe nearby. It also adds information about other events that may interest the user, ensuring diversity in the suggestions.

[0301] 6. Receiving and Displaying Proposals

[0302] The device receives the suggestions from the server and displays them to the user in a visually understandable manner. The user interface (UI) presents the suggestions in a card format, allowing easy access to detailed information.

[0303] 7. Collecting User Feedback

[0304] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0305] 8. Repeat the cycle

[0306] The server uses the received feedback to refine the large-scale language model (e.g., OpenAI GPT, Google BERT) and the entire system to improve the accuracy of the next recommendation. By constantly iterating optimization based on the data provided by the user, the system's recommendation accuracy and user satisfaction are continuously improved.

[0307] Specific examples

[0308] For example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time it makes suggestions, it will recommend a different cafe or new menu that is more likely to make the user feel happy.

[0309] Example prompts for generative AI models

[0310] 1. Based on the user's emotions and behavioral patterns, suggest fun events for this weekend.

[0311] 2. Based on recent lifelog data, make a list of new cafes recommended for this user.

[0312] 3. Suggest new music albums based on this user's favorite music genres.

[0313] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[0314] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0315] Step 1: Collecting Lifelogging Data

[0316] The device collects life log data. The device automatically collects data from various sensors (GPS, accelerometer, gyroscope, etc.) and apps on the user's smartphone.

[0317] Input: Raw data from smartphone sensors (e.g., location information, app usage history, music playback history, social media posts, calendar information)

[0318] Processing: Acquires data from sensors and stores it in a unified format (e.g. JSON).

[0319] Output: Collected lifelog data (JSON format)

[0320] Specific operation: The device uses the GPS sensor to obtain the user's location information and saves it as location data (e.g., latitude, longitude) in JSON format.

[0321] Step 2: Collecting emotion data

[0322] The terminal activates an emotion engine and collects the user's emotion data.

[0323] Input: Data based on facial recognition, voice analysis, and input text analysis

[0324] Processing: An emotion engine analyzes the collected data, identifies the emotional state, and stores it as data (e.g., using Affectiva or IBM Watson).

[0325] Output: Emotion data (JSON format)

[0326] Specific operation: The device uses the camera to analyze the user's facial expressions and collect emotional data such as "happiness" or "sadness."

[0327] Step 3: Preprocessing the data

[0328] Preprocess the life log data and emotion data collected by the device.

[0329] Input: Collected life log data and emotion data

[0330] Processing: Cleansing, anonymizing, and formatting the data.

[0331] Data cleansing specifically: Detecting and removing incomplete data.

[0332] Specific actions of anonymization: Removing user names and contact information from social media post data.

[0333] Specific behavior of format conversion: The location data is rounded within a specific radius and converted into a unified format (e.g., JSON).

[0334] Output: Preprocessed data (JSON format)

[0335] Step 4: Sending data

[0336] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0337] Input: Preprocessed data (JSON format)

[0338] Processing: Encrypting data and transmitting it using secure communications protocols.

[0339] What encryption does: Encrypts data using AES encryption.

[0340] Specific transmission behavior: Encrypted data is sent to the server using the HTTPS protocol.

[0341] Output: Encrypted data sent to the server

[0342] Step 5: Analyze the data

[0343] The server decrypts the data received from the terminal and distributes it to each system.

[0344] Input: Encrypted data

[0345] Processing: The data is decoded and analyzed using a large-scale language model (e.g., OpenAI GPT, Google BERT).

[0346] Specific operations for decrypting data: Decrypt the data using AES decryption.

[0347] Specific analysis steps: Behavioral pattern data is input into a large-scale language model to extract patterns such as "frequently visiting a specific cafe on weekdays."

[0348] Output: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[0349] Step 6: Generate personalized suggestions

[0350] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions.

[0351] Input: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[0352] Processing: Generate suggestions based on behavioral patterns, preferences, and emotions.

[0353] Specific operation of suggestion generation: Based on the user's museum-loving behavioral patterns and emotional data, suggestions are generated to visit a new special exhibition at a museum and a popular cafe nearby.

[0354] Output: Personalized suggestions

[0355] Step 7: Receive and view proposals

[0356] The terminal receives the proposal from the server and displays it to the user in a visually easy-to-understand manner.

[0357] Input: Proposal sent by the server

[0358] Action: Display the suggestion in the UI.

[0359] Display behavior: The suggestion is displayed in the UI as a card, with links to more information and a map.

[0360] Output: Visualized proposal

[0361] Step 8: Gather user feedback

[0362] The terminal collects the actions the user takes in response to the suggestions and subsequent feedback.

[0363] Input: User actions and feedback data

[0364] Processing: Analyze behavioral records and collect feedback.

[0365] Specific feedback behavior: The device detects from location information whether the user has visited the suggested cafe and records that behavior.

[0366] Output: Collected feedback (in JSON format)

[0367] Step 9: Send feedback and optimize the system

[0368] The device sends feedback to the server, which then optimizes the entire system.

[0369] Input: Collected feedback data

[0370] Processing: Analyzing feedback data and adjusting algorithms and parameters across the system.

[0371] Specific behavior of feedback analysis: Adjusting the system's suggestion algorithm based on feedback data.

[0372] Output: Optimized system settings

[0373] This makes it possible to continue to provide optimal suggestions to users at all times.

[0374] (Application example 2)

[0375] 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."

[0376] In modern society, users are surrounded by a vast amount of information, making it difficult to properly manage their health. In particular, the lack of personalized advice that takes into account each user's daily life and health condition makes it difficult to effectively maintain health. Furthermore, existing health management systems have difficulty providing advice that fully reflects the user's emotions and preferences.

[0377] The identification processing 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 life log data and emotional data, means for preprocessing the collected data, and means for analyzing the preprocessed data and generating personalized health advice based on the behavioral patterns, preferences, and emotional data. This enables highly accurate health advice that takes into account the user's individual health data and emotional state.

[0378] "Life log data" is digital data that records the actions and activities of a user in their daily life.

[0379] "Emotional data" is data that reflects the user's emotional state, and is information collected using techniques such as facial recognition, voice analysis, and input text analysis.

[0380] "Preprocessing" refers to the process of cleansing, anonymizing, and formatting collected data to make it easier to analyze.

[0381] "Behavioral patterns" are data extracted from repetitive actions and habits in a user's daily life.

[0382] "Preferences" are data that refer to the preferences and tendencies that a user shows toward specific things.

[0383] "Personalized health advice" refers to health management suggestions that are individually provided based on a user's behavioral patterns, preferences, and emotional data.

[0384] "Feedback" is data that indicates the actions taken by the user in response to the advice provided and their impressions.

[0385] "Data analysis" is the process of automatically extracting and analyzing user behavioral patterns and preferences based on collected and preprocessed data.

[0386] This invention is a system that solves the problem of information overload and improves the quality of a user's life by providing personalized suggestions that take into account the user's daily activities, preferences, and emotions. The implementation of this system consists of the following elements:

[0387] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc., and uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects them as emotional data. This data is saved in a specific format (e.g., JSON).

[0388] The collected lifelog data and emotion data are preprocessed on the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0389] The preprocessed data is encrypted on the device and sent to the server using a secure communication protocol (e.g., HTTPS). The server decrypts the data received from the device and distributes it to various systems. The data is then analyzed using a large-scale language model (LLM). This analysis process automatically extracts user behavioral patterns, preferences, and emotions.

[0390] Based on the analysis results, the server generates personalized health advice. For example, it provides specific advice on the appropriate amount of exercise, diet, and sleep time based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is presented to the user on the device in a visually easy-to-understand format.

[0391] The user takes action on the advice they receive and provides feedback. This allows the user to collect the actions they take in response to the advice (e.g., whether they accepted or rejected the advice) and their subsequent thoughts (e.g., whether the advice was helpful). The collected feedback data is formatted on the device and then sent back to the server.

[0392] The server uses the received feedback to refine the large-scale language model and the entire system, improving the accuracy of the next proposal, creating a cycle of continuous improvement in the system's proposal accuracy and user satisfaction.

[0393] As a specific example, after a user finishes their morning walk, data such as the number of steps, heart rate, dietary content, and sleep time are recorded on the smartphone, and the app generates suggestions based on this data and displays them to the user using the example prompt, "Please provide me with health management advice based on my daily health data." In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them make fun and new choices without worrying about information overload.

[0394] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0395] Step 1:

[0396] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Inputs include location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (e.g., JSON format).

[0397] Step 2:

[0398] The device preprocesses the collected lifelog data and emotion data. The data collected in step 1 is used as input. Specific operations include cleansing the data, for example, rounding location information to a specific radius and removing personally identifiable information (names and contact details). The device also converts the data into a unified format. The output is the cleansed, anonymized data.

[0399] Step 3:

[0400] The terminal encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS). As input, the preprocessed data from step 2 is used, and the output is the encrypted data sent to the server.

[0401] Step 4:

[0402] The server decrypts the encrypted data received from the terminal. As input, it uses the encrypted data sent in step 3. After decryption, the data is distributed to various systems within the server. The output is the decrypted data.

[0403] Step 5:

[0404] The server analyzes the data using a large-scale language model (LLM). The distributed decoded data is used as input. This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. Specifically, it analyzes information such as the user's tendency to frequently visit a particular cafe on weekdays or their preference for a particular music genre. The analysis results are then output.

[0405] Step 6:

[0406] The server generates personalized health advice based on the analysis results. The analysis results obtained in step 5 are used as input. Specific operations include providing specific advice on appropriate amounts of exercise, dietary content, sleep duration, etc. based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is output.

[0407] Step 7:

[0408] The terminal receives the advice from the server and presents it to the user in a visually easy-to-understand format. The health advice generated in step 6 is used as input. Specifically, the advice is presented to the user in an easy-to-read format, such as a card. The output is the advice visually presented to the user.

[0409] Step 8:

[0410] The user takes action on the presented advice and provides feedback. The advice presented in step 7 is used as input. Specific actions include inputting into the terminal whether the advice was acted upon or rejected, and subsequent impressions (e.g., whether the advice was helpful). The output is the collected feedback data.

[0411] Step 9:

[0412] The terminal sends the collected feedback data to the server. The feedback data collected in step 8 is used as input. As a specific operation, the feedback data is reformatted and sent to the server again. The output is the feedback data sent to the server.

[0413] Step 10:

[0414] The server uses the received feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. The feedback data received in step 9 is used as input. Specific operations include analyzing the feedback data and adjusting system settings and model parameters. The output is a refined system and improved accuracy of the next proposal.

[0415] 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.

[0416] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0417] 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.

[0418] [Second embodiment]

[0419] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0420] 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.

[0421] 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).

[0422] 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.

[0423] 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.

[0424] 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).

[0425] 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.

[0426] 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.

[0427] 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.

[0428] 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.

[0429] 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.

[0430] 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."

[0431] The system of the present invention collects and analyzes users' daily activities and preferences, and provides personalized suggestions to resolve the problem of information overload and improve the quality of their lives. This system is mainly composed of the following elements:

[0432] 1. Collection of life log data

[0433] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0434] 2. Data Preprocessing

[0435] The data collected by the device is cleansed and anonymized to prevent personal information from being identified. Specifically, location information is rounded to within a specific range and personal information such as names is removed. The data is also converted to a specified format (e.g., JSON), encrypted, and securely sent to the server.

[0436] 3. Data Analysis

[0437] The server decodes the received data and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM) to extract user behavior patterns and preferences with high accuracy. This analysis utilizes data mining and machine learning technologies.

[0438] 4. Generating personalized suggestions

[0439] The server generates optimal suggestions for users based on their behavioral patterns and preferences. For example, for a user who enjoys visiting art museums, it generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. It also ensures diversity in the suggestions by adding information about other events that may pique a user's interest.

[0440] 5. Receiving and Displaying Proposals

[0441] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[0442] 6. Collecting User Feedback

[0443] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0444] 7. Repeat the cycle

[0445] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[0446] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[0447] The processing flow will be explained below.

[0448] Step 1:

[0449] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (such as JSON).

[0450] Step 2:

[0451] Preprocessing the lifelog data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0452] Step 3:

[0453] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0454] Step 4:

[0455] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[0456] Step 5:

[0457] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns and preferences are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or that they prefer a particular music genre.

[0458] Step 6:

[0459] Based on the analysis, the server generates personalized suggestions for the user, including relevant new activities and events based on the user's behavior and preferences, such as literary events or newly opened cafes.

[0460] Step 7:

[0461] The server generates a proposal and sends it to the terminal.

[0462] Step 8:

[0463] The device receives the suggestions from the server and notifies the user through a user interface (UI). The suggestions are displayed in a visually easy-to-understand format (e.g., a card format) and provide a link to more information.

[0464] Step 9:

[0465] The user reviews the suggestions and takes action, such as attending a suggested event or visiting a suggested location.

[0466] Step 10:

[0467] The device collects user actions and feedback, such as "I accepted the suggestion" or "The suggested location was crowded" and other information.

[0468] Step 11:

[0469] The feedback collected by the device is encrypted and sent to a server.

[0470] Step 12:

[0471] The server analyzes the feedback and refines the system's large-scale language model (LLM) and suggestion generation algorithms, improving the accuracy of suggestions based on user preferences and behavior.

[0472] Step 13:

[0473] The server and device continuously repeat this cycle to optimize the user experience.

[0474] Example 1

[0475] 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."

[0476] In modern society, users are overwhelmed with information, making it difficult to effectively obtain the information they need and make efficient decisions in their daily lives. In particular, information overload makes it difficult to find the best option for them, resulting in a decline in quality of life.

[0477] 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.

[0478] In this invention, the server includes means for collecting life log data related to the user's daily activities and preferences, means for cleansing the collected life log data to anonymize personal information, converting it into a specified data format, and then encrypting it, and means for decrypting the data and analyzing the behavioral patterns and preferences with high accuracy using a large-scale language model, thereby making it possible to provide personalized suggestions to the user.

[0479] "Life log data" is data that includes information about the user's daily activities and preferences.

[0480] "Cleansing" is a process of removing noise and missing values ​​from data and normalizing the data.

[0481] "Anonymization" is the process of removing information that can identify a specific individual and de-identifying personal information.

[0482] A "specified data format" is a format (e.g., JSON) that allows data to be easily processed and exchanged.

[0483] "Encryption" is the process of encoding data using a specific algorithm to protect it from unauthorized access.

[0484] "Decryption" is the process of restoring encrypted data to its original form.

[0485] A "large-scale language model" is a machine learning model with natural language processing capabilities that is trained using massive amounts of data.

[0486] "Behavior patterns" refer to the user's daily behavioral tendencies and recurring patterns of behavior.

[0487] "Preferences" are data that indicate specific tendencies and interests that a user has.

[0488] "Personalized suggestions" refer to information and suggestions that are customized based on a user's individual behavioral patterns and preferences.

[0489] "Visually easy to understand" means using a user interface to present the suggestions in a format that is easy for the user to understand.

[0490] "Feedback" refers to the actions taken by the user in response to the proposal, as well as their subsequent impressions and evaluations.

[0491] "Used for cyclical improvement" means continually optimizing the system's algorithms and models based on collected feedback.

[0492] MODE FOR CARRYING OUT THE INVENTION

[0493] The system of the present invention collects and analyzes a user's daily activities and preferences to provide personalized suggestions, thereby eliminating the problem of information overload and improving the user's quality of life. This system is implemented mainly using the following hardware and software.

[0494] System Configuration

[0495] The system consists of the following components:

[0496] 1. Device: User's smartphone or tablet

[0497] 2. Server: Cloud-based data analysis system

[0498] 3. Software: Smartphone sensor APIs, large-scale language models (LLMs), machine learning frameworks (e.g., PyTorch, TensorFlow)

[0499] Specific processing

[0500] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it obtains location information using GPS, usage history from app usage history APIs, music playback history from music app APIs, post content from SNS APIs, and schedule information from calendar APIs.

[0501] The collected data is cleansed on the device and processed to anonymize personal information. For example, location information is rounded to a specific precision and fields such as name and phone number are removed. The data is then converted to a specified format (e.g., JSON) and encrypted using an encryption algorithm (e.g., AES). The encrypted data is then sent to a server using a secure communication protocol (e.g., HTTPS).

[0502] The server decrypts the received data and inputs it into a machine learning model. Specifically, the data is read using a Python data processing library (e.g., Pandas), and large-scale language models (LLMs) and data mining techniques are used to analyze user behavior patterns and preferences with high accuracy.

[0503] Based on the analysis results, the server generates personalized suggestions for the user. For example, for a user who likes to visit art museums, it provides information on new exhibitions and nearby cafes. This ensures the diversity and accuracy of suggestions and efficiently provides information that catches the user's interest.

[0504] The generated suggestions are sent to the device and displayed visually to the user through a user interface (UI), often in the form of cards that allow easy access to detailed information.

[0505] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and subsequent feedback (e.g., whether they enjoyed the event), and then sends the formatted data back to the server.

[0506] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next suggestions.Continual optimization based on user feedback leads to sustained improvements in the system's suggestion accuracy and user satisfaction.

[0507] Prompt Sentence Examples

[0508] Examples of prompts include:

[0509] "Consider that the user has a habit of jogging on Saturdays, suggest jogging events they can join on sunny Saturdays, and nearby healthy restaurants."

[0510] In this way, the system of the present invention improves the quality of a user's life by providing personalized suggestions based on the user's daily activities and preferences.

[0511] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0512] Step 1: Collecting Lifelogging Data

[0513] The device collects life log data from various sensors and apps on the smartphone. The device obtains location information from the GPS sensor and app usage history from the app usage history API. It also obtains music playback history from the music app API and schedule information from the calendar API. This data is saved along with time information and used as input data.

[0514] Specific behavior:

[0515] Input: GPS sensor location information, app usage history API data, music playback history API data, calendar API data.

[0516] What it does: The device periodically reads location information from the GPS sensor, saves the current location data, and simultaneously sends a request to the music app's play history API to retrieve a list of recently played songs.

[0517] Output: Life log data such as location information, app usage history, music playback history, and calendar information.

[0518] Step 2: Preprocessing the data

[0519] The lifelog data collected by the device is cleansed and personal information is anonymized. For example, location information is rounded to a specific accuracy and fields such as name and phone number are removed. The data is then converted to JSON format and encrypted using the AES encryption algorithm. The encrypted data is then sent to a server using a secure communication protocol (HTTPS).

[0520] Specific behavior:

[0521] Input: Collected life log data (location information, app usage history, music playback history, calendar information).

[0522] What it does: It rounds the longitude and latitude of the location data to two decimal places, replaces all name and phone number fields with blanks or "", then converts the data to JSON format and encrypts it with the AES encryption algorithm.

[0523] Output: Encrypted data in JSON format.

[0524] Step 3: Analyze the data

[0525] The server decrypts the encrypted data it receives and distributes it to various systems. Next, the received data is input into a machine learning model (e.g., a large-scale language model trained using PyTorch) to analyze user behavior patterns and preferences. This analysis involves data mining techniques.

[0526] Specific behavior:

[0527] Input: Encrypted JSON formatted data.

[0528] How it works: The server decrypts the AES-encrypted data, reads the obtained JSON data using a Python data processing library (e.g., Pandas), and inputs the data into a machine learning model. After that, data mining techniques are used to analyze user behavior patterns and preferences.

[0529] Output: Analysis results of user behavior patterns and preferences.

[0530] Step 4: Generate personalized suggestions

[0531] The server generates optimal suggestions for the user based on the analysis results. These suggestions include new events and suitable spots. For example, for a user who likes to visit art museums, the server can provide information on special exhibitions at new museums and nearby cafes.

[0532] Specific behavior:

[0533] Input: Analysis results of user behavior patterns and preferences.

[0534] Specific operation: Based on the results of behavioral pattern analysis, queries are sent to event databases that are likely to be of interest to the user (e.g., museum APIs, event calendar APIs) to obtain appropriate event information.

[0535] Output: Personalized suggestions for the user.

[0536] Step 5: Receive and view proposals

[0537] The device receives the proposal data from the server and displays it on the user interface (UI). The proposals are presented to the user in a visually easy-to-understand format, such as a card, allowing for easy access to detailed information.

[0538] Specific behavior:

[0539] Input: Personalized suggestions for the user.

[0540] Specific operation: The device analyzes the proposal data received from the server and displays it in a card-style UI within the mobile app. Tapping the card opens the event details page.

[0541] Output: A visually friendly suggestion to the user.

[0542] Step 6: Gather user feedback

[0543] The device collects the user's actions and feedback regarding the proposal, such as whether the proposal was accepted or rejected, and subsequent impressions and evaluations, and sends the collected data to the server.

[0544] Specific behavior:

[0545] Input: User actions and feedback on suggestions.

[0546] Specific behavior: Records the user's reaction when clicking on an event suggestion and accepts feedback using rating buttons (e.g., star ratings). Feedback is saved in JSON format, AES encrypted, and sent to the server.

[0547] Output: Feedback data.

[0548] Step 7: Repeat the cycle

[0549] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. Based on the feedback, the system's algorithms and models are continuously optimized.

[0550] Specific behavior:

[0551] Input: Feedback data.

[0552] Specific operation: The feedback data is input back into the machine learning model and the model weights are updated, allowing the model to learn from the new data and improve the accuracy of the next prediction.

[0553] Output: Optimized large-scale language models and systems.

[0554] (Application example 1)

[0555] 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."

[0556] Conventional autonomous vehicle systems are inconvenient for users because they cannot provide individually optimized route guidance based on the user's life log data and preferences. Therefore, to improve the user experience, it is necessary to provide individually optimized route guidance through the collection and analysis of life log data.

[0557] 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.

[0558] In this invention, the server includes means for collecting life log data, means for preprocessing the collected life log data, means for analyzing the preprocessed data and extracting behavioral patterns and preferences, means for generating optimal route guidance based on the extracted behavioral patterns and preferences, and means for providing the optimal route guidance to an autonomous vehicle, thereby enabling the provision of individually optimized route guidance based on the user's behavioral patterns and preferences.

[0559] "Life log data" refers to information about a user's daily activities and preferences.

[0560] "Pre-processing" refers to the process of anonymizing and cleansing collected data.

[0561] "Behavioral patterns" refer to regularities regarding a user's specific behavior and its frequency, time, location, etc.

[0562] "Preferences" refer to the preferences or interests a user has for particular activities, places, or things.

[0563] "Personalized suggestions" refers to suggestions that are specifically tailored based on a user's behavioral patterns and preferences.

[0564] "Optimal route guidance" refers to the most appropriate route and destination for the user, generated based on the user's life log data and preferences.

[0565] "Feedback" refers to the reactions and opinions of users regarding suggestions they receive.

[0566] An "autonomous vehicle" refers to a vehicle that operates autonomously without the need for human operation.

[0567] The system of this invention collects and analyzes the user's daily activities and preferences, and provides optimal route guidance to an autonomous vehicle. To implement the invention, the following hardware and software are used for processing.

[0568] Hardware and software used

[0569] 1. Smartphone: A device that collects life log data from users (location information, app usage history, music history, etc.).

[0570] 2. Server: A device that preprocesses and analyzes lifelog data, generates personalized suggestions, and generates optimal route guidance.

[0571] 3. Self-driving vehicle: A vehicle that receives the generated optimal route guidance and drives autonomously.

[0572] 4. Large-scale language models (e.g., GPT-3): Machine learning models for analyzing user behavioral patterns and preferences and generating personalized suggestions.

[0573] Data collection and preprocessing

[0574] Smartphones periodically collect life log data such as user location information, app usage history, and music history. The collected data is pre-processed by anonymization and encryption and then securely transmitted to a server.

[0575] Data analysis and proposal generation

[0576] The server analyzes the received data using a large-scale language model (LLM) to extract the user's behavioral patterns and preferences. Based on the results of this analysis, it generates individually optimized route guidance. For example, it can provide recommended routes via cafes frequently visited by the user or tourist spots for potential customers.

[0577] Providing optimal route guidance

[0578] The server generates optimal route guidance, which is then sent to the autonomous vehicle, which then drives autonomously according to the guidance. The user can visually check the proposed route guidance through a smartphone app.

[0579] Gathering feedback and optimizing the system

[0580] The smartphone collects user feedback, which is then sent to the server, which then uses the feedback to optimize the entire system and improve the accuracy of the next suggestions.

[0581] Examples of concrete examples and prompts

[0582] Example 1:

[0583] Based on the user's behavioral data when commuting each morning, the system suggests a route that avoids traffic jams and allows them to stop at their favorite cafe.

[0584] Example 2:

[0585] While the user is visiting a tourist spot, the system suggests new tourist spots and restaurants based on past visit data.

[0586] Example prompt sentence:

[0587] Based on the user's behavioral data, generate optimal route guidance based on the following information:

[0588] Current location information (latitude: 35.6895, longitude: 139.6917)

[0589] Past visit history (e.g. cafes, restaurants, tourist spots)

[0590] Commuting route history

[0591] Example 1:

[0592] Destination:Office

[0593] Commuting time: 8:00-9:00 AM

[0594] Favorite stop: Cafe

[0595] Example 2:

[0596] Tourist attractions: famous places

[0597] Favorite restaurant: Restaurant

[0598] Past visit history: tourist attractions, shopping malls

[0599] Each suggestion should be returned in JSON format.

[0600] As described above, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without worrying about information overload.

[0601] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0602] Step 1: Collecting Lifelogging Data

[0603] Smartphone devices periodically collect life log data, such as location information, app usage history, and music history, from users' daily lives. Raw data obtained from smartphone sensors and apps is used as input. This data is collected using location information services, app monitoring services, music playback history services, etc.

[0604] Step 2: Preprocessing the data

[0605] The collected lifelog data is anonymized and cleansed on the smartphone device. This includes rounding location data to a specific range and removing personally identifiable information. The cleansed data is converted into a specified format (e.g., JSON) and securely sent to a server. The input is raw lifelog data, and the output is anonymized and formatted data.

[0606] Step 3: Analyze the data

[0607] The received data is decoded on the server side and analyzed using a large-scale language model (LLM). Machine learning algorithms and data mining techniques are used for the analysis to extract user behavioral patterns and preferences. The input for the analysis is preprocessed data, and the output is information indicating the user's behavioral patterns and preferences.

[0608] Step 4: Generate optimal route directions

[0609] The server generates optimal route guidance for the user based on the extracted behavioral patterns and preferences. This generation utilizes a large-scale language model and takes into account the user's past behavioral data and preferences. The input is the behavioral patterns and preferences resulting from the analysis, and the output is personalized route guidance.

[0610] Step 5: Provide and display directions

[0611] The route guidance generated by the server is sent to the autonomous vehicle. The autonomous vehicle then drives autonomously according to the received route guidance. The user can visually check the proposed route guidance via a smartphone app. The input is the generated route guidance, and the output is the vehicle's driving instructions and the display on the app.

[0612] Step 6: Gather user feedback

[0613] The user provides feedback on the proposed route guidance through a smartphone app. For example, they send information such as whether the route was comfortable or if there are areas for improvement. The input is the user's feedback, and the output is the feedback data. The feedback data is sent to the server and used to improve the accuracy of the next proposal.

[0614] Step 7: Optimize your system

[0615] The server analyzes the collected feedback data and refines the generative AI model and the entire system, improving the accuracy of the next proposal. The input is the feedback data, and the output is an improved system model and proposal accuracy.

[0616] 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.

[0617] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[0618] 1. Collection of life log data

[0619] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0620] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the data as emotion data. This data is also saved in a specific format (e.g., JSON).

[0621] 2. Data Preprocessing

[0622] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (names and contact details) is removed, and the collected data is converted into a unified format.

[0623] 3. Data transmission

[0624] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0625] 4. Data Analysis

[0626] The server decodes the data received from the device and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM). This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[0627] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[0628] 5. Generating personalized suggestions

[0629] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. The server also adds information about other events that may pique a user's interest, ensuring diversity in the suggestions.

[0630] 6. Receiving and Displaying Proposals

[0631] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[0632] 7. Collecting User Feedback

[0633] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0634] 8. Repeat the cycle

[0635] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[0636] As a specific example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time suggestions are made, it will be more likely to recommend a different cafe or new menu that makes the user feel happy. In this way, the system of the present invention supports the user's daily life in various ways, helping them make fun and new choices without worrying about information overload.

[0637] The processing flow will be explained below.

[0638] Step 1:

[0639] The device collects life log data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0640] Step 2:

[0641] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the emotion data. This emotion data is also saved in a specific format (e.g., JSON).

[0642] Step 3:

[0643] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0644] Step 4:

[0645] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0646] Step 5:

[0647] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[0648] Step 6:

[0649] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis results may include "a user tends to frequently visit a particular cafe on weekdays" or "a user prefers a particular music genre."

[0650] Step 7:

[0651] Based on the analysis results, the server generates personalized suggestions suited to the user. These suggestions include new activities and events that are highly relevant based on the user's behavior and preferences. For example, it generates "information about special exhibitions for a person who likes to visit art museums" and "suggestions to visit popular cafes nearby." It also adds information about other events that may pique the user's interest, ensuring diversity in the suggestions.

[0652] Step 8:

[0653] The device receives the suggestion data from the server, which includes detailed information related to the user's behavior, preferences, and emotions.

[0654] Step 9:

[0655] The device will display the suggestions in a user interface (UI) in a visually understandable format, such as a card, allowing the user to easily access the details.

[0656] Step 10:

[0657] The user takes action on the suggested activity or event, for example, attending an event based on the suggestion or visiting a suggested location.

[0658] Step 11:

[0659] The device collects user actions and feedback, such as "I accepted the suggestion," "The suggested location was crowded," and "I enjoyed the event."

[0660] Step 12:

[0661] The device formats the collected feedback, encrypts it, and sends it to a server.

[0662] Step 13:

[0663] The server analyzes the feedback data and refines the system's large-scale language model (LLM) and proposal generation algorithm, which improves the accuracy of the next proposal.

[0664] Step 14:

[0665] The server and device continuously repeat this process to optimize the user experience, and the data acquired at each step is constantly fed back into the model, continuously improving the accuracy of the system's suggestions and user satisfaction.

[0666] Example 2

[0667] 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."

[0668] In modern society, the problem of information overload is serious, making it difficult for users to find the information they truly need in their daily lives. Furthermore, there is a lack of systems that provide personalized suggestions based on users' behavior, preferences, and emotions. Furthermore, there is a need for technology development that can appropriately reflect user feedback to improve the quality of suggestions.

[0669] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0670] In this invention, the server includes a means for collecting life log data, a means for preprocessing the collected life log data, and a means for analyzing the preprocessed data to extract behavioral patterns, preferences, and emotions, thereby providing users with optimized personalized suggestions, solving the problem of information overload, and improving the quality of life of users.

[0671] "Life log data" refers to a variety of data collected from a user's daily life, such as location information, app usage history, music playback history, social media posts, and calendar information.

[0672] "Preprocessing" refers to the process of cleansing collected data, anonymizing it, standardizing its format, and so on.

[0673] "Behavioral patterns" are data that indicate regular tendencies and habits in the user's daily activities.

[0674] "Preferences" is data that indicates the things a user likes and areas of interest.

[0675] "Emotion" is data that indicates the user's emotional state and is collected using techniques such as facial recognition and voice analysis.

[0676] "Personalized suggestions" are suggestions or recommendations that are individually provided based on a user's behavioral patterns, preferences, and emotions.

[0677] "Feedback" refers to data such as the actions taken by users in response to suggestions and the ratings and opinions they provide thereafter.

[0678] "Anonymization" is the process of protecting personal information by removing any information that could identify an individual from the data.

[0679] A "proposal generator" is a mechanism for creating user-specific proposals based on collected and analyzed data.

[0680] "Optimizing the whole system" is the process of adjusting the system's algorithms and parameters based on collected feedback to improve the accuracy of suggestions.

[0681] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[0682] 1. Collection of life log data

[0683] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0684] The device uses an emotion engine (e.g., Affectiva, IBM Watson) to recognize the user's emotions through facial recognition, voice analysis, input text analysis, etc., and collects the emotion data. This data is stored in JSON format.

[0685] 2. Data Preprocessing

[0686] The device preprocesses the collected lifelog data and emotion data. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0687] 3. Data transmission

[0688] The terminal encrypts the preprocessed data and transmits it to the server using a secure communication protocol (e.g., HTTPS).

[0689] 4. Data Analysis

[0690] The server decodes the data received from the device and distributes it to each system. Next, it analyzes the data using a large-scale language model (e.g., OpenAI GPT, Google BERT). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[0691] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[0692] 5. Generating personalized suggestions

[0693] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at a museum and a popular cafe nearby. It also adds information about other events that may interest the user, ensuring diversity in the suggestions.

[0694] 6. Receiving and Displaying Proposals

[0695] The device receives the suggestions from the server and displays them to the user in a visually understandable manner. The user interface (UI) presents the suggestions in a card format, allowing easy access to detailed information.

[0696] 7. Collecting User Feedback

[0697] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0698] 8. Repeat the cycle

[0699] The server uses the received feedback to refine the large-scale language model (e.g., OpenAI GPT, Google BERT) and the entire system to improve the accuracy of the next recommendation. By constantly iterating optimization based on the data provided by the user, the system's recommendation accuracy and user satisfaction are continuously improved.

[0700] Specific examples

[0701] For example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time it makes suggestions, it will recommend a different cafe or new menu that is more likely to make the user feel happy.

[0702] Example prompts for generative AI models

[0703] 1. Based on the user's emotions and behavioral patterns, suggest fun events for this weekend.

[0704] 2. Based on recent lifelog data, make a list of new cafes recommended for this user.

[0705] 3. Suggest new music albums based on this user's favorite music genres.

[0706] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[0707] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0708] Step 1: Collecting Lifelogging Data

[0709] The device collects life log data. The device automatically collects data from various sensors (GPS, accelerometer, gyroscope, etc.) and apps on the user's smartphone.

[0710] Input: Raw data from smartphone sensors (e.g., location information, app usage history, music playback history, social media posts, calendar information)

[0711] Processing: Acquires data from sensors and stores it in a unified format (e.g. JSON).

[0712] Output: Collected lifelog data (JSON format)

[0713] Specific operation: The device uses the GPS sensor to obtain the user's location information and saves it as location data (e.g., latitude, longitude) in JSON format.

[0714] Step 2: Collecting emotion data

[0715] The terminal activates an emotion engine and collects the user's emotion data.

[0716] Input: Data based on facial recognition, voice analysis, and input text analysis

[0717] Processing: An emotion engine analyzes the collected data, identifies the emotional state, and stores it as data (e.g., using Affectiva or IBM Watson).

[0718] Output: Emotion data (JSON format)

[0719] Specific operation: The device uses the camera to analyze the user's facial expressions and collect emotional data such as "happiness" or "sadness."

[0720] Step 3: Preprocessing the data

[0721] Preprocess the life log data and emotion data collected by the device.

[0722] Input: Collected life log data and emotion data

[0723] Processing: Cleansing, anonymizing, and formatting the data.

[0724] Data cleansing specifically: Detecting and removing incomplete data.

[0725] Specific actions of anonymization: Removing user names and contact information from social media post data.

[0726] Specific behavior of format conversion: The location data is rounded within a specific radius and converted into a unified format (e.g., JSON).

[0727] Output: Preprocessed data (JSON format)

[0728] Step 4: Sending data

[0729] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0730] Input: Preprocessed data (JSON format)

[0731] Processing: Encrypting data and transmitting it using secure communications protocols.

[0732] What encryption does: Encrypts data using AES encryption.

[0733] Specific transmission behavior: Encrypted data is sent to the server using the HTTPS protocol.

[0734] Output: Encrypted data sent to the server

[0735] Step 5: Analyze the data

[0736] The server decrypts the data received from the terminal and distributes it to each system.

[0737] Input: Encrypted data

[0738] Processing: The data is decoded and analyzed using a large-scale language model (e.g., OpenAI GPT, Google BERT).

[0739] Specific operations for decrypting data: Decrypt the data using AES decryption.

[0740] Specific analysis steps: Behavioral pattern data is input into a large-scale language model to extract patterns such as "frequently visiting a specific cafe on weekdays."

[0741] Output: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[0742] Step 6: Generate personalized suggestions

[0743] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions.

[0744] Input: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[0745] Processing: Generate suggestions based on behavioral patterns, preferences, and emotions.

[0746] Specific operation of suggestion generation: Based on the user's museum-loving behavioral patterns and emotional data, suggestions are generated to visit a new special exhibition at a museum and a popular cafe nearby.

[0747] Output: Personalized suggestions

[0748] Step 7: Receive and view proposals

[0749] The terminal receives the proposal from the server and displays it to the user in a visually easy-to-understand manner.

[0750] Input: Proposal sent by the server

[0751] Action: Display the suggestion in the UI.

[0752] Display behavior: The suggestion is displayed in the UI as a card, with links to more information and a map.

[0753] Output: Visualized proposal

[0754] Step 8: Gather user feedback

[0755] The terminal collects the actions the user takes in response to the suggestions and subsequent feedback.

[0756] Input: User actions and feedback data

[0757] Processing: Analyze behavioral records and collect feedback.

[0758] Specific feedback behavior: The device detects from location information whether the user has visited the suggested cafe and records that behavior.

[0759] Output: Collected feedback (in JSON format)

[0760] Step 9: Send feedback and optimize the system

[0761] The device sends feedback to the server, which then optimizes the entire system.

[0762] Input: Collected feedback data

[0763] Processing: Analyzing feedback data and adjusting algorithms and parameters across the system.

[0764] Specific behavior of feedback analysis: Adjusting the system's suggestion algorithm based on feedback data.

[0765] Output: Optimized system settings

[0766] This makes it possible to continue to provide optimal suggestions to users at all times.

[0767] (Application example 2)

[0768] 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."

[0769] In modern society, users are surrounded by a vast amount of information, making it difficult to properly manage their health. In particular, the lack of personalized advice that takes into account each user's daily life and health condition makes it difficult to effectively maintain health. Furthermore, existing health management systems have difficulty providing advice that fully reflects the user's emotions and preferences.

[0770] The identification processing 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 life log data and emotional data, means for preprocessing the collected data, and means for analyzing the preprocessed data and generating personalized health advice based on the behavioral patterns, preferences, and emotional data. This enables highly accurate health advice that takes into account the user's individual health data and emotional state.

[0771] "Life log data" is digital data that records the actions and activities of a user in their daily life.

[0772] "Emotional data" is data that reflects the user's emotional state, and is information collected using techniques such as facial recognition, voice analysis, and input text analysis.

[0773] "Preprocessing" refers to the process of cleansing, anonymizing, and formatting collected data to make it easier to analyze.

[0774] "Behavioral patterns" are data extracted from repetitive actions and habits in a user's daily life.

[0775] "Preferences" are data that refer to the preferences and tendencies that a user shows toward specific things.

[0776] "Personalized health advice" refers to health management suggestions that are individually provided based on a user's behavioral patterns, preferences, and emotional data.

[0777] "Feedback" is data that indicates the actions taken by the user in response to the advice provided and their impressions.

[0778] "Data analysis" is the process of automatically extracting and analyzing user behavioral patterns and preferences based on collected and preprocessed data.

[0779] This invention is a system that solves the problem of information overload and improves the quality of a user's life by providing personalized suggestions that take into account the user's daily activities, preferences, and emotions. The implementation of this system consists of the following elements:

[0780] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc., and uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects them as emotional data. This data is saved in a specific format (e.g., JSON).

[0781] The collected lifelog data and emotion data are preprocessed on the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0782] The preprocessed data is encrypted on the device and sent to the server using a secure communication protocol (e.g., HTTPS). The server decrypts the data received from the device and distributes it to various systems. The data is then analyzed using a large-scale language model (LLM). This analysis process automatically extracts user behavioral patterns, preferences, and emotions.

[0783] Based on the analysis results, the server generates personalized health advice. For example, it provides specific advice on the appropriate amount of exercise, diet, and sleep time based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is presented to the user on the device in a visually easy-to-understand format.

[0784] The user takes action on the advice they receive and provides feedback. This allows the user to collect the actions they take in response to the advice (e.g., whether they accepted or rejected the advice) and their subsequent thoughts (e.g., whether the advice was helpful). The collected feedback data is formatted on the device and then sent back to the server.

[0785] The server uses the received feedback to refine the large-scale language model and the entire system, improving the accuracy of the next proposal, creating a cycle of continuous improvement in the system's proposal accuracy and user satisfaction.

[0786] As a specific example, after a user finishes their morning walk, data such as the number of steps, heart rate, dietary content, and sleep time are recorded on the smartphone, and the app generates suggestions based on this data and displays them to the user using the example prompt, "Please provide me with health management advice based on my daily health data." In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them make fun and new choices without worrying about information overload.

[0787] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0788] Step 1:

[0789] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Inputs include location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (e.g., JSON format).

[0790] Step 2:

[0791] The device preprocesses the collected lifelog data and emotion data. The data collected in step 1 is used as input. Specific operations include cleansing the data, for example, rounding location information to a specific radius and removing personally identifiable information (names and contact details). The device also converts the data into a unified format. The output is the cleansed, anonymized data.

[0792] Step 3:

[0793] The terminal encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS). As input, the preprocessed data from step 2 is used, and the output is the encrypted data sent to the server.

[0794] Step 4:

[0795] The server decrypts the encrypted data received from the terminal. As input, it uses the encrypted data sent in step 3. After decryption, the data is distributed to various systems within the server. The output is the decrypted data.

[0796] Step 5:

[0797] The server analyzes the data using a large-scale language model (LLM). The distributed decoded data is used as input. This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. Specifically, it analyzes information such as the user's tendency to frequently visit a particular cafe on weekdays or their preference for a particular music genre. The analysis results are then output.

[0798] Step 6:

[0799] The server generates personalized health advice based on the analysis results. The analysis results obtained in step 5 are used as input. Specific operations include providing specific advice on appropriate amounts of exercise, dietary content, sleep duration, etc. based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is output.

[0800] Step 7:

[0801] The terminal receives the advice from the server and presents it to the user in a visually easy-to-understand format. The health advice generated in step 6 is used as input. Specifically, the advice is presented to the user in an easy-to-read format, such as a card. The output is the advice visually presented to the user.

[0802] Step 8:

[0803] The user takes action on the presented advice and provides feedback. The advice presented in step 7 is used as input. Specific actions include inputting into the terminal whether the advice was acted upon or rejected, and subsequent impressions (e.g., whether the advice was helpful). The output is the collected feedback data.

[0804] Step 9:

[0805] The terminal sends the collected feedback data to the server. The feedback data collected in step 8 is used as input. As a specific operation, the feedback data is reformatted and sent to the server again. The output is the feedback data sent to the server.

[0806] Step 10:

[0807] The server uses the received feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. The feedback data received in step 9 is used as input. Specific operations include analyzing the feedback data and adjusting system settings and model parameters. The output is a refined system and improved accuracy of the next proposal.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] [Third embodiment]

[0812] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0813] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0814] 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).

[0815] 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.

[0816] 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.

[0817] 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).

[0818] 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.

[0819] 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.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] 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."

[0824] The system of the present invention collects and analyzes users' daily activities and preferences, and provides personalized suggestions to resolve the problem of information overload and improve the quality of their lives. This system is mainly composed of the following elements:

[0825] 1. Collection of life log data

[0826] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[0827] 2. Data Preprocessing

[0828] The data collected by the device is cleansed and anonymized to prevent personal information from being identified. Specifically, location information is rounded to within a specific range and personal information such as names is removed. The data is also converted to a specified format (e.g., JSON), encrypted, and securely sent to the server.

[0829] 3. Data Analysis

[0830] The server decodes the received data and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM) to extract user behavior patterns and preferences with high accuracy. This analysis utilizes data mining and machine learning technologies.

[0831] 4. Generating personalized suggestions

[0832] The server generates optimal suggestions for users based on their behavioral patterns and preferences. For example, for a user who enjoys visiting art museums, it generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. It also ensures diversity in the suggestions by adding information about other events that may pique a user's interest.

[0833] 5. Receiving and Displaying Proposals

[0834] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[0835] 6. Collecting User Feedback

[0836] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[0837] 7. Repeat the cycle

[0838] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[0839] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[0840] The processing flow will be explained below.

[0841] Step 1:

[0842] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (such as JSON).

[0843] Step 2:

[0844] Preprocessing the lifelog data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[0845] Step 3:

[0846] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[0847] Step 4:

[0848] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[0849] Step 5:

[0850] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns and preferences are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or that they prefer a particular music genre.

[0851] Step 6:

[0852] Based on the analysis, the server generates personalized suggestions for the user, including relevant new activities and events based on the user's behavior and preferences, such as literary events or newly opened cafes.

[0853] Step 7:

[0854] The server generates a proposal and sends it to the terminal.

[0855] Step 8:

[0856] The device receives the suggestions from the server and notifies the user through a user interface (UI). The suggestions are displayed in a visually easy-to-understand format (e.g., a card format) and provide a link to more information.

[0857] Step 9:

[0858] The user reviews the suggestions and takes action, such as attending a suggested event or visiting a suggested location.

[0859] Step 10:

[0860] The device collects user actions and feedback, such as "I accepted the suggestion" or "The suggested location was crowded" and other information.

[0861] Step 11:

[0862] The feedback collected by the device is encrypted and sent to a server.

[0863] Step 12:

[0864] The server analyzes the feedback and refines the system's large-scale language model (LLM) and suggestion generation algorithms, improving the accuracy of suggestions based on user preferences and behavior.

[0865] Step 13:

[0866] The server and device continuously repeat this cycle to optimize the user experience.

[0867] Example 1

[0868] 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."

[0869] In modern society, users are overwhelmed with information, making it difficult to effectively obtain the information they need and make efficient decisions in their daily lives. In particular, information overload makes it difficult to find the best option for them, resulting in a decline in quality of life.

[0870] 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.

[0871] In this invention, the server includes means for collecting life log data related to the user's daily activities and preferences, means for cleansing the collected life log data to anonymize personal information, converting it into a specified data format, and then encrypting it, and means for decrypting the data and analyzing the behavioral patterns and preferences with high accuracy using a large-scale language model, thereby making it possible to provide personalized suggestions to the user.

[0872] "Life log data" is data that includes information about the user's daily activities and preferences.

[0873] "Cleansing" is a process of removing noise and missing values ​​from data and normalizing the data.

[0874] "Anonymization" is the process of removing information that can identify a specific individual and de-identifying personal information.

[0875] A "specified data format" is a format (e.g., JSON) that allows data to be easily processed and exchanged.

[0876] "Encryption" is the process of encoding data using a specific algorithm to protect it from unauthorized access.

[0877] "Decryption" is the process of restoring encrypted data to its original form.

[0878] A "large-scale language model" is a machine learning model with natural language processing capabilities that is trained using massive amounts of data.

[0879] "Behavior patterns" refer to the user's daily behavioral tendencies and recurring patterns of behavior.

[0880] "Preferences" are data that indicate specific tendencies and interests that a user has.

[0881] "Personalized suggestions" refer to information and suggestions that are customized based on a user's individual behavioral patterns and preferences.

[0882] "Visually easy to understand" means using a user interface to present the suggestions in a format that is easy for the user to understand.

[0883] "Feedback" refers to the actions taken by the user in response to the proposal, as well as their subsequent impressions and evaluations.

[0884] "Used for cyclical improvement" means continually optimizing the system's algorithms and models based on collected feedback.

[0885] MODE FOR CARRYING OUT THE INVENTION

[0886] The system of the present invention collects and analyzes a user's daily activities and preferences to provide personalized suggestions, thereby eliminating the problem of information overload and improving the user's quality of life. This system is implemented mainly using the following hardware and software.

[0887] System Configuration

[0888] The system consists of the following components:

[0889] 1. Device: User's smartphone or tablet

[0890] 2. Server: Cloud-based data analysis system

[0891] 3. Software: Smartphone sensor APIs, large-scale language models (LLMs), machine learning frameworks (e.g., PyTorch, TensorFlow)

[0892] Specific processing

[0893] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it obtains location information using GPS, usage history from app usage history APIs, music playback history from music app APIs, post content from SNS APIs, and schedule information from calendar APIs.

[0894] The collected data is cleansed on the device and processed to anonymize personal information. For example, location information is rounded to a specific precision and fields such as name and phone number are removed. The data is then converted to a specified format (e.g., JSON) and encrypted using an encryption algorithm (e.g., AES). The encrypted data is then sent to a server using a secure communication protocol (e.g., HTTPS).

[0895] The server decrypts the received data and inputs it into a machine learning model. Specifically, the data is read using a Python data processing library (e.g., Pandas), and large-scale language models (LLMs) and data mining techniques are used to analyze user behavior patterns and preferences with high accuracy.

[0896] Based on the analysis results, the server generates personalized suggestions for the user. For example, for a user who likes to visit art museums, it provides information on new exhibitions and nearby cafes. This ensures the diversity and accuracy of suggestions and efficiently provides information that catches the user's interest.

[0897] The generated suggestions are sent to the device and displayed visually to the user through a user interface (UI), often in the form of cards that allow easy access to detailed information.

[0898] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and subsequent feedback (e.g., whether they enjoyed the event), and then sends the formatted data back to the server.

[0899] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next suggestions.Continual optimization based on user feedback leads to sustained improvements in the system's suggestion accuracy and user satisfaction.

[0900] Prompt Sentence Examples

[0901] Examples of prompts include:

[0902] "Consider that the user has a habit of jogging on Saturdays, suggest jogging events they can join on sunny Saturdays, and nearby healthy restaurants."

[0903] In this way, the system of the present invention improves the quality of a user's life by providing personalized suggestions based on the user's daily activities and preferences.

[0904] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0905] Step 1: Collecting Lifelogging Data

[0906] The device collects life log data from various sensors and apps on the smartphone. The device obtains location information from the GPS sensor and app usage history from the app usage history API. It also obtains music playback history from the music app API and schedule information from the calendar API. This data is saved along with time information and used as input data.

[0907] Specific behavior:

[0908] Input: GPS sensor location information, app usage history API data, music playback history API data, calendar API data.

[0909] What it does: The device periodically reads location information from the GPS sensor, saves the current location data, and simultaneously sends a request to the music app's play history API to retrieve a list of recently played songs.

[0910] Output: Life log data such as location information, app usage history, music playback history, and calendar information.

[0911] Step 2: Preprocessing the data

[0912] The lifelog data collected by the device is cleansed and personal information is anonymized. For example, location information is rounded to a specific accuracy and fields such as name and phone number are removed. The data is then converted to JSON format and encrypted using the AES encryption algorithm. The encrypted data is then sent to a server using a secure communication protocol (HTTPS).

[0913] Specific behavior:

[0914] Input: Collected life log data (location information, app usage history, music playback history, calendar information).

[0915] What it does: It rounds the longitude and latitude of the location data to two decimal places, replaces all name and phone number fields with blanks or "", then converts the data to JSON format and encrypts it with the AES encryption algorithm.

[0916] Output: Encrypted data in JSON format.

[0917] Step 3: Analyze the data

[0918] The server decrypts the encrypted data it receives and distributes it to various systems. Next, the received data is input into a machine learning model (e.g., a large-scale language model trained using PyTorch) to analyze user behavior patterns and preferences. This analysis involves data mining techniques.

[0919] Specific behavior:

[0920] Input: Encrypted JSON formatted data.

[0921] How it works: The server decrypts the AES-encrypted data, reads the obtained JSON data using a Python data processing library (e.g., Pandas), and inputs the data into a machine learning model. After that, data mining techniques are used to analyze user behavior patterns and preferences.

[0922] Output: Analysis results of user behavior patterns and preferences.

[0923] Step 4: Generate personalized suggestions

[0924] The server generates optimal suggestions for the user based on the analysis results. These suggestions include new events and suitable spots. For example, for a user who likes to visit art museums, the server can provide information on special exhibitions at new museums and nearby cafes.

[0925] Specific behavior:

[0926] Input: Analysis results of user behavior patterns and preferences.

[0927] Specific operation: Based on the results of behavioral pattern analysis, queries are sent to event databases that are likely to be of interest to the user (e.g., museum APIs, event calendar APIs) to obtain appropriate event information.

[0928] Output: Personalized suggestions for the user.

[0929] Step 5: Receive and view proposals

[0930] The device receives the proposal data from the server and displays it on the user interface (UI). The proposals are presented to the user in a visually easy-to-understand format, such as a card, allowing for easy access to detailed information.

[0931] Specific behavior:

[0932] Input: Personalized suggestions for the user.

[0933] Specific operation: The device analyzes the proposal data received from the server and displays it in a card-style UI within the mobile app. Tapping the card opens the event details page.

[0934] Output: A visually friendly suggestion to the user.

[0935] Step 6: Gather user feedback

[0936] The device collects the user's actions and feedback regarding the proposal, such as whether the proposal was accepted or rejected, and subsequent impressions and evaluations, and sends the collected data to the server.

[0937] Specific behavior:

[0938] Input: User actions and feedback on suggestions.

[0939] Specific behavior: Records the user's reaction when clicking on an event suggestion and accepts feedback using rating buttons (e.g., star ratings). Feedback is saved in JSON format, AES encrypted, and sent to the server.

[0940] Output: Feedback data.

[0941] Step 7: Repeat the cycle

[0942] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. Based on the feedback, the system's algorithms and models are continuously optimized.

[0943] Specific behavior:

[0944] Input: Feedback data.

[0945] Specific operation: The feedback data is input back into the machine learning model and the model weights are updated, allowing the model to learn from the new data and improve the accuracy of the next prediction.

[0946] Output: Optimized large-scale language models and systems.

[0947] (Application example 1)

[0948] 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."

[0949] Conventional autonomous vehicle systems are inconvenient for users because they cannot provide individually optimized route guidance based on the user's life log data and preferences. Therefore, to improve the user experience, it is necessary to provide individually optimized route guidance through the collection and analysis of life log data.

[0950] 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.

[0951] In this invention, the server includes means for collecting life log data, means for preprocessing the collected life log data, means for analyzing the preprocessed data and extracting behavioral patterns and preferences, means for generating optimal route guidance based on the extracted behavioral patterns and preferences, and means for providing the optimal route guidance to an autonomous vehicle, thereby enabling the provision of individually optimized route guidance based on the user's behavioral patterns and preferences.

[0952] "Life log data" refers to information about a user's daily activities and preferences.

[0953] "Pre-processing" refers to the process of anonymizing and cleansing collected data.

[0954] "Behavioral patterns" refer to regularities regarding a user's specific behavior and its frequency, time, location, etc.

[0955] "Preferences" refer to the preferences or interests a user has for particular activities, places, or things.

[0956] "Personalized suggestions" refers to suggestions that are specifically tailored based on a user's behavioral patterns and preferences.

[0957] "Optimal route guidance" refers to the most appropriate route and destination for the user, generated based on the user's life log data and preferences.

[0958] "Feedback" refers to the reactions and opinions of users regarding suggestions they receive.

[0959] An "autonomous vehicle" refers to a vehicle that operates autonomously without the need for human operation.

[0960] The system of this invention collects and analyzes the user's daily activities and preferences, and provides optimal route guidance to an autonomous vehicle. To implement the invention, the following hardware and software are used for processing.

[0961] Hardware and software used

[0962] 1. Smartphone: A device that collects life log data from users (location information, app usage history, music history, etc.).

[0963] 2. Server: A device that preprocesses and analyzes lifelog data, generates personalized suggestions, and generates optimal route guidance.

[0964] 3. Self-driving vehicle: A vehicle that receives the generated optimal route guidance and drives autonomously.

[0965] 4. Large-scale language models (e.g., GPT-3): Machine learning models for analyzing user behavioral patterns and preferences and generating personalized suggestions.

[0966] Data collection and preprocessing

[0967] Smartphones periodically collect life log data such as user location information, app usage history, and music history. The collected data is pre-processed by anonymization and encryption and then securely transmitted to a server.

[0968] Data analysis and proposal generation

[0969] The server analyzes the received data using a large-scale language model (LLM) to extract the user's behavioral patterns and preferences. Based on the results of this analysis, it generates individually optimized route guidance. For example, it can provide recommended routes via cafes frequently visited by the user or tourist spots for potential customers.

[0970] Providing optimal route guidance

[0971] The server generates optimal route guidance, which is then sent to the autonomous vehicle, which then drives autonomously according to the guidance. The user can visually check the proposed route guidance through a smartphone app.

[0972] Gathering feedback and optimizing the system

[0973] The smartphone collects user feedback, which is then sent to the server, which then uses the feedback to optimize the entire system and improve the accuracy of the next suggestions.

[0974] Examples of concrete examples and prompts

[0975] Example 1:

[0976] Based on the user's behavioral data when commuting each morning, the system suggests a route that avoids traffic jams and allows them to stop at their favorite cafe.

[0977] Example 2:

[0978] While the user is visiting a tourist spot, the system suggests new tourist spots and restaurants based on past visit data.

[0979] Example prompt sentence:

[0980] Based on the user's behavioral data, generate optimal route guidance based on the following information:

[0981] Current location information (latitude: 35.6895, longitude: 139.6917)

[0982] Past visit history (e.g. cafes, restaurants, tourist spots)

[0983] Commuting route history

[0984] Example 1:

[0985] Destination:Office

[0986] Commuting time: 8:00-9:00 AM

[0987] Favorite stop: Cafe

[0988] Example 2:

[0989] Tourist attractions: famous places

[0990] Favorite restaurant: Restaurant

[0991] Past visit history: tourist attractions, shopping malls

[0992] Each suggestion should be returned in JSON format.

[0993] As described above, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without worrying about information overload.

[0994] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0995] Step 1: Collecting Lifelogging Data

[0996] Smartphone devices periodically collect life log data, such as location information, app usage history, and music history, from users' daily lives. Raw data obtained from smartphone sensors and apps is used as input. This data is collected using location information services, app monitoring services, music playback history services, etc.

[0997] Step 2: Preprocessing the data

[0998] The collected lifelog data is anonymized and cleansed on the smartphone device. This includes rounding location data to a specific range and removing personally identifiable information. The cleansed data is converted into a specified format (e.g., JSON) and securely sent to a server. The input is raw lifelog data, and the output is anonymized and formatted data.

[0999] Step 3: Analyze the data

[1000] The received data is decoded on the server side and analyzed using a large-scale language model (LLM). Machine learning algorithms and data mining techniques are used for the analysis to extract user behavioral patterns and preferences. The input for the analysis is preprocessed data, and the output is information indicating the user's behavioral patterns and preferences.

[1001] Step 4: Generate optimal route directions

[1002] The server generates optimal route guidance for the user based on the extracted behavioral patterns and preferences. This generation utilizes a large-scale language model and takes into account the user's past behavioral data and preferences. The input is the behavioral patterns and preferences resulting from the analysis, and the output is personalized route guidance.

[1003] Step 5: Provide and display directions

[1004] The route guidance generated by the server is sent to the autonomous vehicle. The autonomous vehicle then drives autonomously according to the received route guidance. The user can visually check the proposed route guidance via a smartphone app. The input is the generated route guidance, and the output is the vehicle's driving instructions and the display on the app.

[1005] Step 6: Gather user feedback

[1006] The user provides feedback on the proposed route guidance through a smartphone app. For example, they send information such as whether the route was comfortable or if there are areas for improvement. The input is the user's feedback, and the output is the feedback data. The feedback data is sent to the server and used to improve the accuracy of the next proposal.

[1007] Step 7: Optimize your system

[1008] The server analyzes the collected feedback data and refines the generative AI model and the entire system, improving the accuracy of the next proposal. The input is the feedback data, and the output is an improved system model and proposal accuracy.

[1009] 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.

[1010] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[1011] 1. Collection of life log data

[1012] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1013] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the data as emotion data. This data is also saved in a specific format (e.g., JSON).

[1014] 2. Data Preprocessing

[1015] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (names and contact details) is removed, and the collected data is converted into a unified format.

[1016] 3. Data transmission

[1017] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1018] 4. Data Analysis

[1019] The server decodes the data received from the device and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM). This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[1020] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[1021] 5. Generating personalized suggestions

[1022] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. The server also adds information about other events that may pique a user's interest, ensuring diversity in the suggestions.

[1023] 6. Receiving and Displaying Proposals

[1024] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[1025] 7. Collecting User Feedback

[1026] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[1027] 8. Repeat the cycle

[1028] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[1029] As a specific example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time suggestions are made, it will be more likely to recommend a different cafe or new menu that makes the user feel happy. In this way, the system of the present invention supports the user's daily life in various ways, helping them make fun and new choices without worrying about information overload.

[1030] The processing flow will be explained below.

[1031] Step 1:

[1032] The device collects life log data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1033] Step 2:

[1034] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the emotion data. This emotion data is also saved in a specific format (e.g., JSON).

[1035] Step 3:

[1036] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1037] Step 4:

[1038] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1039] Step 5:

[1040] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[1041] Step 6:

[1042] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis results may include "a user tends to frequently visit a particular cafe on weekdays" or "a user prefers a particular music genre."

[1043] Step 7:

[1044] Based on the analysis results, the server generates personalized suggestions suited to the user. These suggestions include new activities and events that are highly relevant based on the user's behavior and preferences. For example, it generates "information about special exhibitions for a person who likes to visit art museums" and "suggestions to visit popular cafes nearby." It also adds information about other events that may pique the user's interest, ensuring diversity in the suggestions.

[1045] Step 8:

[1046] The device receives the suggestion data from the server, which includes detailed information related to the user's behavior, preferences, and emotions.

[1047] Step 9:

[1048] The device will display the suggestions in a user interface (UI) in a visually understandable format, such as a card, allowing the user to easily access the details.

[1049] Step 10:

[1050] The user takes action on the suggested activity or event, for example, attending an event based on the suggestion or visiting a suggested location.

[1051] Step 11:

[1052] The device collects user actions and feedback, such as "I accepted the suggestion," "The suggested location was crowded," and "I enjoyed the event."

[1053] Step 12:

[1054] The device formats the collected feedback, encrypts it, and sends it to a server.

[1055] Step 13:

[1056] The server analyzes the feedback data and refines the system's large-scale language model (LLM) and proposal generation algorithm, which improves the accuracy of the next proposal.

[1057] Step 14:

[1058] The server and device continuously repeat this process to optimize the user experience, and the data acquired at each step is constantly fed back into the model, continuously improving the accuracy of the system's suggestions and user satisfaction.

[1059] Example 2

[1060] 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."

[1061] In modern society, the problem of information overload is serious, making it difficult for users to find the information they truly need in their daily lives. Furthermore, there is a lack of systems that provide personalized suggestions based on users' behavior, preferences, and emotions. Furthermore, there is a need for technology development that can appropriately reflect user feedback to improve the quality of suggestions.

[1062] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1063] In this invention, the server includes a means for collecting life log data, a means for preprocessing the collected life log data, and a means for analyzing the preprocessed data to extract behavioral patterns, preferences, and emotions, thereby providing users with optimized personalized suggestions, solving the problem of information overload, and improving the quality of life of users.

[1064] "Life log data" refers to a variety of data collected from a user's daily life, such as location information, app usage history, music playback history, social media posts, and calendar information.

[1065] "Preprocessing" refers to the process of cleansing collected data, anonymizing it, standardizing its format, and so on.

[1066] "Behavioral patterns" are data that indicate regular tendencies and habits in the user's daily activities.

[1067] "Preferences" is data that indicates the things a user likes and areas of interest.

[1068] "Emotion" is data that indicates the user's emotional state and is collected using techniques such as facial recognition and voice analysis.

[1069] "Personalized suggestions" are suggestions or recommendations that are individually provided based on a user's behavioral patterns, preferences, and emotions.

[1070] "Feedback" refers to data such as the actions taken by users in response to suggestions and the ratings and opinions they provide thereafter.

[1071] "Anonymization" is the process of protecting personal information by removing any information that could identify an individual from the data.

[1072] A "proposal generator" is a mechanism for creating user-specific proposals based on collected and analyzed data.

[1073] "Optimizing the whole system" is the process of adjusting the system's algorithms and parameters based on collected feedback to improve the accuracy of suggestions.

[1074] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[1075] 1. Collection of life log data

[1076] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1077] The device uses an emotion engine (e.g., Affectiva, IBM Watson) to recognize the user's emotions through facial recognition, voice analysis, input text analysis, etc., and collects the emotion data. This data is stored in JSON format.

[1078] 2. Data Preprocessing

[1079] The device preprocesses the collected lifelog data and emotion data. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1080] 3. Data transmission

[1081] The terminal encrypts the preprocessed data and transmits it to the server using a secure communication protocol (e.g., HTTPS).

[1082] 4. Data Analysis

[1083] The server decodes the data received from the device and distributes it to each system. Next, it analyzes the data using a large-scale language model (e.g., OpenAI GPT, Google BERT). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[1084] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[1085] 5. Generating personalized suggestions

[1086] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at a museum and a popular cafe nearby. It also adds information about other events that may interest the user, ensuring diversity in the suggestions.

[1087] 6. Receiving and Displaying Proposals

[1088] The device receives the suggestions from the server and displays them to the user in a visually understandable manner. The user interface (UI) presents the suggestions in a card format, allowing easy access to detailed information.

[1089] 7. Collecting User Feedback

[1090] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[1091] 8. Repeat the cycle

[1092] The server uses the received feedback to refine the large-scale language model (e.g., OpenAI GPT, Google BERT) and the entire system to improve the accuracy of the next recommendation. By constantly iterating optimization based on the data provided by the user, the system's recommendation accuracy and user satisfaction are continuously improved.

[1093] Specific examples

[1094] For example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time it makes suggestions, it will recommend a different cafe or new menu that is more likely to make the user feel happy.

[1095] Example prompts for generative AI models

[1096] 1. Based on the user's emotions and behavioral patterns, suggest fun events for this weekend.

[1097] 2. Based on recent lifelog data, make a list of new cafes recommended for this user.

[1098] 3. Suggest new music albums based on this user's favorite music genres.

[1099] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[1100] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1101] Step 1: Collecting Lifelogging Data

[1102] The device collects life log data. The device automatically collects data from various sensors (GPS, accelerometer, gyroscope, etc.) and apps on the user's smartphone.

[1103] Input: Raw data from smartphone sensors (e.g., location information, app usage history, music playback history, social media posts, calendar information)

[1104] Processing: Acquires data from sensors and stores it in a unified format (e.g. JSON).

[1105] Output: Collected lifelog data (JSON format)

[1106] Specific operation: The device uses the GPS sensor to obtain the user's location information and saves it as location data (e.g., latitude, longitude) in JSON format.

[1107] Step 2: Collecting emotion data

[1108] The terminal activates an emotion engine and collects the user's emotion data.

[1109] Input: Data based on facial recognition, voice analysis, and input text analysis

[1110] Processing: An emotion engine analyzes the collected data, identifies the emotional state, and stores it as data (e.g., using Affectiva or IBM Watson).

[1111] Output: Emotion data (JSON format)

[1112] Specific operation: The device uses the camera to analyze the user's facial expressions and collect emotional data such as "happiness" or "sadness."

[1113] Step 3: Preprocessing the data

[1114] Preprocess the life log data and emotion data collected by the device.

[1115] Input: Collected life log data and emotion data

[1116] Processing: Cleansing, anonymizing, and formatting the data.

[1117] Data cleansing specifically: Detecting and removing incomplete data.

[1118] Specific actions of anonymization: Removing user names and contact information from social media post data.

[1119] Specific behavior of format conversion: The location data is rounded within a specific radius and converted into a unified format (e.g., JSON).

[1120] Output: Preprocessed data (JSON format)

[1121] Step 4: Sending data

[1122] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1123] Input: Preprocessed data (JSON format)

[1124] Processing: Encrypting data and transmitting it using secure communications protocols.

[1125] What encryption does: Encrypts data using AES encryption.

[1126] Specific transmission behavior: Encrypted data is sent to the server using the HTTPS protocol.

[1127] Output: Encrypted data sent to the server

[1128] Step 5: Analyze the data

[1129] The server decrypts the data received from the terminal and distributes it to each system.

[1130] Input: Encrypted data

[1131] Processing: The data is decoded and analyzed using a large-scale language model (e.g., OpenAI GPT, Google BERT).

[1132] Specific operations for decrypting data: Decrypt the data using AES decryption.

[1133] Specific analysis steps: Behavioral pattern data is input into a large-scale language model to extract patterns such as "frequently visiting a specific cafe on weekdays."

[1134] Output: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[1135] Step 6: Generate personalized suggestions

[1136] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions.

[1137] Input: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[1138] Processing: Generate suggestions based on behavioral patterns, preferences, and emotions.

[1139] Specific operation of suggestion generation: Based on the user's museum-loving behavioral patterns and emotional data, suggestions are generated to visit a new special exhibition at a museum and a popular cafe nearby.

[1140] Output: Personalized suggestions

[1141] Step 7: Receive and view proposals

[1142] The terminal receives the proposal from the server and displays it to the user in a visually easy-to-understand manner.

[1143] Input: Proposal sent by the server

[1144] Action: Display the suggestion in the UI.

[1145] Display behavior: The suggestion is displayed in the UI as a card, with links to more information and a map.

[1146] Output: Visualized proposal

[1147] Step 8: Gather user feedback

[1148] The terminal collects the actions the user takes in response to the suggestions and subsequent feedback.

[1149] Input: User actions and feedback data

[1150] Processing: Analyze behavioral records and collect feedback.

[1151] Specific feedback behavior: The device detects from location information whether the user has visited the suggested cafe and records that behavior.

[1152] Output: Collected feedback (in JSON format)

[1153] Step 9: Send feedback and optimize the system

[1154] The device sends feedback to the server, which then optimizes the entire system.

[1155] Input: Collected feedback data

[1156] Processing: Analyzing feedback data and adjusting algorithms and parameters across the system.

[1157] Specific behavior of feedback analysis: Adjusting the system's suggestion algorithm based on feedback data.

[1158] Output: Optimized system settings

[1159] This makes it possible to continue to provide optimal suggestions to users at all times.

[1160] (Application example 2)

[1161] 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."

[1162] In modern society, users are surrounded by a vast amount of information, making it difficult to properly manage their health. In particular, the lack of personalized advice that takes into account each user's daily life and health condition makes it difficult to effectively maintain health. Furthermore, existing health management systems have difficulty providing advice that fully reflects the user's emotions and preferences.

[1163] The identification processing 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 life log data and emotional data, means for preprocessing the collected data, and means for analyzing the preprocessed data and generating personalized health advice based on the behavioral patterns, preferences, and emotional data. This enables highly accurate health advice that takes into account the user's individual health data and emotional state.

[1164] "Life log data" is digital data that records the actions and activities of a user in their daily life.

[1165] "Emotional data" is data that reflects the user's emotional state, and is information collected using techniques such as facial recognition, voice analysis, and input text analysis.

[1166] "Preprocessing" refers to the process of cleansing, anonymizing, and formatting collected data to make it easier to analyze.

[1167] "Behavioral patterns" are data extracted from repetitive actions and habits in a user's daily life.

[1168] "Preferences" are data that refer to the preferences and tendencies that a user shows toward specific things.

[1169] "Personalized health advice" refers to health management suggestions that are individually provided based on a user's behavioral patterns, preferences, and emotional data.

[1170] "Feedback" is data that indicates the actions taken by the user in response to the advice provided and their impressions.

[1171] "Data analysis" is the process of automatically extracting and analyzing user behavioral patterns and preferences based on collected and preprocessed data.

[1172] This invention is a system that solves the problem of information overload and improves the quality of a user's life by providing personalized suggestions that take into account the user's daily activities, preferences, and emotions. The implementation of this system consists of the following elements:

[1173] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc., and uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects them as emotional data. This data is saved in a specific format (e.g., JSON).

[1174] The collected lifelog data and emotion data are preprocessed on the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1175] The preprocessed data is encrypted on the device and sent to the server using a secure communication protocol (e.g., HTTPS). The server decrypts the data received from the device and distributes it to various systems. The data is then analyzed using a large-scale language model (LLM). This analysis process automatically extracts user behavioral patterns, preferences, and emotions.

[1176] Based on the analysis results, the server generates personalized health advice. For example, it provides specific advice on the appropriate amount of exercise, diet, and sleep time based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is presented to the user on the device in a visually easy-to-understand format.

[1177] The user takes action on the advice they receive and provides feedback. This allows the user to collect the actions they take in response to the advice (e.g., whether they accepted or rejected the advice) and their subsequent thoughts (e.g., whether the advice was helpful). The collected feedback data is formatted on the device and then sent back to the server.

[1178] The server uses the received feedback to refine the large-scale language model and the entire system, improving the accuracy of the next proposal, creating a cycle of continuous improvement in the system's proposal accuracy and user satisfaction.

[1179] As a specific example, after a user finishes their morning walk, data such as the number of steps, heart rate, dietary content, and sleep time are recorded on the smartphone, and the app generates suggestions based on this data and displays them to the user using the example prompt, "Please provide me with health management advice based on my daily health data." In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them make fun and new choices without worrying about information overload.

[1180] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1181] Step 1:

[1182] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Inputs include location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (e.g., JSON format).

[1183] Step 2:

[1184] The device preprocesses the collected lifelog data and emotion data. The data collected in step 1 is used as input. Specific operations include cleansing the data, for example, rounding location information to a specific radius and removing personally identifiable information (names and contact details). The device also converts the data into a unified format. The output is the cleansed, anonymized data.

[1185] Step 3:

[1186] The terminal encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS). As input, the preprocessed data from step 2 is used, and the output is the encrypted data sent to the server.

[1187] Step 4:

[1188] The server decrypts the encrypted data received from the terminal. As input, it uses the encrypted data sent in step 3. After decryption, the data is distributed to various systems within the server. The output is the decrypted data.

[1189] Step 5:

[1190] The server analyzes the data using a large-scale language model (LLM). The distributed decoded data is used as input. This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. Specifically, it analyzes information such as the user's tendency to frequently visit a particular cafe on weekdays or their preference for a particular music genre. The analysis results are then output.

[1191] Step 6:

[1192] The server generates personalized health advice based on the analysis results. The analysis results obtained in step 5 are used as input. Specific operations include providing specific advice on appropriate amounts of exercise, dietary content, sleep duration, etc. based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is output.

[1193] Step 7:

[1194] The terminal receives the advice from the server and presents it to the user in a visually easy-to-understand format. The health advice generated in step 6 is used as input. Specifically, the advice is presented to the user in an easy-to-read format, such as a card. The output is the advice visually presented to the user.

[1195] Step 8:

[1196] The user takes action on the presented advice and provides feedback. The advice presented in step 7 is used as input. Specific actions include inputting into the terminal whether the advice was acted upon or rejected, and subsequent impressions (e.g., whether the advice was helpful). The output is the collected feedback data.

[1197] Step 9:

[1198] The terminal sends the collected feedback data to the server. The feedback data collected in step 8 is used as input. As a specific operation, the feedback data is reformatted and sent to the server again. The output is the feedback data sent to the server.

[1199] Step 10:

[1200] The server uses the received feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. The feedback data received in step 9 is used as input. Specific operations include analyzing the feedback data and adjusting system settings and model parameters. The output is a refined system and improved accuracy of the next proposal.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] [Fourth embodiment]

[1205] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1206] 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.

[1207] 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).

[1208] 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.

[1209] 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.

[1210] 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).

[1211] 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.

[1212] 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.

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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.

[1217] 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."

[1218] The system of the present invention collects and analyzes users' daily activities and preferences, and provides personalized suggestions to resolve the problem of information overload and improve the quality of their lives. This system is mainly composed of the following elements:

[1219] 1. Collection of life log data

[1220] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1221] 2. Data Preprocessing

[1222] The data collected by the device is cleansed and anonymized to prevent personal information from being identified. Specifically, location information is rounded to within a specific range and personal information such as names is removed. The data is also converted to a specified format (e.g., JSON), encrypted, and securely sent to the server.

[1223] 3. Data Analysis

[1224] The server decodes the received data and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM) to extract user behavior patterns and preferences with high accuracy. This analysis utilizes data mining and machine learning technologies.

[1225] 4. Generating personalized suggestions

[1226] The server generates optimal suggestions for users based on their behavioral patterns and preferences. For example, for a user who enjoys visiting art museums, it generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. It also ensures diversity in the suggestions by adding information about other events that may pique a user's interest.

[1227] 5. Receiving and Displaying Proposals

[1228] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[1229] 6. Collecting User Feedback

[1230] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[1231] 7. Repeat the cycle

[1232] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[1233] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[1234] The processing flow will be explained below.

[1235] Step 1:

[1236] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (such as JSON).

[1237] Step 2:

[1238] Preprocessing the lifelog data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1239] Step 3:

[1240] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1241] Step 4:

[1242] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[1243] Step 5:

[1244] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns and preferences are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or that they prefer a particular music genre.

[1245] Step 6:

[1246] Based on the analysis, the server generates personalized suggestions for the user, including relevant new activities and events based on the user's behavior and preferences, such as literary events or newly opened cafes.

[1247] Step 7:

[1248] The server generates a proposal and sends it to the terminal.

[1249] Step 8:

[1250] The device receives the suggestions from the server and notifies the user through a user interface (UI). The suggestions are displayed in a visually easy-to-understand format (e.g., a card format) and provide a link to more information.

[1251] Step 9:

[1252] The user reviews the suggestions and takes action, such as attending a suggested event or visiting a suggested location.

[1253] Step 10:

[1254] The device collects user actions and feedback, such as "I accepted the suggestion" or "The suggested location was crowded" and other information.

[1255] Step 11:

[1256] The feedback collected by the device is encrypted and sent to a server.

[1257] Step 12:

[1258] The server analyzes the feedback and refines the system's large-scale language model (LLM) and suggestion generation algorithms, improving the accuracy of suggestions based on user preferences and behavior.

[1259] Step 13:

[1260] The server and device continuously repeat this cycle to optimize the user experience.

[1261] Example 1

[1262] 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."

[1263] In modern society, users are overwhelmed with information, making it difficult to effectively obtain the information they need and make efficient decisions in their daily lives. In particular, information overload makes it difficult to find the best option for them, resulting in a decline in quality of life.

[1264] 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.

[1265] In this invention, the server includes means for collecting life log data related to the user's daily activities and preferences, means for cleansing the collected life log data to anonymize personal information, converting it into a specified data format, and then encrypting it, and means for decrypting the data and analyzing the behavioral patterns and preferences with high accuracy using a large-scale language model, thereby making it possible to provide personalized suggestions to the user.

[1266] "Life log data" is data that includes information about the user's daily activities and preferences.

[1267] "Cleansing" is a process of removing noise and missing values ​​from data and normalizing the data.

[1268] "Anonymization" is the process of removing information that can identify a specific individual and de-identifying personal information.

[1269] A "specified data format" is a format (e.g., JSON) that allows data to be easily processed and exchanged.

[1270] "Encryption" is the process of encoding data using a specific algorithm to protect it from unauthorized access.

[1271] "Decryption" is the process of restoring encrypted data to its original form.

[1272] A "large-scale language model" is a machine learning model with natural language processing capabilities that is trained using massive amounts of data.

[1273] "Behavior patterns" refer to the user's daily behavioral tendencies and recurring patterns of behavior.

[1274] "Preferences" are data that indicate specific tendencies and interests that a user has.

[1275] "Personalized suggestions" refer to information and suggestions that are customized based on a user's individual behavioral patterns and preferences.

[1276] "Visually easy to understand" means using a user interface to present the suggestions in a format that is easy for the user to understand.

[1277] "Feedback" refers to the actions taken by the user in response to the proposal, as well as their subsequent impressions and evaluations.

[1278] "Used for cyclical improvement" means continually optimizing the system's algorithms and models based on collected feedback.

[1279] MODE FOR CARRYING OUT THE INVENTION

[1280] The system of the present invention collects and analyzes a user's daily activities and preferences to provide personalized suggestions, thereby eliminating the problem of information overload and improving the user's quality of life. This system is implemented mainly using the following hardware and software.

[1281] System Configuration

[1282] The system consists of the following components:

[1283] 1. Device: User's smartphone or tablet

[1284] 2. Server: Cloud-based data analysis system

[1285] 3. Software: Smartphone sensor APIs, large-scale language models (LLMs), machine learning frameworks (e.g., PyTorch, TensorFlow)

[1286] Specific processing

[1287] The device collects life log data from various sensors and apps on the user's smartphone. Specifically, it obtains location information using GPS, usage history from app usage history APIs, music playback history from music app APIs, post content from SNS APIs, and schedule information from calendar APIs.

[1288] The collected data is cleansed on the device and processed to anonymize personal information. For example, location information is rounded to a specific precision and fields such as name and phone number are removed. The data is then converted to a specified format (e.g., JSON) and encrypted using an encryption algorithm (e.g., AES). The encrypted data is then sent to a server using a secure communication protocol (e.g., HTTPS).

[1289] The server decrypts the received data and inputs it into a machine learning model. Specifically, the data is read using a Python data processing library (e.g., Pandas), and large-scale language models (LLMs) and data mining techniques are used to analyze user behavior patterns and preferences with high accuracy.

[1290] Based on the analysis results, the server generates personalized suggestions for the user. For example, for a user who likes to visit art museums, it provides information on new exhibitions and nearby cafes. This ensures the diversity and accuracy of suggestions and efficiently provides information that catches the user's interest.

[1291] The generated suggestions are sent to the device and displayed visually to the user through a user interface (UI), often in the form of cards that allow easy access to detailed information.

[1292] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and subsequent feedback (e.g., whether they enjoyed the event), and then sends the formatted data back to the server.

[1293] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next suggestions.Continual optimization based on user feedback leads to sustained improvements in the system's suggestion accuracy and user satisfaction.

[1294] Prompt Sentence Examples

[1295] Examples of prompts include:

[1296] "Consider that the user has a habit of jogging on Saturdays, suggest jogging events they can join on sunny Saturdays, and nearby healthy restaurants."

[1297] In this way, the system of the present invention improves the quality of a user's life by providing personalized suggestions based on the user's daily activities and preferences.

[1298] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1299] Step 1: Collecting Lifelogging Data

[1300] The device collects life log data from various sensors and apps on the smartphone. The device obtains location information from the GPS sensor and app usage history from the app usage history API. It also obtains music playback history from the music app API and schedule information from the calendar API. This data is saved along with time information and used as input data.

[1301] Specific behavior:

[1302] Input: GPS sensor location information, app usage history API data, music playback history API data, calendar API data.

[1303] What it does: The device periodically reads location information from the GPS sensor, saves the current location data, and simultaneously sends a request to the music app's play history API to retrieve a list of recently played songs.

[1304] Output: Life log data such as location information, app usage history, music playback history, and calendar information.

[1305] Step 2: Preprocessing the data

[1306] The lifelog data collected by the device is cleansed and personal information is anonymized. For example, location information is rounded to a specific accuracy and fields such as name and phone number are removed. The data is then converted to JSON format and encrypted using the AES encryption algorithm. The encrypted data is then sent to a server using a secure communication protocol (HTTPS).

[1307] Specific behavior:

[1308] Input: Collected life log data (location information, app usage history, music playback history, calendar information).

[1309] What it does: It rounds the longitude and latitude of the location data to two decimal places, replaces all name and phone number fields with blanks or "", then converts the data to JSON format and encrypts it with the AES encryption algorithm.

[1310] Output: Encrypted data in JSON format.

[1311] Step 3: Analyze the data

[1312] The server decrypts the encrypted data it receives and distributes it to various systems. Next, the received data is input into a machine learning model (e.g., a large-scale language model trained using PyTorch) to analyze user behavior patterns and preferences. This analysis involves data mining techniques.

[1313] Specific behavior:

[1314] Input: Encrypted JSON formatted data.

[1315] How it works: The server decrypts the AES-encrypted data, reads the obtained JSON data using a Python data processing library (e.g., Pandas), and inputs the data into a machine learning model. After that, data mining techniques are used to analyze user behavior patterns and preferences.

[1316] Output: Analysis results of user behavior patterns and preferences.

[1317] Step 4: Generate personalized suggestions

[1318] The server generates optimal suggestions for the user based on the analysis results. These suggestions include new events and suitable spots. For example, for a user who likes to visit art museums, the server can provide information on special exhibitions at new museums and nearby cafes.

[1319] Specific behavior:

[1320] Input: Analysis results of user behavior patterns and preferences.

[1321] Specific operation: Based on the results of behavioral pattern analysis, queries are sent to event databases that are likely to be of interest to the user (e.g., museum APIs, event calendar APIs) to obtain appropriate event information.

[1322] Output: Personalized suggestions for the user.

[1323] Step 5: Receive and view proposals

[1324] The device receives the proposal data from the server and displays it on the user interface (UI). The proposals are presented to the user in a visually easy-to-understand format, such as a card, allowing for easy access to detailed information.

[1325] Specific behavior:

[1326] Input: Personalized suggestions for the user.

[1327] Specific operation: The device analyzes the proposal data received from the server and displays it in a card-style UI within the mobile app. Tapping the card opens the event details page.

[1328] Output: A visually friendly suggestion to the user.

[1329] Step 6: Gather user feedback

[1330] The device collects the user's actions and feedback regarding the proposal, such as whether the proposal was accepted or rejected, and subsequent impressions and evaluations, and sends the collected data to the server.

[1331] Specific behavior:

[1332] Input: User actions and feedback on suggestions.

[1333] Specific behavior: Records the user's reaction when clicking on an event suggestion and accepts feedback using rating buttons (e.g., star ratings). Feedback is saved in JSON format, AES encrypted, and sent to the server.

[1334] Output: Feedback data.

[1335] Step 7: Repeat the cycle

[1336] The server uses the collected feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. Based on the feedback, the system's algorithms and models are continuously optimized.

[1337] Specific behavior:

[1338] Input: Feedback data.

[1339] Specific operation: The feedback data is input back into the machine learning model and the model weights are updated, allowing the model to learn from the new data and improve the accuracy of the next prediction.

[1340] Output: Optimized large-scale language models and systems.

[1341] (Application example 1)

[1342] 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."

[1343] Conventional autonomous vehicle systems are inconvenient for users because they cannot provide individually optimized route guidance based on the user's life log data and preferences. Therefore, to improve the user experience, it is necessary to provide individually optimized route guidance through the collection and analysis of life log data.

[1344] 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.

[1345] In this invention, the server includes means for collecting life log data, means for preprocessing the collected life log data, means for analyzing the preprocessed data and extracting behavioral patterns and preferences, means for generating optimal route guidance based on the extracted behavioral patterns and preferences, and means for providing the optimal route guidance to an autonomous vehicle, thereby enabling the provision of individually optimized route guidance based on the user's behavioral patterns and preferences.

[1346] "Life log data" refers to information about a user's daily activities and preferences.

[1347] "Pre-processing" refers to the process of anonymizing and cleansing collected data.

[1348] "Behavioral patterns" refer to regularities regarding a user's specific behavior and its frequency, time, location, etc.

[1349] "Preferences" refer to the preferences or interests a user has for particular activities, places, or things.

[1350] "Personalized suggestions" refers to suggestions that are specifically tailored based on a user's behavioral patterns and preferences.

[1351] "Optimal route guidance" refers to the most appropriate route and destination for the user, generated based on the user's life log data and preferences.

[1352] "Feedback" refers to the reactions and opinions of users regarding suggestions they receive.

[1353] An "autonomous vehicle" refers to a vehicle that operates autonomously without the need for human operation.

[1354] The system of this invention collects and analyzes the user's daily activities and preferences, and provides optimal route guidance to an autonomous vehicle. To implement the invention, the following hardware and software are used for processing.

[1355] Hardware and software used

[1356] 1. Smartphone: A device that collects life log data from users (location information, app usage history, music history, etc.).

[1357] 2. Server: A device that preprocesses and analyzes lifelog data, generates personalized suggestions, and generates optimal route guidance.

[1358] 3. Self-driving vehicle: A vehicle that receives the generated optimal route guidance and drives autonomously.

[1359] 4. Large-scale language models (e.g., GPT-3): Machine learning models for analyzing user behavioral patterns and preferences and generating personalized suggestions.

[1360] Data collection and preprocessing

[1361] Smartphones periodically collect life log data such as user location information, app usage history, and music history. The collected data is pre-processed by anonymization and encryption and then securely transmitted to a server.

[1362] Data analysis and proposal generation

[1363] The server analyzes the received data using a large-scale language model (LLM) to extract the user's behavioral patterns and preferences. Based on the results of this analysis, it generates individually optimized route guidance. For example, it can provide recommended routes via cafes frequently visited by the user or tourist spots for potential customers.

[1364] Providing optimal route guidance

[1365] The server generates optimal route guidance, which is then sent to the autonomous vehicle, which then drives autonomously according to the guidance. The user can visually check the proposed route guidance through a smartphone app.

[1366] Gathering feedback and optimizing the system

[1367] The smartphone collects user feedback, which is then sent to the server, which then uses the feedback to optimize the entire system and improve the accuracy of the next suggestions.

[1368] Examples of concrete examples and prompts

[1369] Example 1:

[1370] Based on the user's behavioral data when commuting each morning, the system suggests a route that avoids traffic jams and allows them to stop at their favorite cafe.

[1371] Example 2:

[1372] While the user is visiting a tourist spot, the system suggests new tourist spots and restaurants based on past visit data.

[1373] Example prompt sentence:

[1374] Based on the user's behavioral data, generate optimal route guidance based on the following information:

[1375] Current location information (latitude: 35.6895, longitude: 139.6917)

[1376] Past visit history (e.g. cafes, restaurants, tourist spots)

[1377] Commuting route history

[1378] Example 1:

[1379] Destination:Office

[1380] Commuting time: 8:00-9:00 AM

[1381] Favorite stop: Cafe

[1382] Example 2:

[1383] Tourist attractions: famous places

[1384] Favorite restaurant: Restaurant

[1385] Past visit history: tourist attractions, shopping malls

[1386] Each suggestion should be returned in JSON format.

[1387] As described above, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without worrying about information overload.

[1388] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1389] Step 1: Collecting Lifelogging Data

[1390] Smartphone devices periodically collect life log data, such as location information, app usage history, and music history, from users' daily lives. Raw data obtained from smartphone sensors and apps is used as input. This data is collected using location information services, app monitoring services, music playback history services, etc.

[1391] Step 2: Preprocessing the data

[1392] The collected lifelog data is anonymized and cleansed on the smartphone device. This includes rounding location data to a specific range and removing personally identifiable information. The cleansed data is converted into a specified format (e.g., JSON) and securely sent to a server. The input is raw lifelog data, and the output is anonymized and formatted data.

[1393] Step 3: Analyze the data

[1394] The received data is decoded on the server side and analyzed using a large-scale language model (LLM). Machine learning algorithms and data mining techniques are used for the analysis to extract user behavioral patterns and preferences. The input for the analysis is preprocessed data, and the output is information indicating the user's behavioral patterns and preferences.

[1395] Step 4: Generate optimal route directions

[1396] The server generates optimal route guidance for the user based on the extracted behavioral patterns and preferences. This generation utilizes a large-scale language model and takes into account the user's past behavioral data and preferences. The input is the behavioral patterns and preferences resulting from the analysis, and the output is personalized route guidance.

[1397] Step 5: Provide and display directions

[1398] The route guidance generated by the server is sent to the autonomous vehicle. The autonomous vehicle then drives autonomously according to the received route guidance. The user can visually check the proposed route guidance via a smartphone app. The input is the generated route guidance, and the output is the vehicle's driving instructions and the display on the app.

[1399] Step 6: Gather user feedback

[1400] The user provides feedback on the proposed route guidance through a smartphone app. For example, they send information such as whether the route was comfortable or if there are areas for improvement. The input is the user's feedback, and the output is the feedback data. The feedback data is sent to the server and used to improve the accuracy of the next proposal.

[1401] Step 7: Optimize your system

[1402] The server analyzes the collected feedback data and refines the generative AI model and the entire system, improving the accuracy of the next proposal. The input is the feedback data, and the output is an improved system model and proposal accuracy.

[1403] 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.

[1404] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[1405] 1. Collection of life log data

[1406] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1407] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the data as emotion data. This data is also saved in a specific format (e.g., JSON).

[1408] 2. Data Preprocessing

[1409] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (names and contact details) is removed, and the collected data is converted into a unified format.

[1410] 3. Data transmission

[1411] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1412] 4. Data Analysis

[1413] The server decodes the data received from the device and distributes it to various systems. Next, the data is analyzed using a large-scale language model (LLM). This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[1414] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[1415] 5. Generating personalized suggestions

[1416] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at the museum and a popular cafe nearby. The server also adds information about other events that may pique a user's interest, ensuring diversity in the suggestions.

[1417] 6. Receiving and Displaying Proposals

[1418] The device receives suggestions from the server and displays them visually to the user in an easy-to-understand manner. In the user interface (UI), the suggestions are presented in a card format, allowing easy access to detailed information.

[1419] 7. Collecting User Feedback

[1420] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[1421] 8. Repeat the cycle

[1422] The server uses the feedback it receives to refine the large-scale language model (LLM) and the entire system, improving the accuracy of the next proposal. By constantly iterating optimization based on user-supplied data, the system's proposal accuracy and user satisfaction are continuously improved.

[1423] As a specific example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time suggestions are made, it will be more likely to recommend a different cafe or new menu that makes the user feel happy. In this way, the system of the present invention supports the user's daily life in various ways, helping them make fun and new choices without worrying about information overload.

[1424] The processing flow will be explained below.

[1425] Step 1:

[1426] The device collects life log data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1427] Step 2:

[1428] The device uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects the emotion data. This emotion data is also saved in a specific format (e.g., JSON).

[1429] Step 3:

[1430] Preprocessing of life log data and emotion data collected by the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1431] Step 4:

[1432] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1433] Step 5:

[1434] The server decodes the data received from the device and feeds the decoded data into a large-scale language model (LLM) and a data analysis system.

[1435] Step 6:

[1436] The server analyzes the received data using a large-scale language model (LLM). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis results may include "a user tends to frequently visit a particular cafe on weekdays" or "a user prefers a particular music genre."

[1437] Step 7:

[1438] Based on the analysis results, the server generates personalized suggestions suited to the user. These suggestions include new activities and events that are highly relevant based on the user's behavior and preferences. For example, it generates "information about special exhibitions for a person who likes to visit art museums" and "suggestions to visit popular cafes nearby." It also adds information about other events that may pique the user's interest, ensuring diversity in the suggestions.

[1439] Step 8:

[1440] The device receives the suggestion data from the server, which includes detailed information related to the user's behavior, preferences, and emotions.

[1441] Step 9:

[1442] The device will display the suggestions in a user interface (UI) in a visually understandable format, such as a card, allowing the user to easily access the details.

[1443] Step 10:

[1444] The user takes action on the suggested activity or event, for example, attending an event based on the suggestion or visiting a suggested location.

[1445] Step 11:

[1446] The device collects user actions and feedback, such as "I accepted the suggestion," "The suggested location was crowded," and "I enjoyed the event."

[1447] Step 12:

[1448] The device formats the collected feedback, encrypts it, and sends it to a server.

[1449] Step 13:

[1450] The server analyzes the feedback data and refines the system's large-scale language model (LLM) and proposal generation algorithm, which improves the accuracy of the next proposal.

[1451] Step 14:

[1452] The server and device continuously repeat this process to optimize the user experience, and the data acquired at each step is constantly fed back into the model, continuously improving the accuracy of the system's suggestions and user satisfaction.

[1453] Example 2

[1454] 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."

[1455] In modern society, the problem of information overload is serious, making it difficult for users to find the information they truly need in their daily lives. Furthermore, there is a lack of systems that provide personalized suggestions based on users' behavior, preferences, and emotions. Furthermore, there is a need for technology development that can appropriately reflect user feedback to improve the quality of suggestions.

[1456] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1457] In this invention, the server includes a means for collecting life log data, a means for preprocessing the collected life log data, and a means for analyzing the preprocessed data to extract behavioral patterns, preferences, and emotions, thereby providing users with optimized personalized suggestions, solving the problem of information overload, and improving the quality of life of users.

[1458] "Life log data" refers to a variety of data collected from a user's daily life, such as location information, app usage history, music playback history, social media posts, and calendar information.

[1459] "Preprocessing" refers to the process of cleansing collected data, anonymizing it, standardizing its format, and so on.

[1460] "Behavioral patterns" are data that indicate regular tendencies and habits in the user's daily activities.

[1461] "Preferences" is data that indicates the things a user likes and areas of interest.

[1462] "Emotion" is data that indicates the user's emotional state and is collected using techniques such as facial recognition and voice analysis.

[1463] "Personalized suggestions" are suggestions or recommendations that are individually provided based on a user's behavioral patterns, preferences, and emotions.

[1464] "Feedback" refers to data such as the actions taken by users in response to suggestions and the ratings and opinions they provide thereafter.

[1465] "Anonymization" is the process of protecting personal information by removing any information that could identify an individual from the data.

[1466] A "proposal generator" is a mechanism for creating user-specific proposals based on collected and analyzed data.

[1467] "Optimizing the whole system" is the process of adjusting the system's algorithms and parameters based on collected feedback to improve the accuracy of suggestions.

[1468] The system of the present invention solves the problem of information overload and improves the quality of life of users by providing personalized suggestions that take into account their daily activities, preferences, and emotions. This system mainly consists of the following elements:

[1469] 1. Collection of life log data

[1470] The device collects data from various sensors and apps on the user's smartphone, including location information, app usage history, music playback history, social media posts, and calendar information.

[1471] The device uses an emotion engine (e.g., Affectiva, IBM Watson) to recognize the user's emotions through facial recognition, voice analysis, input text analysis, etc., and collects the emotion data. This data is stored in JSON format.

[1472] 2. Data Preprocessing

[1473] The device preprocesses the collected lifelog data and emotion data. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1474] 3. Data transmission

[1475] The terminal encrypts the preprocessed data and transmits it to the server using a secure communication protocol (e.g., HTTPS).

[1476] 4. Data Analysis

[1477] The server decodes the data received from the device and distributes it to each system. Next, it analyzes the data using a large-scale language model (e.g., OpenAI GPT, Google BERT). During this analysis process, the user's behavioral patterns, preferences, and emotions are automatically extracted. For example, the analysis output may show that the user tends to frequently visit a particular cafe on weekdays or prefers a particular music genre.

[1478] Emotional data is also incorporated into the analysis process, taking into account detailed behavioral and preference biases based on the user's emotional state.

[1479] 5. Generating personalized suggestions

[1480] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions. For example, for a user who enjoys visiting art museums, the server generates suggestions for visiting a new special exhibition at a museum and a popular cafe nearby. It also adds information about other events that may interest the user, ensuring diversity in the suggestions.

[1481] 6. Receiving and Displaying Proposals

[1482] The device receives the suggestions from the server and displays them to the user in a visually understandable manner. The user interface (UI) presents the suggestions in a card format, allowing easy access to detailed information.

[1483] 7. Collecting User Feedback

[1484] The device collects the actions the user takes in response to the proposal (e.g., whether they accepted or rejected the proposal) and their subsequent feedback (e.g., whether the event was fun or the cafe was crowded). The collected feedback is formatted and sent to the server.

[1485] 8. Repeat the cycle

[1486] The server uses the received feedback to refine the large-scale language model (e.g., OpenAI GPT, Google BERT) and the entire system to improve the accuracy of the next recommendation. By constantly iterating optimization based on the data provided by the user, the system's recommendation accuracy and user satisfaction are continuously improved.

[1487] Specific examples

[1488] For example, if a user drinks coffee at the same cafe every morning and feels "happy" at the time, the emotion engine will recognize that emotion and record it as data. As a result, the next time it makes suggestions, it will recommend a different cafe or new menu that is more likely to make the user feel happy.

[1489] Example prompts for generative AI models

[1490] 1. Based on the user's emotions and behavioral patterns, suggest fun events for this weekend.

[1491] 2. Based on recent lifelog data, make a list of new cafes recommended for this user.

[1492] 3. Suggest new music albums based on this user's favorite music genres.

[1493] In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them to make fun and new choices without being overwhelmed by information overload.

[1494] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1495] Step 1: Collecting Lifelogging Data

[1496] The device collects life log data. The device automatically collects data from various sensors (GPS, accelerometer, gyroscope, etc.) and apps on the user's smartphone.

[1497] Input: Raw data from smartphone sensors (e.g., location information, app usage history, music playback history, social media posts, calendar information)

[1498] Processing: Acquires data from sensors and stores it in a unified format (e.g. JSON).

[1499] Output: Collected lifelog data (JSON format)

[1500] Specific operation: The device uses the GPS sensor to obtain the user's location information and saves it as location data (e.g., latitude, longitude) in JSON format.

[1501] Step 2: Collecting emotion data

[1502] The terminal activates an emotion engine and collects the user's emotion data.

[1503] Input: Data based on facial recognition, voice analysis, and input text analysis

[1504] Processing: An emotion engine analyzes the collected data, identifies the emotional state, and stores it as data (e.g., using Affectiva or IBM Watson).

[1505] Output: Emotion data (JSON format)

[1506] Specific operation: The device uses the camera to analyze the user's facial expressions and collect emotional data such as "happiness" or "sadness."

[1507] Step 3: Preprocessing the data

[1508] Preprocess the life log data and emotion data collected by the device.

[1509] Input: Collected life log data and emotion data

[1510] Processing: Cleansing, anonymizing, and formatting the data.

[1511] Data cleansing specifically: Detecting and removing incomplete data.

[1512] Specific actions of anonymization: Removing user names and contact information from social media post data.

[1513] Specific behavior of format conversion: The location data is rounded within a specific radius and converted into a unified format (e.g., JSON).

[1514] Output: Preprocessed data (JSON format)

[1515] Step 4: Sending data

[1516] The device encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS).

[1517] Input: Preprocessed data (JSON format)

[1518] Processing: Encrypting data and transmitting it using secure communications protocols.

[1519] What encryption does: Encrypts data using AES encryption.

[1520] Specific transmission behavior: Encrypted data is sent to the server using the HTTPS protocol.

[1521] Output: Encrypted data sent to the server

[1522] Step 5: Analyze the data

[1523] The server decrypts the data received from the terminal and distributes it to each system.

[1524] Input: Encrypted data

[1525] Processing: The data is decoded and analyzed using a large-scale language model (e.g., OpenAI GPT, Google BERT).

[1526] Specific operations for decrypting data: Decrypt the data using AES decryption.

[1527] Specific analysis steps: Behavioral pattern data is input into a large-scale language model to extract patterns such as "frequently visiting a specific cafe on weekdays."

[1528] Output: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[1529] Step 6: Generate personalized suggestions

[1530] The server generates optimal suggestions for users based on their behavioral patterns, preferences, and emotions.

[1531] Input: Analyzed data (analysis results of behavioral patterns, preferences, and emotions)

[1532] Processing: Generate suggestions based on behavioral patterns, preferences, and emotions.

[1533] Specific operation of suggestion generation: Based on the user's museum-loving behavioral patterns and emotional data, suggestions are generated to visit a new special exhibition at a museum and a popular cafe nearby.

[1534] Output: Personalized suggestions

[1535] Step 7: Receive and view proposals

[1536] The terminal receives the proposal from the server and displays it to the user in a visually easy-to-understand manner.

[1537] Input: Proposal sent by the server

[1538] Action: Display the suggestion in the UI.

[1539] Display behavior: The suggestion is displayed in the UI as a card, with links to more information and a map.

[1540] Output: Visualized proposal

[1541] Step 8: Gather user feedback

[1542] The terminal collects the actions the user takes in response to the suggestions and subsequent feedback.

[1543] Input: User actions and feedback data

[1544] Processing: Analyze behavioral records and collect feedback.

[1545] Specific feedback behavior: The device detects from location information whether the user has visited the suggested cafe and records that behavior.

[1546] Output: Collected feedback (in JSON format)

[1547] Step 9: Send feedback and optimize the system

[1548] The device sends feedback to the server, which then optimizes the entire system.

[1549] Input: Collected feedback data

[1550] Processing: Analyzing feedback data and adjusting algorithms and parameters across the system.

[1551] Specific behavior of feedback analysis: Adjusting the system's suggestion algorithm based on feedback data.

[1552] Output: Optimized system settings

[1553] This makes it possible to continue to provide optimal suggestions to users at all times.

[1554] (Application example 2)

[1555] 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."

[1556] In modern society, users are surrounded by a vast amount of information, making it difficult to properly manage their health. In particular, the lack of personalized advice that takes into account each user's daily life and health condition makes it difficult to effectively maintain health. Furthermore, existing health management systems have difficulty providing advice that fully reflects the user's emotions and preferences.

[1557] The identification processing 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 life log data and emotional data, means for preprocessing the collected data, and means for analyzing the preprocessed data and generating personalized health advice based on the behavioral patterns, preferences, and emotional data. This enables highly accurate health advice that takes into account the user's individual health data and emotional state.

[1558] "Life log data" is digital data that records the actions and activities of a user in their daily life.

[1559] "Emotional data" is data that reflects the user's emotional state, and is information collected using techniques such as facial recognition, voice analysis, and input text analysis.

[1560] "Preprocessing" refers to the process of cleansing, anonymizing, and formatting collected data to make it easier to analyze.

[1561] "Behavioral patterns" are data extracted from repetitive actions and habits in a user's daily life.

[1562] "Preferences" are data that refer to the preferences and tendencies that a user shows toward specific things.

[1563] "Personalized health advice" refers to health management suggestions that are individually provided based on a user's behavioral patterns, preferences, and emotional data.

[1564] "Feedback" is data that indicates the actions taken by the user in response to the advice provided and their impressions.

[1565] "Data analysis" is the process of automatically extracting and analyzing user behavioral patterns and preferences based on collected and preprocessed data.

[1566] This invention is a system that solves the problem of information overload and improves the quality of a user's life by providing personalized suggestions that take into account the user's daily activities, preferences, and emotions. The implementation of this system consists of the following elements:

[1567] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Specifically, it acquires location information, app usage history, music playback history, social media posts, calendar information, etc., and uses an emotion engine to recognize the user's emotions using techniques such as facial recognition, voice analysis, and input text analysis, and collects them as emotional data. This data is saved in a specific format (e.g., JSON).

[1568] The collected lifelog data and emotion data are preprocessed on the device. Preprocessing includes data cleansing, anonymization, and format conversion. For example, location information is rounded to within a specific radius, personally identifiable information (such as names and contact details) is removed, and the collected data is converted into a unified format.

[1569] The preprocessed data is encrypted on the device and sent to the server using a secure communication protocol (e.g., HTTPS). The server decrypts the data received from the device and distributes it to various systems. The data is then analyzed using a large-scale language model (LLM). This analysis process automatically extracts user behavioral patterns, preferences, and emotions.

[1570] Based on the analysis results, the server generates personalized health advice. For example, it provides specific advice on the appropriate amount of exercise, diet, and sleep time based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is presented to the user on the device in a visually easy-to-understand format.

[1571] The user takes action on the advice they receive and provides feedback. This allows the user to collect the actions they take in response to the advice (e.g., whether they accepted or rejected the advice) and their subsequent thoughts (e.g., whether the advice was helpful). The collected feedback data is formatted on the device and then sent back to the server.

[1572] The server uses the received feedback to refine the large-scale language model and the entire system, improving the accuracy of the next proposal, creating a cycle of continuous improvement in the system's proposal accuracy and user satisfaction.

[1573] As a specific example, after a user finishes their morning walk, data such as the number of steps, heart rate, dietary content, and sleep time are recorded on the smartphone, and the app generates suggestions based on this data and displays them to the user using the example prompt, "Please provide me with health management advice based on my daily health data." In this way, the system of the present invention provides multifaceted support for the user's daily life, helping them make fun and new choices without worrying about information overload.

[1574] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1575] Step 1:

[1576] The device collects life log data and emotional data from various sensors and apps on the user's smartphone. Inputs include location information, app usage history, music playback history, social media posts, calendar information, etc. This data is saved in a specific format (e.g., JSON format).

[1577] Step 2:

[1578] The device preprocesses the collected lifelog data and emotion data. The data collected in step 1 is used as input. Specific operations include cleansing the data, for example, rounding location information to a specific radius and removing personally identifiable information (names and contact details). The device also converts the data into a unified format. The output is the cleansed, anonymized data.

[1579] Step 3:

[1580] The terminal encrypts the preprocessed data and sends it to the server using a secure communication protocol (e.g., HTTPS). As input, the preprocessed data from step 2 is used, and the output is the encrypted data sent to the server.

[1581] Step 4:

[1582] The server decrypts the encrypted data received from the terminal. As input, it uses the encrypted data sent in step 3. After decryption, the data is distributed to various systems within the server. The output is the decrypted data.

[1583] Step 5:

[1584] The server analyzes the data using a large-scale language model (LLM). The distributed decoded data is used as input. This analysis process automatically extracts the user's behavioral patterns, preferences, and emotions. Specifically, it analyzes information such as the user's tendency to frequently visit a particular cafe on weekdays or their preference for a particular music genre. The analysis results are then output.

[1585] Step 6:

[1586] The server generates personalized health advice based on the analysis results. The analysis results obtained in step 5 are used as input. Specific operations include providing specific advice on appropriate amounts of exercise, dietary content, sleep duration, etc. based on the user's daily behavioral patterns, preferences, and emotional state. The generated advice is output.

[1587] Step 7:

[1588] The terminal receives the advice from the server and presents it to the user in a visually easy-to-understand format. The health advice generated in step 6 is used as input. Specifically, the advice is presented to the user in an easy-to-read format, such as a card. The output is the advice visually presented to the user.

[1589] Step 8:

[1590] The user takes action on the presented advice and provides feedback. The advice presented in step 7 is used as input. Specific actions include inputting into the terminal whether the advice was acted upon or rejected, and subsequent impressions (e.g., whether the advice was helpful). The output is the collected feedback data.

[1591] Step 9:

[1592] The terminal sends the collected feedback data to the server. The feedback data collected in step 8 is used as input. As a specific operation, the feedback data is reformatted and sent to the server again. The output is the feedback data sent to the server.

[1593] Step 10:

[1594] The server uses the received feedback data to refine the large-scale language model and the entire system, improving the accuracy of the next proposal. The feedback data received in step 9 is used as input. Specific operations include analyzing the feedback data and adjusting system settings and model parameters. The output is a refined system and improved accuracy of the next proposal.

[1595] 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.

[1596] 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.

[1597] 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.

[1598] 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.

[1599] 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.

[1600] 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.

[1601] 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).

[1602] 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.

[1603] 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."

[1604] 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.

[1605] 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).

[1606] 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.

[1607] 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.

[1608] 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.

[1609] 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.

[1610] 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.

[1611] 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.

[1612] 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.

[1613] 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.

[1614] 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.

[1615] 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.

[1616] The following is further disclosed regarding the above embodiment.

[1617] (Claim 1)

[1618] a means for collecting life log data;

[1619] A means for preprocessing the collected life log data;

[1620] means for analyzing the pre-processed data to extract behavioral patterns and preferences;

[1621] means for generating personalized suggestions based on the extracted behavioral patterns and preferences;

[1622] means for presenting personalized suggestions to the user;

[1623] A system including a means for collecting user feedback.

[1624] (Claim 2)

[1625] 2. The system according to claim 1, wherein the preprocessing means includes a process for anonymizing personal information.

[1626] (Claim 3)

[1627] 2. The system according to claim 1, wherein the suggestion generating means includes a process for incorporating additional resource data to enrich the user's interests and preferences in a multifaceted manner.

[1628] "Example 1"

[1629] (Claim 1)

[1630] A means for collecting life log data relating to the user's daily activities and preferences;

[1631] A method for cleansing the collected life log data, anonymizing personal information, converting it into a specified data format, and then encrypting it;

[1632] means for decoding the data and analyzing behavioral patterns and preferences with high accuracy using large-scale language models;

[1633] means for generating personalized suggestions based on the extracted behavioral patterns and preferences;

[1634] a means for visually displaying the generated proposal to a user in an easy-to-understand manner;

[1635] A system that includes a means of collecting feedback on user suggestions and using it to make cyclical improvements.

[1636] (Claim 2)

[1637] 2. The system of claim 1, further comprising a process for anonymizing personal information.

[1638] (Claim 3)

[1639] 10. The system of claim 1, further comprising a process for incorporating additional resource data to enrich user interests and preferences in a multifaceted manner.

[1640] "Application Example 1"

[1641] (Claim 1)

[1642] a means for collecting life log data;

[1643] A means for preprocessing the collected life log data;

[1644] means for analyzing the pre-processed data to extract behavioral patterns and preferences;

[1645] means for generating personalized suggestions based on the extracted behavioral patterns and preferences;

[1646] means for presenting personalized suggestions to the user;

[1647] a means for collecting user feedback;

[1648] means for generating optimal route guidance based on the behavioral patterns and preferences;

[1649] means for providing the optimal route guidance to an autonomous vehicle.

[1650] (Claim 2)

[1651] 2. The system according to claim 1, wherein the preprocessing means includes a process for anonymizing personal information.

[1652] (Claim 3)

[1653] 2. The system according to claim 1, wherein the suggestion generating means includes a process for incorporating additional resource data to enrich the user's interests and preferences in a multifaceted manner.

[1654] "Example 2: Combining Emotion Engines"

[1655] (Claim 1)

[1656] a means for collecting life log data;

[1657] A means for preprocessing the collected life log data;

[1658] means for analyzing the pre-processed data to extract behavioral patterns, preferences, and sentiments;

[1659] means for generating personalized recommendations based on the extracted behavioral patterns, preferences, and emotions;

[1660] means for presenting personalized suggestions to the user;

[1661] a means for collecting user feedback;

[1662] A system that includes a means for using collected feedback to optimize the entire system.

[1663] (Claim 2)

[1664] 2. The system according to claim 1, wherein the preprocessing means includes a process for anonymizing personal information.

[1665] (Claim 3)

[1666] 2. The system according to claim 1, wherein the suggestion generating means includes a process for incorporating additional resource data to enrich the user's interests and preferences in a multifaceted manner.

[1667] "Application example 2 when combining emotion engines"

[1668] (Claim 1)

[1669] a means for collecting life log data;

[1670] means for preprocessing the collected life log data and emotion data;

[1671] means for analyzing the pre-processed data to extract behavioral patterns and preferences;

[1672] means for generating personalized health advice based on the extracted behavioral patterns, preferences, and emotion data;

[1673] means for presenting personalized health advice to a user;

[1674] A system including a means for collecting user feedback.

[1675] (Claim 2)

[1676] 2. The system according to claim 1, wherein the preprocessing means includes a process for anonymizing personal information.

[1677] (Claim 3)

[1678] 2. The system according to claim 1, wherein the suggestion generating means includes a process for incorporating additional resource data to enrich the user's interests and preferences in a multifaceted manner. [Explanation of symbols]

[1679] 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. a means for collecting life log data; A means for preprocessing the collected life log data; means for analyzing the pre-processed data to extract behavioral patterns and preferences; means for generating personalized suggestions based on the extracted behavioral patterns and preferences; means for presenting personalized suggestions to the user; A system including a means for collecting user feedback.

2. 2. The system according to claim 1, wherein the preprocessing means includes a process for anonymizing personal information.

3. 2. The system of claim 1, wherein the suggestion generating means includes a process for incorporating additional resource data to enrich the user's interests and preferences in a multifaceted manner.

Images