System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting an operation log of a user; means for analyzing the collected operation log to identify a behavior pattern of the user; means for predicting a next behavior of the user based on the identified behavior pattern; means for prefetching a resource of an application corresponding to the next behavior of the user; means for delivering the prefetched resource to a user terminal; and means for caching the delivered resource in the terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In conventional applications, performance optimization based on user behavior often requires developers to manually identify and improve individual performance bottlenecks. This approach limits the scope of improvement, limiting the extent to which the overall performance of an application can be improved. Furthermore, performance varies significantly depending on the user's operating environment and characteristics, making it unreasonable to apply the same improvement measures to all users, making it difficult to consistently improve the user experience. The objective of this invention is to solve these problems and provide a system that enables real-time performance optimization based on user behavior. [Means for solving the problem]

[0005] The present invention provides a means for collecting and analyzing user operation logs to identify user behavior patterns. It also includes a means for predicting the user's next action based on the identified behavior pattern and pre-fetching application resources corresponding to the predicted action. It also provides a means for delivering the pre-fetched resources to the user's terminal and caching them on the terminal. This enables immediate response to the user's next action and optimizes application performance. Furthermore, by including a function for determining optimization methods according to the user's network conditions, optimization suited to each individual user is achieved. These configurations can improve the user experience and achieve overall performance improvements.

[0006] A "user operation log" is data that records a series of operations such as clicking, scrolling, and input when a user operates an application.

[0007] "Analysis" refers to the process of finding features and patterns in collected data and identifying regularities and trends.

[0008] "Behavioral patterns" refers to a collection of data that indicates the user's past operations, behavioral tendencies, and habits.

[0009] A "behavioral characteristic model" is a mathematical model that predicts future behavior based on a user's behavioral patterns.

[0010] "Behavior prediction" is the process of using a behavioral characteristic model to predict the operations or actions that a user is likely to perform next.

[0011] "Application resources" refers to the information and functions provided by the application, specifically data such as articles, images, and product information.

[0012] "Pre-reading" refers to retrieving data that is likely to be needed next by the user in advance.

[0013] "Delivery" means that the server sends data and instructions to the user terminal.

[0014] "Caching" refers to temporarily storing data on a device for quick later access.

[0015] "Network Status" is information that indicates the speed and stability of the Internet connection to which the user's device is connected.

[0016] "Optimization" refers to making adjustments or improvements to maximize the performance of a system or application. [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 present invention relates to a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience.

[0039] Program processing overview

[0040] This system works in cooperation with a server and a terminal. When a user operates an application, the operation log is collected and sent to the server. The server analyzes the collected operation log, identifies behavioral patterns, and predicts the next action based on the identified behavioral patterns. Based on the prediction results, the server distributes the next required resource to the terminal, and the terminal stores it in its cache.

[0041] 1. Collecting and sending operation logs

[0042] The device monitors user operations (clicks, scrolls, inputs, etc.) in real time and records the operation logs internally. The operation logs are sent to the server periodically or when certain conditions are met.

[0043] 2. Analyzing data and identifying behavioral patterns

[0044] The server analyzes the received operation logs to identify user behavior patterns. This analysis uses machine learning algorithms to extract trends, such as the time of day when users tend to use certain functions.

[0045] 3. Behavioral prediction and resource preemption

[0046] The server predicts the next action a user is likely to take based on their behavioral patterns. For example, in the case of a news app, it predicts that a user who frequently views articles in a particular category is likely to click on the next article in that category. The server then pre-fetches the next required resources (articles, images, product information, etc.) and delivers them to the device.

[0047] 4. Resource Delivery and Caching

[0048] The server instructs the device to deliver the predicted resource, which the device stores in a local cache and displays immediately when the user actually needs it.

[0049] Specific examples

[0050] Example 1: News app

[0051] A user opens a news app and browses to a specific category (e.g., sports).

[0052] The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[0053] The server analyzes the received operation log and identifies that the user tends to frequently view articles in the sports category.

[0054] Based on this behavioral pattern, the server predicts that the user is likely to view an article in the sports category next.

[0055] The server pre-reads the latest sports articles and delivers them to the device, which then stores the articles in its cache.

[0056] When a user clicks on an article in the sports category, the device immediately displays the cached article.

[0057] Example 2: Shopping app

[0058] A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[0059] The terminal records these operation logs and periodically sends them to the server.

[0060] The server analyzes the data and determines that the user frequently views products in the cosmetics category.

[0061] The server predicts the user's next action and pre-fetches the next required product data.

[0062] The server delivers pre-fetched product data to the terminal, which then stores the data in its cache.

[0063] When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[0064] In this way, the present invention realizes a system that improves user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and distributing resources, and caching.

[0065] The processing flow will be explained below.

[0066] Step 1:

[0067] The user interacts with the application: within the news app, the user scrolls and clicks to browse different articles or navigate to a particular category.

[0068] Step 2:

[0069] The device logs all user actions (clicks, scrolls, typing, etc.), including timestamps and specific actions.

[0070] Step 3:

[0071] When a device has been in operation for a certain period of time or when certain conditions are met, the collected operation logs are sent to a server, and security measures such as encryption are taken when sending the data.

[0072] Step 4:

[0073] The server stores the received operation logs in a database. The stored data includes user identification information, operation details, timestamps, etc.

[0074] Step 5:

[0075] The server analyzes the operation logs using machine learning algorithms at regular intervals or when new data arrives, which involves extracting user behavior patterns, such as how often a user visits articles in a particular category at a particular time of day.

[0076] Step 6:

[0077] The server updates the user's behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and is used to predict the user's next actions.

[0078] Step 7:

[0079] The server uses the behavioral trait model to predict the next action the user is likely to take, for example determining that the user is likely to view an article in the sports category next.

[0080] Step 8:

[0081] The server determines the next required resource (article, image, product information, etc.) based on the predicted behavior. If the prediction is highly accurate, it issues a command to prefetch that resource.

[0082] Step 9:

[0083] The server delivers prefetched resource data to the device, including specific article content and image files that will be needed next.

[0084] Step 10:

[0085] The terminal caches the received resource data, allowing the resource to be accessed quickly without network delays.

[0086] Step 11:

[0087] A user actually takes the following action: for example, clicks on an article in the sports category.

[0088] Step 12:

[0089] The device will instantly display the cached resources, resulting in a smoother and faster operating experience for users.

[0090] Step 13:

[0091] The server receives the user's real-time operation log again and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[0092] Through the above processing steps, this system predicts user behavior in real time and pre-fetches and caches necessary resources, significantly improving the user experience.

[0093] Example 1

[0094] 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."

[0095] Conventional systems simply collect user operation logs, but lack the ability to predict the next operation or prefetch resources based on the predicted operation. As a result, the next resource the user needs is not provided immediately, resulting in a poor user experience. Furthermore, the lack of consideration for network conditions and real-time environmental data makes efficient data transfer and cache management difficult.

[0096] 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.

[0097] In this invention, the server includes means for transmitting a user's operation log to the server, means for analyzing the received operation log to identify the user's behavior pattern, and means for predicting the user's next behavior based on the identified behavior pattern. This makes it possible to analyze the user's operation log in real time, predict the next required resources, and deliver the predicted resources to the terminal.

[0098] "User" refers to any individual or corporation that uses this system.

[0099] An "operation log" is a record of operations such as clicking, scrolling, and input that occur when a user operates an application.

[0100] "Server" refers to a computer system that analyzes collected operation logs, identifies behavioral patterns, and distributes resources.

[0101] "Terminal" refers to a device operated by a user, and is hardware that collects operation logs and receives and displays resources.

[0102] "Behavioral patterns" refer to behavioral tendencies or characteristics identified based on a series of actions a user has taken in the past.

[0103] "Behavior prediction" refers to estimating the operations a user is likely to perform in the future based on identified behavioral patterns.

[0104] "Resources" refers to data or information that a user might need next (e.g., news articles, product information, images, etc.).

[0105] "Looking ahead" means predicting and preparing the resources a user will need next.

[0106] "Delivery" refers to sending prefetched resources from the server to the terminal.

[0107] "Cache" is a mechanism that temporarily stores distributed resources inside the device and provides them quickly when the user needs them.

[0108] The present invention is a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience. This system operates in cooperation with a server and terminals.

[0109] When a user operates an application, an operation log is collected and temporarily saved on the device. For example, it is common for Android and iOS devices to save operation logs in an SQLite database. The device sends the operation log to the server based on certain conditions. Specifically, it sends a POST request to the server when the log size exceeds 1MB or every certain time (e.g., every hour).

[0110] The server analyzes the received operation logs and identifies user behavior patterns. This analysis uses machine learning algorithms such as Python's scikit-learn and TensorFlow. Libraries such as Pandas and NumPy are used for data processing. Specifically, the server converts the received operation logs into a Pandas data frame and fits it to a machine learning model to identify user behavior patterns.

[0111] Once a behavioral pattern is identified, the server predicts the next action the user is likely to take. For example, in a news app, if it determines that a user tends to frequently view articles in a particular category (e.g., sports), it predicts that the user is likely to click on an article in that category next.

[0112] The server then prefetches the resources needed based on the predicted behavior. These resources can include news articles, images, product information, etc. The server queries the database to retrieve the next required resource and prepares it. Specifically, the server executes a query to retrieve the latest news article and related image data.

[0113] The server delivers the prepared resources to the device. The device stores the received resources in a local cache. For caching, internal storage is used on Android devices, and NSCache is used on iOS devices. When the user actually needs the resource, the device immediately displays the cached resource.

[0114] Specific examples

[0115] Specific examples of news apps

[0116] 1. A user opens a news app and browses to a specific category (e.g., sports).

[0117] 2. The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[0118] 3. The server reads the received operation log as a Pandas data frame and analyzes the data.

[0119] 4. Based on the analysis results, the server identifies a tendency for users to frequently view articles in the sports category.

[0120] 5. The server predicts the user's next action and determines that they are likely to view articles in the sports category.

[0121] 6. The server prefetches the next sports category article and image it needs and prepares it.

[0122] 7. The server delivers the prepared article and image data to the terminal.

[0123] 8. The device stores the received articles and images in a local cache.

[0124] 9. When a user clicks on an article in the Sports category, the device immediately displays the cached article.

[0125] Example prompts for specific generative AI models:

[0126] "We provide an operation log showing users frequently viewing news articles in the sports category. Please predict their next action and deliver the necessary resources to the device in advance."

[0127] Example of a shopping app

[0128] 1. A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[0129] 2. The terminal records these operation logs and periodically sends them to the server.

[0130] 3. Analyze the operation log data received by the server using a Pandas data frame.

[0131] 4. The server clusters user behavior patterns and identifies a tendency to frequently browse products in the cosmetics category.

[0132] 5. The server uses the predictive model to predict which cosmetic product is likely to be viewed next.

[0133] 6. The server queries the database to retrieve the next required product data and prepares it.

[0134] 7. The server delivers the prepared product data to the terminal.

[0135] 8. The device caches the received product data in its internal storage.

[0136] 9. When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[0137] Example prompts for specific generative AI models:

[0138] "We will provide an operation log of users frequently browsing products in the cosmetics category. Please predict their next action, pre-read the necessary product data, and deliver it to the device."

[0139] In this way, this system can improve the user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and delivering resources, and caching.

[0140] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0141] Step 1: Collecting operational logs

[0142] The device monitors user operations (clicks, scrolling, input, etc.) in real time and records the operation log in its internal storage. Specifically, the device records the user's tap and scroll position and time.

[0143] Input: User actions (clicks, scrolling, typing, etc.)

[0144] Data processing: Log location, time, and type of operation

[0145] Output: Operation log saved in internal storage

[0146] Step 2: Send the operation log

[0147] The terminal sends the operation log to the server at regular intervals or when certain conditions are met, for example, when the log size exceeds 1MB.

[0148] Input: Operation log saved in internal storage

[0149] Data processing: Converts operation logs into a format that can be sent to the server

[0150] Output: Operation log sent to the server

[0151] Step 3: Analyze the data

[0152] The server analyzes the received operation logs using machine learning algorithms such as Python's scikit-learn and TensorFlow.

[0153] Input: Operation log sent to the server

[0154] Data processing: Convert to a Pandas data frame and fit to a machine learning model

[0155] Output: Analysis results (user behavior patterns)

[0156] Step 4: Identify patterns of behavior

[0157] The server uses the analysis results to identify user behavior patterns, such as the tendency to use certain functions at certain times of the day.

[0158] Input: Analysis results (user behavior patterns)

[0159] Data processing: Identifying behavioral patterns using clustering and pattern mining techniques

[0160] Output: Identified behavioral patterns

[0161] Step 5: Predict behavior

[0162] Based on the identified behavioral patterns, the server predicts the user's likely next action.

[0163] Input: Identified behavioral patterns

[0164] Data processing: Using predictive models to estimate the next steps

[0165] Output: Predicted next action

[0166] Step 6: Prefetching resources

[0167] Based on predicted behavior, the server pre-fetches resources that will be needed next, including, for example, news articles or product information that are likely to be viewed next.

[0168] Input: Predicted next action

[0169] Data processing: Query the relevant database to obtain the required resources

[0170] Output: Prefetched resources (news articles, product information, etc.)

[0171] Step 7: Deliver resources

[0172] The server delivers the prefetched resources to the terminal.

[0173] Input: Prefetched resources

[0174] Data processing: Convert resources into a distribution format and send them to the terminal

[0175] Output: Resources delivered to the device

[0176] Step 8: Saving the cache

[0177] The device stores the received resource in a local cache and displays it immediately when the user actually needs it.

[0178] Input: Resources delivered from the server

[0179] Data processing: Save resources to internal storage or NSCache

[0180] Output: Cached resources

[0181] Through these steps, user operation logs are collected, analyzed, and the next action is predicted, and necessary resources are pre-loaded and delivered quickly, improving the user experience.

[0182] (Application example 1)

[0183] 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."

[0184] Existing user experience improvement systems collect user operation logs and identify behavioral patterns to predict the next action and prefetch resources. However, there is a lack of systems that ensure immediate responsiveness to user actions while also having extensive caching capabilities. Furthermore, in scenarios such as virtual stores, optimization methods are required to instantly display the products and information that the user will be interested in next. Conventional systems lack sufficient resource prediction and caching capabilities for identified actions, making it difficult to provide end users with prompt services. Furthermore, the lack of optimization based on network conditions and user environments contributes to a decline in the quality of the user experience.

[0185] 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.

[0186] In this invention, the server includes means for collecting user operation logs, means for analyzing the collected operation logs to identify the user's behavioral patterns, means for predicting the user's next action based on the identified behavioral patterns, means for prefetching resources corresponding to the user's next action, means for delivering the prefetched resources to the user terminal, means for caching the delivered resources in the terminal, and means for instantly displaying the retained resources on the interface. This makes it possible to instantly provide products and information that the user will be interested in next, improving the shopping experience in virtual stores, etc. Furthermore, by providing optimized resources taking into account the user's network status, the quality of the user experience can be improved.

[0187] A "user operation log" is data that records the operation history of a user, such as clicking, scrolling, and inputting, when the user operates an application.

[0188] A "behavioral pattern" is a tendency or tendency for a particular operation or behavior to be repeated, extracted by analyzing a user's operation log.

[0189] "Next Action" refers to the action or choice a user is likely to take next, predicted based on identified behavioral patterns.

[0190] "Resources" refers to data or information that corresponds to the user's next action, specifically articles, images, product information, etc.

[0191] "Predictive load" means preparing and delivering the resources needed in advance for the next predicted action, so that they are available immediately when the user requests them.

[0192] "Cache" refers to a memory area or the act of temporarily storing data that has been retrieved, and is used to quickly provide that data when the user needs it again.

[0193] "Immediate display on the interface" means that cached resources are instantly displayed on the display screen or application in response to user actions.

[0194] "Network status" refers to the communication status between the user terminal and the server, specifically information about network performance such as communication speed, latency, and bandwidth.

[0195] "Optimization measures" are mechanisms for selecting and implementing optimal methods and techniques to improve user experience based on the user's network conditions and behavioral patterns.

[0196] "Predicted resources" are data or information that are prepared in advance by predicting the user's next action.

[0197] "Retained resources" refers to data or information that is cached on the device through prediction or prefetching, and is immediately available when needed.

[0198] This invention relates to a system that collects user operation logs, analyzes them to identify behavioral patterns, predicts future actions, and ultimately improves the user experience. This system mainly operates in cooperation with a server and a terminal. The server is responsible for analyzing operation logs, identifying behavioral patterns, predicting actions, and pre-fetching and distributing resources, while the terminal is responsible for recording, caching, and immediately displaying the collected operation logs.

[0199] The system works as follows: First, when a user operates an application, the device monitors the operation (clicking, scrolling, input, etc.) in real time and records the operation log internally. The collected operation log is then periodically sent to the server.

[0200] The server then analyzes the received operation logs using machine learning algorithms to identify user behavior patterns, such as the time of day when users tend to use certain features or frequently browse products in a particular category.

[0201] Next, based on the behavioral patterns, the server predicts the user's likely next action. Based on the prediction results, the server pre-fetches the next required resource (such as product information or article data) and delivers it to the device. The delivered resource is cached on the device and is displayed immediately when the user actually needs it.

[0202] As a concrete example, consider a shopping experience in a virtual store. When a user puts on a head-mounted display (HMD) and enters the virtual store, an operation log is recorded in real time. When the user browses multiple products (shoes, shirts, etc.), the operation log is sent to a server, which analyzes it and predicts that the user is likely to browse "pants" next. The server then pre-fetches the product information for "pants," which will be needed next, delivers it to the HMD, and stores it in a cache. When the user moves to the "pants" section, the cached product information is immediately displayed, providing a fast shopping experience.

[0203] (Example of a prompt)

[0204] User records: [view_shoes, view_shirt]

[0205] Next action prediction: recommend_pants

[0206] Cached resources: pants information

[0207] The system uses hardware and software such as a head-mounted display (HMD), a program implemented in Python, data exchange in JSON format, and the requests library for HTTP requests.

[0208] In this way, the present invention realizes a system that improves user experience by combining analysis of collected operation logs, identification of behavioral patterns, behavior prediction, resource prefetching and distribution, and caching functions.

[0209] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0210] Step 1:

[0211] The user interacts with the application.

[0212] At this time, the device monitors and records in real time the user's operations, such as clicking, scrolling, and input. This recorded operation log becomes a dataset that will be used for subsequent analysis.

[0213] Input: User actions

[0214] Output: Operation log recorded internally on the device

[0215] Step 2:

[0216] The recorded operation log is sent to the server periodically or when certain conditions are met.

[0217] The terminal sends operation logs to the server at specified intervals or based on specific trigger conditions, using HTTP requests to transfer data in JSON format.

[0218] Input: Operation log (internal to the terminal)

[0219] Output: Operation log (sent to server)

[0220] Step 3:

[0221] The server analyzes the received operation logs and identifies the user's behavior patterns.

[0222] The server inputs the operation logs into a machine learning algorithm to extract user operation trends. This analysis reveals behavioral patterns, such as frequent use of certain functions at certain times of the day.

[0223] Input: Operation log (data sent to the server)

[0224] Output: Identified behavioral patterns

[0225] Step 4:

[0226] The server predicts the user's next action based on the identified behavioral patterns.

[0227] Based on machine learning models, the server predicts what the user is likely to do next, and this prediction determines the next resource required.

[0228] Input: Behavioral pattern analysis results

[0229] Output: Predicted next action

[0230] Step 5:

[0231] The server pre-fetches the required resources based on the predicted next action and delivers them to the terminal.

[0232] The prefetched resources (e.g., product information, article data, etc.) are sent to the terminal via an HTTP request.

[0233] Input: Predicted next action

[0234] Output: Prefetched resources (delivered to device)

[0235] Step 6:

[0236] The device caches the delivered resources.

[0237] The device caches the received resources in local storage, making them immediately available when the user needs them.

[0238] Input: Prefetched resources

[0239] Output: Cached resources

[0240] Step 7:

[0241] The device will immediately display the cached resource when the user next performs an action.

[0242] For example, when a user browses for "pants" in a virtual store, the cached product information for "pants" is immediately displayed, providing a fast browsing experience for the user.

[0243] Input: User's next action

[0244] Output: View cached resources

[0245] In this way, the system of the present invention provides end users with fast and efficient services through a series of processes, from analyzing collected operation logs to predicting behavior, pre-reading and distributing resources, caching, and displaying them immediately. This entire process significantly improves the user experience.

[0246] 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.

[0247] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[0248] Program processing overview

[0249] This system works in cooperation with a server, a device, and an emotion engine. When a user operates an application, an operation log is collected, along with the user's emotional data. This data is sent to the server, where it is analyzed. The server identifies the user's behavioral patterns, predicts the user's next action, and pre-reads and delivers the next required resource to the device. The device stores this data in a cache and responds immediately to the user's operations.

[0250] 1. Collection and transmission of operation logs and emotion data

[0251] When a user operates an application, the operation details are recorded as an operation log. In addition, the emotion engine analyzes the user's facial expressions and voice to obtain emotional data. This is done in real time.

[0252] 2. Data collection and transmission

[0253] The device collects the user's operation log and emotional data and sends it to the server periodically or when certain conditions are met.

[0254] 3. Analyzing data and identifying behavioral patterns

[0255] The server stores the received operation logs and emotion data in a database and analyzes them using machine learning algorithms, thereby identifying user behavior patterns and emotional trends.

[0256] 4. Behavioral prediction and emotion-based compensation

[0257] The server updates the behavioral model based on the analysis results. The behavioral model reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Emotional data is an important factor in correcting behavioral predictions.

[0258] 5. Resource Prefetching and Delivery

[0259] The server uses a behavioral characteristic model to predict the user's next action, and pre-fetches and delivers the required resources (articles, images, product information, etc.) to the device.

[0260] 6. Resource caching and fast display

[0261] The device can cache the received resources and display them immediately when the user takes their next action.

[0262] Specific examples

[0263] Example 1: News app

[0264] 1. A user interacts with a news app and frequently browses political news. Facial expression analysis detects that the user is feeling stressed while reading a specific article.

[0265] 2. The device records these operation logs and facial expression data and periodically transmits them to the server.

[0266] 3. The server analyzes the operation log to identify patterns in which users frequently view political news, and at the same time, detects content that is likely to cause stress to users.

[0267] 4. Based on the analysis results, the server predicts the next news article that is likely to be displayed and predicts content that takes the user's emotions into consideration.

[0268] 5. The server delivers the next required resource to the device, which stores it in its cache.

[0269] 6. The next time a user clicks on a political news item, the device will instantly display the cached article, along with relaxing content if the article is expected to cause stress.

[0270] Example 2: Shopping app

[0271] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative emotions are detected when a user cancels a purchase just before it is due to be made.

[0272] 2. The device records these operation logs and emotion data and sends them to the server.

[0273] 3. The server analyzes the data and determines that users frequently browse the cosmetics category and tend to be hesitant to make a purchase.

[0274] 4. The server uses the behavioral characteristic model to predict the next required product data and resources to increase the user's purchasing intent.

[0275] 5. The server delivers this data to the device, which stores it in its cache.

[0276] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[0277] In this way, the present invention provides a system that uses operation logs and emotion data to achieve more accurate behavior prediction and resource preemption, thereby improving the user experience.

[0278] The processing flow will be explained below.

[0279] Step 1:

[0280] The user interacts with the application: The user opens the news app, scrolls, and clicks to read a particular article.

[0281] Step 2:

[0282] The device records user operations (clicks, scrolls, timestamps, etc.) as a log, while also collecting facial expression and voice data using the device's built-in camera and microphone.

[0283] Step 3:

[0284] The emotion engine analyzes the collected facial and voice data to recognize the user's emotions. The emotion recognition results (e.g., "excitement," "stress," "joy," etc.) are recorded along with the operation log.

[0285] Step 4:

[0286] The terminal transmits operation logs and emotion data to the server for a certain period of time or under certain conditions (for example, a specific time or page transition).

[0287] Step 5:

[0288] The server stores the received operation logs and emotion data in a database, including the user's identification information, operation details, timestamps, and emotion recognition results.

[0289] Step 6:

[0290] The server uses machine learning algorithms to analyze past operation logs and emotional data stored in a database, thereby identifying user behavior patterns and emotional trends.

[0291] Step 7:

[0292] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data and is used to predict the next action.

[0293] Step 8:

[0294] The server uses the behavioral trait model to predict what a user is likely to do next, for example determining that they are likely to view an article in a particular category next.

[0295] Step 9:

[0296] The server identifies the required resources (specific articles, images, product information, etc.) based on predicted behavior. Based on the emotional data, the server adjusts the resources according to the user's current emotional state. For example, if the emotional state is positive, relevant articles will be selected, and if the emotional state is negative, stress-reducing content will be selected.

[0297] Step 10:

[0298] The server delivers the predicted resource data to the device, including the specific article content and image files that will be needed next.

[0299] Step 11:

[0300] The device caches the received resource data, allowing it to be accessed quickly the next time the user takes action.

[0301] Step 12:

[0302] The user actually takes the following action: for example, clicks on an article in a particular category.

[0303] Step 13:

[0304] The device will instantly display cached resources, providing a better user experience by presenting content that takes into account the user's emotional state.

[0305] Step 14:

[0306] The server receives the user's real-time operation log and emotion data again, and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[0307] Through the above processing steps, the system predicts user behavior and emotions in real time, pre-fetching and caching necessary resources, thereby significantly improving the user experience.

[0308] Example 2

[0309] 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."

[0310] Conventional user behavior prediction systems identify behavioral patterns based on user operation logs and predict the user's next actions. However, the behavioral predictions obtained in this way are often inaccurate because they do not take into account the user's internal emotional state. In particular, when a user's psychological state significantly influences their behavior, conventional systems are unable to properly capture these fluctuations, and an improvement in the user experience cannot be expected. For this reason, more accurate behavior predictions and the provision of appropriate resources are needed.

[0311] 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.

[0312] In this invention, the server includes means for collecting user operation logs, means for acquiring user emotion data in real time, and means for periodically transmitting the collected operation logs and emotion data to the server, thereby enabling behavior prediction that takes into account both the user operation logs and emotion data, thereby improving the user experience.

[0313] A "user operation log" is data that records information such as a timestamp, operation content, and operation target when a user operates an application.

[0314] "Emotion data" is data that includes emotion labels such as joy, sadness, anger, and surprise, and their intensities, obtained by analyzing the user's facial expressions and voice.

[0315] The "server" is a central processing unit that receives, stores, and analyzes data sent from the terminal to identify trends in user behavior patterns and emotional fluctuations.

[0316] A "terminal" is a device that is directly operated by the user and is responsible for collecting operation logs and emotional data and transmitting them to a server.

[0317] A "machine learning algorithm" is a computational method for analyzing large data sets and identifying patterns, a technique used to build models of user behavioral characteristics.

[0318] A "behavioral characteristic model" is a mathematical model that learns a user's past operation patterns and emotional data and predicts future behavior.

[0319] "Resources" refers to digital content such as articles, images, and product information that are needed for the user's next action.

[0320] A "cache" is a high-speed storage area that stores frequently accessed data in advance, and is used to respond quickly to user operations.

[0321] "Preprocessing" is the process of preparing data for analysis, and includes data cleansing and filtering.

[0322] "User emotion fluctuation tendency" is an index that shows how a user's emotions change over time and depending on specific operations.

[0323] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[0324] The system consists of a device operated by the user, a server that analyzes data, and an emotion engine that acquires emotional data. The device is a user device such as a smartphone, tablet, or PC, and the emotion engine is a software module that uses facial expression and voice recognition. The server includes a cloud server and a dedicated database server for large-scale data processing.

[0325] When a user operates the application, the operation details are recorded as an operation log with a timestamp. In addition, the emotion engine analyzes the user's facial expressions and voice in real time and records them as emotion data. This data is sent to the server periodically or when a specific trigger event occurs.

[0326] The server accumulates the received operation logs and emotion data in a data store and saves it in a database. Next, preprocessing such as data cleansing and filtering is performed, and machine learning algorithms are applied. Specifically, a time series analysis model (e.g., LSTM) is used to identify behavioral patterns, and a natural language processing model (e.g., BERT) is used to analyze emotion data.

[0327] The behavioral model built based on the analysis results reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Using this behavioral model, the server prefetches the required resources (such as news articles or product information that are likely to be viewed next) and sends them to the device.

[0328] The device can cache the received resources and respond quickly to the user's next operation, thereby shortening loading times and improving the user experience.

[0329] Specific examples

[0330] Example 1: News app

[0331] 1. A user frequently browses political news on a news app. The timestamp and operation details are recorded as an operation log, and stress is detected by facial expression analysis.

[0332] 2. The device periodically sends operation logs and emotion data to the server. The trigger conditions for sending are a certain viewing time or number of operations.

[0333] 3. The server accumulates this data, performs preprocessing, and then analyzes it using a machine learning algorithm. From past data, the system learns behavioral patterns when the user feels stressed.

[0334] 4. The server predicts which article is likely to be displayed next and adjusts it based on sentiment data, including selecting relaxing content to reduce stress.

[0335] 5. The server pre-fetches the required resources and delivers them to the device.

[0336] 6. The device will cache it and display it immediately the next time the user clicks on a political news item, resulting in faster loading times.

[0337] Example 2: Shopping app

[0338] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative sentiment is detected when a user cancels a purchase just before it is due to be made.

[0339] 2. The device records the operation log and emotion data and sends them to the server.

[0340] 3. The server stores this data, preprocesses it, and then analyzes it using machine learning algorithms to identify trends that could cause users to hesitate before making a purchase.

[0341] 4. The server uses the behavioral characteristic model to predict the next product data needed and resources that will increase purchasing motivation.

[0342] 5. The server delivers this data to the device, which stores it in its cache.

[0343] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[0344] Example prompts to input to the generative AI model

[0345] "Please provide us with the operation logs and sentiment data necessary to predict the next user action."

[0346] In this way, the present invention provides a system that uses operation logs and emotion data to predict user behavior and predict optimal resources, thereby improving the user experience.

[0347] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0348] Step 1:

[0349] When a user interacts with an application, the details of the interaction are recorded in an operation log along with a timestamp. Specifically, every time a user clicks a button or navigates to a different page, the interaction is saved as a log. This operation log is part of the data used for subsequent analysis.

[0350] Input: User actions (clicks, page transitions, etc.)

[0351] Output: Operation log (timestamp and operation details)

[0352] Step 2:

[0353] The device analyzes the user's facial expressions and voice in real time using an emotion engine to obtain emotional data. The emotion engine is a software module that captures the user's facial expressions and voice using a camera and microphone and analyzes the data. The analysis results are output as a label and intensity value that indicates the user's emotional state.

[0354] Input: User's facial expressions, voice

[0355] Output: Emotion data (emotion labels and their intensities)

[0356] Step 3:

[0357] The operation log and emotion data collected by the device are sent to a server periodically or according to trigger conditions. For example, data is compiled and sent after a certain number of operations or at certain time intervals. The data sent to the server is used for subsequent analysis.

[0358] Input: Operation log, emotion data

[0359] Output: The dataset sent to the server

[0360] Step 4:

[0361] The server accumulates the received operation logs and emotion data in a large-scale data store and saves it in a database. This data is necessary for later high-speed query execution. Specifically, Amazon S3 or Google Cloud Storage can be used.

[0362] Input: The dataset sent to the server

[0363] Output: Accumulated data

[0364] Step 5:

[0365] The server preprocesses the accumulated data. Preprocessing involves cleansing and filtering the data to remove incomplete or duplicate data. It also normalizes and transforms the data, preparing it for machine learning algorithms to work properly.

[0366] Input: Accumulated data

[0367] Output: Preprocessed data

[0368] Step 6:

[0369] The server runs machine learning algorithms on the preprocessed data, such as using an LSTM model to analyze behavioral patterns and a BERT model to analyze sentiment data, to build a model of user behavioral characteristics.

[0370] Input: Preprocessed data

[0371] Output: Behavioral characteristics model

[0372] Step 7:

[0373] The server uses the constructed behavioral characteristic model to predict the user's next action. This prediction is made using past operation logs and emotional data. For example, it predicts what operation the user will perform next based on past patterns of performing specific operations while in a specific emotional state.

[0374] Input: Behavioral characteristics model, past operation logs and emotion data

[0375] Output: Next action prediction result

[0376] Step 8:

[0377] The server pre-reads the next required resource and delivers it to the device. Based on the predicted next action, it selects the articles, images, product information, etc. needed for that action and sends them to the device.

[0378] Input: Next action prediction result

[0379] Output: Prefetched resources

[0380] Step 9:

[0381] The device caches the delivered resources and displays them quickly the next time the user performs an action. This reduces loading time and improves the user experience. Specifically, data that has been cached in advance is displayed immediately.

[0382] Input: Prefetched resources

[0383] Output: Cached data, fast display

[0384] The above steps enable highly accurate behavioral prediction and resource preemption using user operation logs and emotional data.

[0385] (Application example 2)

[0386] 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."

[0387] Autonomous vehicle systems are required to predict the next action in real time using occupant operation logs and emotional data, and to provide optimal driving modes and routes while reducing occupant stress. However, conventional systems do not adequately collect and analyze emotional data, making it difficult to respond according to the occupant's emotional state. Therefore, more accurate behavioral prediction, including emotional data, is necessary to provide a comfortable and safe driving experience.

[0388] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user operation logs, means for identifying the user's behavior pattern by analyzing the collected operation logs and the user's emotional data, means for predicting the user's next behavior based on the identified behavior pattern and emotional data, means for prefetching resources corresponding to the predicted next behavior based on the emotional data, means for delivering the prefetched resources to the user terminal, and means for caching the delivered resources in the terminal. This enables behavior prediction that takes the user's emotional state into consideration in real time, making it possible to provide the optimal driving mode and route while reducing stress on passengers.

[0389] A "user operation log" is a series of operation records generated when a user operates a system or terminal.

[0390] "Emotional data" refers to data that indicates the user's emotional state, obtained from facial expressions, voice, heart rate, etc.

[0391] "Behavioral patterns" refer to the behavioral tendencies and characteristics obtained by analyzing a user's series of operation logs and emotional data.

[0392] "Predicting the next action" means predicting the next action the user will take based on past operation logs and emotional data.

[0393] "Pre-fetching resources" means preparing the data and content needed for the user's next predicted action in advance.

[0394] "User terminal" refers to a device or equipment that is directly operated by a user.

[0395] "Delivering" means that the server sends the prefetched resources to the user terminal.

[0396] "Caching on the device" means temporarily storing the delivered resource on the user's device.

[0397] This invention relates to a system that collects and analyzes the operation logs and emotional data of occupants in an autonomous vehicle, predicts their next actions based on the collected data, and provides the occupants with the optimal driving mode and route. This system is configured as follows.

[0398] When a user boards an autonomous vehicle, sensors (such as an operation panel, buttons, and screen) are used to collect operation log data, and an emotion engine (such as a camera, microphone, and heart rate sensor) is used to collect emotional data from the occupant. The operation log data records the operations performed by the occupant and the order in which they were performed. The emotional data also includes the occupant's facial expressions, voice, heart rate, and so on.

[0399] This data is collected in real time and recorded on the terminal (vehicle's computer). The terminal then sends the collected data to a server periodically or when certain conditions are met. The server then stores the data in a database and analyzes it using machine learning algorithms.

[0400] During the analysis process, the server identifies the user's behavioral patterns and emotional trends. Based on this, the server updates the behavioral characteristic model and predicts the next action to be taken. The behavioral characteristic model reflects past operation logs and emotional data, enabling highly accurate predictions. Based on the predicted behavior, resources that take emotional data into account (such as vehicle speed adjustment information, driving mode change instructions, relaxing music, etc.) are prepared in advance and distributed to the device.

[0401] The device caches the delivered resources and can immediately respond to the occupant's next action. For example, if the occupant's stress level rises while driving on the highway, the device can automatically adjust the atmosphere in the vehicle by selecting relaxing music or a gentle driving mode based on emotional data.

[0402] This series of processes enables behavior prediction and response that takes into account the user's emotional state in real time, improving passenger comfort and safety.

[0403] As a specific example, if a state of stress is detected while driving on a highway, the server analyzes this and operates the system to switch to an appropriate driving mode. An example of a prompt sentence to input to the generative AI model in this case could be, "Please explain a system that detects emotional data indicating a state of stress while driving on a highway and suggests driving modes or route changes." Using this prompt sentence, the system can acquire basic data for optimal processing.

[0404] As described above, this invention achieves more accurate behavior prediction and real-time response by integrating and analyzing operation logs and emotion data, thereby providing a comfortable driving experience for occupants of autonomous vehicles.

[0405] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0406] Step 1:

[0407] A user gets into an autonomous vehicle and operates the control panel and buttons, which generates an operation log. At the same time, the emotion engine uses cameras and microphones to sense the user's facial expressions, voice, heart rate, etc., and collects emotional data. The input is the user's operations and emotional state, and the output is the collected operation log and emotional data.

[0408] Step 2:

[0409] The terminal temporarily stores the collected operation log and emotion data and transmits them to the server at regular intervals (for example, in real time or at regular intervals). The input is the collected data, and the output is the data sent to the server. Specific operations include generating data packets and transmitting them via a communication network.

[0410] Step 3:

[0411] The server receives the operation log and emotion data and stores them in a database. Next, it uses machine learning algorithms to analyze this data and identify the user's behavioral patterns and emotional fluctuation trends. The input is the data sent to the server, and the output is the analysis results, which are specific data on behavioral patterns and emotional states. Specific operations include data cleaning, feature extraction, and model application.

[0412] Step 4:

[0413] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data, and contains information for predicting the user's next action. The input is the analysis results, and the output is the updated behavioral characteristic model. Specific operations include learning and updating the model.

[0414] Step 5:

[0415] The server predicts the next action based on the updated behavioral characteristic model. In particular, taking emotional data into consideration enables highly accurate predictions. Based on the prediction, the server looks ahead and prepares the required resources (e.g., driving mode adjustment information, relaxing music, etc.). The input is the updated behavioral characteristic model, and the output is the predicted next action and the required resources. Specific operations include applying the prediction algorithm and selecting resources.

[0416] Step 6:

[0417] The server delivers prefetched resources to the terminal. The terminal receives them and stores them in its cache. The input is the resource delivered from the server, and the output is the resource stored in the terminal's cache. The specific operations involve receiving data and writing it to the cache.

[0418] Step 7:

[0419] When the user takes the next action, the device immediately uses the cached resources and provides appropriate feedback. For example, if the user is feeling stressed, suggestions for relaxing music or driving modes may be displayed. The input is the user's action and the cached resources, and the output is feedback to the user. Specific operations include reading the resources and displaying them on the user interface.

[0420] In this way, each step is performed consecutively, enabling behavioral prediction and resource provision that takes into account the user's emotional state in real time.

[0421] 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.

[0422] 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.

[0423] 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.

[0424] [Second embodiment]

[0425] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0426] 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.

[0427] 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).

[0428] 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.

[0429] 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.

[0430] 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).

[0431] 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. 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.

[0432] 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.

[0433] 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.

[0434] 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.

[0435] 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.

[0436] 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."

[0437] The present invention relates to a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience.

[0438] Program processing overview

[0439] This system works in cooperation with a server and a terminal. When a user operates an application, the operation log is collected and sent to the server. The server analyzes the collected operation log, identifies behavioral patterns, and predicts the next action based on the identified behavioral patterns. Based on the prediction results, the server distributes the next required resource to the terminal, and the terminal stores it in its cache.

[0440] 1. Collecting and sending operation logs

[0441] The device monitors user operations (clicks, scrolls, inputs, etc.) in real time and records the operation logs internally. The operation logs are sent to the server periodically or when certain conditions are met.

[0442] 2. Analyzing data and identifying behavioral patterns

[0443] The server analyzes the received operation logs to identify user behavior patterns. This analysis uses machine learning algorithms to extract trends, such as the time of day when users tend to use certain functions.

[0444] 3. Behavioral prediction and resource preemption

[0445] The server predicts the next action a user is likely to take based on their behavioral patterns. For example, in the case of a news app, it predicts that a user who frequently views articles in a particular category is likely to click on the next article in that category. The server then pre-fetches the next required resources (articles, images, product information, etc.) and delivers them to the device.

[0446] 4. Resource Delivery and Caching

[0447] The server instructs the device to deliver the predicted resource, which the device stores in a local cache and displays immediately when the user actually needs it.

[0448] Specific examples

[0449] Example 1: News app

[0450] A user opens a news app and browses to a specific category (e.g., sports).

[0451] The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[0452] The server analyzes the received operation log and identifies that the user tends to frequently view articles in the sports category.

[0453] Based on this behavioral pattern, the server predicts that the user is likely to view an article in the sports category next.

[0454] The server pre-reads the latest sports articles and delivers them to the device, which then stores the articles in its cache.

[0455] When a user clicks on an article in the sports category, the device immediately displays the cached article.

[0456] Example 2: Shopping app

[0457] A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[0458] The terminal records these operation logs and periodically sends them to the server.

[0459] The server analyzes the data and determines that the user frequently views products in the cosmetics category.

[0460] The server predicts the user's next action and pre-fetches the next required product data.

[0461] The server delivers pre-fetched product data to the terminal, which then stores the data in its cache.

[0462] When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[0463] In this way, the present invention realizes a system that improves user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and distributing resources, and caching.

[0464] The processing flow will be explained below.

[0465] Step 1:

[0466] The user interacts with the application: within the news app, the user scrolls and clicks to browse different articles or navigate to a particular category.

[0467] Step 2:

[0468] The device logs all user actions (clicks, scrolls, typing, etc.), including timestamps and specific actions.

[0469] Step 3:

[0470] When a device has been in operation for a certain period of time or when certain conditions are met, the collected operation logs are sent to a server, and security measures such as encryption are taken when sending the data.

[0471] Step 4:

[0472] The server stores the received operation logs in a database. The stored data includes user identification information, operation details, timestamps, etc.

[0473] Step 5:

[0474] The server analyzes the operation logs using machine learning algorithms at regular intervals or when new data arrives, which involves extracting user behavior patterns, such as how often a user visits articles in a particular category at a particular time of day.

[0475] Step 6:

[0476] The server updates the user's behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and is used to predict the user's next actions.

[0477] Step 7:

[0478] The server uses the behavioral trait model to predict the next action the user is likely to take, for example determining that the user is likely to view an article in the sports category next.

[0479] Step 8:

[0480] The server determines the next required resource (article, image, product information, etc.) based on the predicted behavior. If the prediction is highly accurate, it issues a command to prefetch that resource.

[0481] Step 9:

[0482] The server delivers prefetched resource data to the device, including specific article content and image files that will be needed next.

[0483] Step 10:

[0484] The terminal caches the received resource data, allowing the resource to be accessed quickly without network delays.

[0485] Step 11:

[0486] A user actually takes the following action: for example, clicks on an article in the sports category.

[0487] Step 12:

[0488] The device will instantly display the cached resources, resulting in a smoother and faster operating experience for users.

[0489] Step 13:

[0490] The server receives the user's real-time operation log again and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[0491] Through the above processing steps, this system predicts user behavior in real time and pre-fetches and caches necessary resources, significantly improving the user experience.

[0492] Example 1

[0493] 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."

[0494] Conventional systems simply collect user operation logs, but lack the ability to predict the next operation or prefetch resources based on the predicted operation. As a result, the next resource the user needs is not provided immediately, resulting in a poor user experience. Furthermore, the lack of consideration for network conditions and real-time environmental data makes efficient data transfer and cache management difficult.

[0495] 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.

[0496] In this invention, the server includes means for transmitting a user's operation log to the server, means for analyzing the received operation log to identify the user's behavior pattern, and means for predicting the user's next behavior based on the identified behavior pattern. This makes it possible to analyze the user's operation log in real time, predict the next required resources, and deliver the predicted resources to the terminal.

[0497] "User" refers to any individual or corporation that uses this system.

[0498] An "operation log" is a record of operations such as clicking, scrolling, and input that occur when a user operates an application.

[0499] "Server" refers to a computer system that analyzes collected operation logs, identifies behavioral patterns, and distributes resources.

[0500] "Terminal" refers to a device operated by a user, and is hardware that collects operation logs and receives and displays resources.

[0501] "Behavioral patterns" refer to behavioral tendencies or characteristics identified based on a series of actions a user has taken in the past.

[0502] "Behavior prediction" refers to estimating the operations a user is likely to perform in the future based on identified behavioral patterns.

[0503] "Resources" refers to data or information that a user might need next (e.g., news articles, product information, images, etc.).

[0504] "Looking ahead" means predicting and preparing the resources a user will need next.

[0505] "Delivery" refers to sending prefetched resources from the server to the terminal.

[0506] "Cache" is a mechanism that temporarily stores distributed resources inside the device and provides them quickly when the user needs them.

[0507] The present invention is a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience. This system operates in cooperation with a server and terminals.

[0508] When a user operates an application, an operation log is collected and temporarily saved on the device. For example, it is common for Android and iOS devices to save operation logs in an SQLite database. The device sends the operation log to the server based on certain conditions. Specifically, it sends a POST request to the server when the log size exceeds 1MB or every certain time (e.g., every hour).

[0509] The server analyzes the received operation logs and identifies user behavior patterns. This analysis uses machine learning algorithms such as Python's scikit-learn and TensorFlow. Libraries such as Pandas and NumPy are used for data processing. Specifically, the server converts the received operation logs into a Pandas data frame and fits it to a machine learning model to identify user behavior patterns.

[0510] Once a behavioral pattern is identified, the server predicts the next action the user is likely to take. For example, in a news app, if it determines that a user tends to frequently view articles in a particular category (e.g., sports), it predicts that the user is likely to click on an article in that category next.

[0511] The server then prefetches the resources needed based on the predicted behavior. These resources can include news articles, images, product information, etc. The server queries the database to retrieve the next required resource and prepares it. Specifically, the server executes a query to retrieve the latest news article and related image data.

[0512] The server delivers the prepared resources to the device. The device stores the received resources in a local cache. For caching, internal storage is used on Android devices, and NSCache is used on iOS devices. When the user actually needs the resource, the device immediately displays the cached resource.

[0513] Specific examples

[0514] Specific examples of news apps

[0515] 1. A user opens a news app and browses to a specific category (e.g., sports).

[0516] 2. The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[0517] 3. The server reads the received operation log as a Pandas data frame and analyzes the data.

[0518] 4. Based on the analysis results, the server identifies a tendency for users to frequently view articles in the sports category.

[0519] 5. The server predicts the user's next action and determines that they are likely to view articles in the sports category.

[0520] 6. The server prefetches the next sports category article and image it needs and prepares it.

[0521] 7. The server delivers the prepared article and image data to the terminal.

[0522] 8. The device stores the received articles and images in a local cache.

[0523] 9. When a user clicks on an article in the Sports category, the device immediately displays the cached article.

[0524] Example prompts for specific generative AI models:

[0525] "We provide an operation log showing users frequently viewing news articles in the sports category. Please predict their next action and deliver the necessary resources to the device in advance."

[0526] Example of a shopping app

[0527] 1. A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[0528] 2. The terminal records these operation logs and periodically sends them to the server.

[0529] 3. Analyze the operation log data received by the server using a Pandas data frame.

[0530] 4. The server clusters user behavior patterns and identifies a tendency to frequently browse products in the cosmetics category.

[0531] 5. The server uses the predictive model to predict which cosmetic product is likely to be viewed next.

[0532] 6. The server queries the database to retrieve the next required product data and prepares it.

[0533] 7. The server delivers the prepared product data to the terminal.

[0534] 8. The device caches the received product data in its internal storage.

[0535] 9. When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[0536] Example prompts for specific generative AI models:

[0537] "We will provide an operation log of users frequently browsing products in the cosmetics category. Please predict their next action, pre-read the necessary product data, and deliver it to the device."

[0538] In this way, this system can improve the user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and delivering resources, and caching.

[0539] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0540] Step 1: Collecting operational logs

[0541] The device monitors user operations (clicks, scrolling, input, etc.) in real time and records the operation log in its internal storage. Specifically, the device records the user's tap and scroll position and time.

[0542] Input: User actions (clicks, scrolling, typing, etc.)

[0543] Data processing: Log location, time, and type of operation

[0544] Output: Operation log saved in internal storage

[0545] Step 2: Send the operation log

[0546] The terminal sends the operation log to the server at regular intervals or when certain conditions are met, for example, when the log size exceeds 1MB.

[0547] Input: Operation log saved in internal storage

[0548] Data processing: Converts operation logs into a format that can be sent to the server

[0549] Output: Operation log sent to the server

[0550] Step 3: Analyze the data

[0551] The server analyzes the received operation logs using machine learning algorithms such as Python's scikit-learn and TensorFlow.

[0552] Input: Operation log sent to the server

[0553] Data processing: Convert to a Pandas data frame and fit to a machine learning model

[0554] Output: Analysis results (user behavior patterns)

[0555] Step 4: Identify patterns of behavior

[0556] The server uses the analysis results to identify user behavior patterns, such as the tendency to use certain functions at certain times of the day.

[0557] Input: Analysis results (user behavior patterns)

[0558] Data processing: Identifying behavioral patterns using clustering and pattern mining techniques

[0559] Output: Identified behavioral patterns

[0560] Step 5: Predict behavior

[0561] Based on the identified behavioral patterns, the server predicts the user's likely next action.

[0562] Input: Identified behavioral patterns

[0563] Data processing: Using predictive models to estimate the next steps

[0564] Output: Predicted next action

[0565] Step 6: Prefetching resources

[0566] Based on predicted behavior, the server pre-fetches resources that will be needed next, including, for example, news articles or product information that are likely to be viewed next.

[0567] Input: Predicted next action

[0568] Data processing: Query the relevant database to obtain the required resources

[0569] Output: Prefetched resources (news articles, product information, etc.)

[0570] Step 7: Deliver resources

[0571] The server delivers the prefetched resources to the terminal.

[0572] Input: Prefetched resources

[0573] Data processing: Convert resources into a distribution format and send them to the terminal

[0574] Output: Resources delivered to the device

[0575] Step 8: Saving the cache

[0576] The device stores the received resource in a local cache and displays it immediately when the user actually needs it.

[0577] Input: Resources delivered from the server

[0578] Data processing: Save resources to internal storage or NSCache

[0579] Output: Cached resources

[0580] Through these steps, user operation logs are collected, analyzed, and the next action is predicted, and necessary resources are pre-loaded and delivered quickly, improving the user experience.

[0581] (Application example 1)

[0582] 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."

[0583] Existing user experience improvement systems collect user operation logs and identify behavioral patterns to predict the next action and prefetch resources. However, there is a lack of systems that ensure immediate responsiveness to user actions while also having extensive caching capabilities. Furthermore, in scenarios such as virtual stores, optimization methods are required to instantly display the products and information that the user will be interested in next. Conventional systems lack sufficient resource prediction and caching capabilities for identified actions, making it difficult to provide end users with prompt services. Furthermore, the lack of optimization based on network conditions and user environments contributes to a decline in the quality of the user experience.

[0584] 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.

[0585] In this invention, the server includes means for collecting user operation logs, means for analyzing the collected operation logs to identify the user's behavioral patterns, means for predicting the user's next action based on the identified behavioral patterns, means for prefetching resources corresponding to the user's next action, means for delivering the prefetched resources to the user terminal, means for caching the delivered resources in the terminal, and means for instantly displaying the retained resources on the interface. This makes it possible to instantly provide products and information that the user will be interested in next, improving the shopping experience in virtual stores, etc. Furthermore, by providing optimized resources taking into account the user's network status, the quality of the user experience can be improved.

[0586] A "user operation log" is data that records the operation history of a user, such as clicking, scrolling, and inputting, when the user operates an application.

[0587] A "behavioral pattern" is a tendency or tendency for a particular operation or behavior to be repeated, extracted by analyzing a user's operation log.

[0588] "Next Action" refers to the action or choice a user is likely to take next, predicted based on identified behavioral patterns.

[0589] "Resources" refers to data or information that corresponds to the user's next action, specifically articles, images, product information, etc.

[0590] "Predictive load" means preparing and delivering the resources needed in advance for the next predicted action, so that they are available immediately when the user requests them.

[0591] "Cache" refers to a memory area or the act of temporarily storing data that has been retrieved, and is used to quickly provide that data when the user needs it again.

[0592] "Immediate display on the interface" means that cached resources are instantly displayed on the display screen or application in response to user actions.

[0593] "Network status" refers to the communication status between the user terminal and the server, specifically information about network performance such as communication speed, latency, and bandwidth.

[0594] "Optimization measures" are mechanisms for selecting and implementing optimal methods and techniques to improve user experience based on the user's network conditions and behavioral patterns.

[0595] "Predicted resources" are data or information that are prepared in advance by predicting the user's next action.

[0596] "Retained resources" refers to data or information that is cached on the device through prediction or prefetching, and is immediately available when needed.

[0597] This invention relates to a system that collects user operation logs, analyzes them to identify behavioral patterns, predicts future actions, and ultimately improves the user experience. This system mainly operates in cooperation with a server and a terminal. The server is responsible for analyzing operation logs, identifying behavioral patterns, predicting actions, and pre-fetching and distributing resources, while the terminal is responsible for recording, caching, and immediately displaying the collected operation logs.

[0598] The system works as follows: First, when a user operates an application, the device monitors the operation (clicking, scrolling, input, etc.) in real time and records the operation log internally. The collected operation log is then periodically sent to the server.

[0599] The server then analyzes the received operation logs using machine learning algorithms to identify user behavior patterns, such as the time of day when users tend to use certain features or frequently browse products in a particular category.

[0600] Next, based on the behavioral patterns, the server predicts the user's likely next action. Based on the prediction results, the server pre-fetches the next required resource (such as product information or article data) and delivers it to the device. The delivered resource is cached on the device and is displayed immediately when the user actually needs it.

[0601] As a concrete example, consider a shopping experience in a virtual store. When a user puts on a head-mounted display (HMD) and enters the virtual store, an operation log is recorded in real time. When the user browses multiple products (shoes, shirts, etc.), the operation log is sent to a server, which analyzes it and predicts that the user is likely to browse "pants" next. The server then pre-fetches the product information for "pants," which will be needed next, delivers it to the HMD, and stores it in a cache. When the user moves to the "pants" section, the cached product information is immediately displayed, providing a fast shopping experience.

[0602] (Example of a prompt)

[0603] User records: [view_shoes, view_shirt]

[0604] Next action prediction: recommend_pants

[0605] Cached resources: pants information

[0606] The system uses hardware and software such as a head-mounted display (HMD), a program implemented in Python, data exchange in JSON format, and the requests library for HTTP requests.

[0607] In this way, the present invention realizes a system that improves user experience by combining analysis of collected operation logs, identification of behavioral patterns, behavior prediction, resource prefetching and distribution, and caching functions.

[0608] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0609] Step 1:

[0610] The user interacts with the application.

[0611] At this time, the device monitors and records in real time the user's operations, such as clicking, scrolling, and input. This recorded operation log becomes a dataset that will be used for subsequent analysis.

[0612] Input: User actions

[0613] Output: Operation log recorded internally on the device

[0614] Step 2:

[0615] The recorded operation log is sent to the server periodically or when certain conditions are met.

[0616] The terminal sends operation logs to the server at specified intervals or based on specific trigger conditions, using HTTP requests to transfer data in JSON format.

[0617] Input: Operation log (internal to the terminal)

[0618] Output: Operation log (sent to server)

[0619] Step 3:

[0620] The server analyzes the received operation logs and identifies the user's behavior patterns.

[0621] The server inputs the operation logs into a machine learning algorithm to extract user operation trends. This analysis reveals behavioral patterns, such as frequent use of certain functions at certain times of the day.

[0622] Input: Operation log (data sent to the server)

[0623] Output: Identified behavioral patterns

[0624] Step 4:

[0625] The server predicts the user's next action based on the identified behavioral patterns.

[0626] Based on machine learning models, the server predicts what the user is likely to do next, and this prediction determines the next resource required.

[0627] Input: Behavioral pattern analysis results

[0628] Output: Predicted next action

[0629] Step 5:

[0630] The server pre-fetches the required resources based on the predicted next action and delivers them to the terminal.

[0631] The prefetched resources (e.g., product information, article data, etc.) are sent to the terminal via an HTTP request.

[0632] Input: Predicted next action

[0633] Output: Prefetched resources (delivered to device)

[0634] Step 6:

[0635] The device caches the delivered resources.

[0636] The device caches the received resources in local storage, making them immediately available when the user needs them.

[0637] Input: Prefetched resources

[0638] Output: Cached resources

[0639] Step 7:

[0640] The device will immediately display the cached resource when the user next performs an action.

[0641] For example, when a user browses for "pants" in a virtual store, the cached product information for "pants" is immediately displayed, providing a fast browsing experience for the user.

[0642] Input: User's next action

[0643] Output: View cached resources

[0644] In this way, the system of the present invention provides end users with fast and efficient services through a series of processes, from analyzing collected operation logs to predicting behavior, pre-reading and distributing resources, caching, and displaying them immediately. This entire process significantly improves the user experience.

[0645] 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.

[0646] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[0647] Program processing overview

[0648] This system works in cooperation with a server, a device, and an emotion engine. When a user operates an application, an operation log is collected, along with the user's emotional data. This data is sent to the server, where it is analyzed. The server identifies the user's behavioral patterns, predicts the user's next action, and pre-reads and delivers the next required resource to the device. The device stores this data in a cache and responds immediately to the user's operations.

[0649] 1. Collection and transmission of operation logs and emotion data

[0650] When a user operates an application, the operation details are recorded as an operation log. In addition, the emotion engine analyzes the user's facial expressions and voice to obtain emotional data. This is done in real time.

[0651] 2. Data collection and transmission

[0652] The device collects the user's operation log and emotional data and sends it to the server periodically or when certain conditions are met.

[0653] 3. Analyzing data and identifying behavioral patterns

[0654] The server stores the received operation logs and emotion data in a database and analyzes them using machine learning algorithms, thereby identifying user behavior patterns and emotional trends.

[0655] 4. Behavioral prediction and emotion-based compensation

[0656] The server updates the behavioral model based on the analysis results. The behavioral model reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Emotional data is an important factor in correcting behavioral predictions.

[0657] 5. Resource Prefetching and Delivery

[0658] The server uses a behavioral characteristic model to predict the user's next action, and pre-fetches and delivers the required resources (articles, images, product information, etc.) to the device.

[0659] 6. Resource caching and fast display

[0660] The device can cache the received resources and display them immediately when the user takes their next action.

[0661] Specific examples

[0662] Example 1: News app

[0663] 1. A user interacts with a news app and frequently browses political news. Facial expression analysis detects that the user is feeling stressed while reading a specific article.

[0664] 2. The device records these operation logs and facial expression data and periodically transmits them to the server.

[0665] 3. The server analyzes the operation log to identify patterns in which users frequently view political news, and at the same time, detects content that is likely to cause stress to users.

[0666] 4. Based on the analysis results, the server predicts the next news article that is likely to be displayed and predicts content that takes the user's emotions into consideration.

[0667] 5. The server delivers the next required resource to the device, which stores it in its cache.

[0668] 6. The next time a user clicks on a political news item, the device will instantly display the cached article, along with relaxing content if the article is expected to cause stress.

[0669] Example 2: Shopping app

[0670] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative emotions are detected when a user cancels a purchase just before it is due to be made.

[0671] 2. The device records these operation logs and emotion data and sends them to the server.

[0672] 3. The server analyzes the data and determines that users frequently browse the cosmetics category and tend to be hesitant to make a purchase.

[0673] 4. The server uses the behavioral characteristic model to predict the next required product data and resources to increase the user's purchasing intent.

[0674] 5. The server delivers this data to the device, which stores it in its cache.

[0675] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[0676] In this way, the present invention provides a system that uses operation logs and emotion data to achieve more accurate behavior prediction and resource preemption, thereby improving the user experience.

[0677] The processing flow will be explained below.

[0678] Step 1:

[0679] The user interacts with the application: The user opens the news app, scrolls, and clicks to read a particular article.

[0680] Step 2:

[0681] The device records user operations (clicks, scrolls, timestamps, etc.) as a log, while also collecting facial expression and voice data using the device's built-in camera and microphone.

[0682] Step 3:

[0683] The emotion engine analyzes the collected facial and voice data to recognize the user's emotions. The emotion recognition results (e.g., "excitement," "stress," "joy," etc.) are recorded along with the operation log.

[0684] Step 4:

[0685] The terminal transmits operation logs and emotion data to the server for a certain period of time or under certain conditions (for example, a specific time or page transition).

[0686] Step 5:

[0687] The server stores the received operation logs and emotion data in a database, including the user's identification information, operation details, timestamps, and emotion recognition results.

[0688] Step 6:

[0689] The server uses machine learning algorithms to analyze past operation logs and emotional data stored in a database, thereby identifying user behavior patterns and emotional trends.

[0690] Step 7:

[0691] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data and is used to predict the next action.

[0692] Step 8:

[0693] The server uses the behavioral trait model to predict what a user is likely to do next, for example determining that they are likely to view an article in a particular category next.

[0694] Step 9:

[0695] The server identifies the required resources (specific articles, images, product information, etc.) based on predicted behavior. Based on the emotional data, the server adjusts the resources according to the user's current emotional state. For example, if the emotional state is positive, relevant articles will be selected, and if the emotional state is negative, stress-reducing content will be selected.

[0696] Step 10:

[0697] The server delivers the predicted resource data to the device, including the specific article content and image files that will be needed next.

[0698] Step 11:

[0699] The device caches the received resource data, allowing it to be accessed quickly the next time the user takes action.

[0700] Step 12:

[0701] The user actually takes the following action: for example, clicks on an article in a particular category.

[0702] Step 13:

[0703] The device will instantly display cached resources, providing a better user experience by presenting content that takes into account the user's emotional state.

[0704] Step 14:

[0705] The server receives the user's real-time operation log and emotion data again, and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[0706] Through the above processing steps, the system predicts user behavior and emotions in real time, pre-fetching and caching necessary resources, thereby significantly improving the user experience.

[0707] Example 2

[0708] 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."

[0709] Conventional user behavior prediction systems identify behavioral patterns based on user operation logs and predict the user's next actions. However, the behavioral predictions obtained in this way are often inaccurate because they do not take into account the user's internal emotional state. In particular, when a user's psychological state significantly influences their behavior, conventional systems are unable to properly capture these fluctuations, and an improvement in the user experience cannot be expected. For this reason, more accurate behavior predictions and the provision of appropriate resources are needed.

[0710] 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.

[0711] In this invention, the server includes means for collecting user operation logs, means for acquiring user emotion data in real time, and means for periodically transmitting the collected operation logs and emotion data to the server, thereby enabling behavior prediction that takes into account both the user operation logs and emotion data, thereby improving the user experience.

[0712] A "user operation log" is data that records information such as a timestamp, operation content, and operation target when a user operates an application.

[0713] "Emotion data" is data that includes emotion labels such as joy, sadness, anger, and surprise, and their intensities, obtained by analyzing the user's facial expressions and voice.

[0714] The "server" is a central processing unit that receives, stores, and analyzes data sent from the terminal to identify trends in user behavior patterns and emotional fluctuations.

[0715] A "terminal" is a device that is directly operated by the user and is responsible for collecting operation logs and emotional data and transmitting them to a server.

[0716] A "machine learning algorithm" is a computational method for analyzing large data sets and identifying patterns, a technique used to build models of user behavioral characteristics.

[0717] A "behavioral characteristic model" is a mathematical model that learns a user's past operation patterns and emotional data and predicts future behavior.

[0718] "Resources" refers to digital content such as articles, images, and product information that are needed for the user's next action.

[0719] A "cache" is a high-speed storage area that stores frequently accessed data in advance, and is used to respond quickly to user operations.

[0720] "Preprocessing" is the process of preparing data for analysis, and includes data cleansing and filtering.

[0721] "User emotion fluctuation tendency" is an index that shows how a user's emotions change over time and depending on specific operations.

[0722] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[0723] The system consists of a device operated by the user, a server that analyzes data, and an emotion engine that acquires emotional data. The device is a user device such as a smartphone, tablet, or PC, and the emotion engine is a software module that uses facial expression and voice recognition. The server includes a cloud server and a dedicated database server for large-scale data processing.

[0724] When a user operates the application, the operation details are recorded as an operation log with a timestamp. In addition, the emotion engine analyzes the user's facial expressions and voice in real time and records them as emotion data. This data is sent to the server periodically or when a specific trigger event occurs.

[0725] The server accumulates the received operation logs and emotion data in a data store and saves it in a database. Next, preprocessing such as data cleansing and filtering is performed, and machine learning algorithms are applied. Specifically, a time series analysis model (e.g., LSTM) is used to identify behavioral patterns, and a natural language processing model (e.g., BERT) is used to analyze emotion data.

[0726] The behavioral model built based on the analysis results reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Using this behavioral model, the server prefetches the required resources (such as news articles or product information that are likely to be viewed next) and sends them to the device.

[0727] The device can cache the received resources and respond quickly to the user's next operation, thereby shortening loading times and improving the user experience.

[0728] Specific examples

[0729] Example 1: News app

[0730] 1. A user frequently browses political news on a news app. The timestamp and operation details are recorded as an operation log, and stress is detected by facial expression analysis.

[0731] 2. The device periodically sends operation logs and emotion data to the server. The trigger conditions for sending are a certain viewing time or number of operations.

[0732] 3. The server accumulates this data, performs preprocessing, and then analyzes it using a machine learning algorithm. From past data, the system learns behavioral patterns when the user feels stressed.

[0733] 4. The server predicts which article is likely to be displayed next and adjusts it based on sentiment data, including selecting relaxing content to reduce stress.

[0734] 5. The server pre-fetches the required resources and delivers them to the device.

[0735] 6. The device will cache it and display it immediately the next time the user clicks on a political news item, resulting in faster loading times.

[0736] Example 2: Shopping app

[0737] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative sentiment is detected when a user cancels a purchase just before it is due to be made.

[0738] 2. The device records the operation log and emotion data and sends them to the server.

[0739] 3. The server stores this data, preprocesses it, and then analyzes it using machine learning algorithms to identify trends that could cause users to hesitate before making a purchase.

[0740] 4. The server uses the behavioral characteristic model to predict the next product data needed and resources that will increase purchasing motivation.

[0741] 5. The server delivers this data to the device, which stores it in its cache.

[0742] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[0743] Example prompts to input to the generative AI model

[0744] "Please provide us with the operation logs and sentiment data necessary to predict the next user action."

[0745] In this way, the present invention provides a system that uses operation logs and emotion data to predict user behavior and predict optimal resources, thereby improving the user experience.

[0746] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0747] Step 1:

[0748] When a user interacts with an application, the details of the interaction are recorded in an operation log along with a timestamp. Specifically, every time a user clicks a button or navigates to a different page, the interaction is saved as a log. This operation log is part of the data used for subsequent analysis.

[0749] Input: User actions (clicks, page transitions, etc.)

[0750] Output: Operation log (timestamp and operation details)

[0751] Step 2:

[0752] The device analyzes the user's facial expressions and voice in real time using an emotion engine to obtain emotional data. The emotion engine is a software module that captures the user's facial expressions and voice using a camera and microphone and analyzes the data. The analysis results are output as a label and intensity value that indicates the user's emotional state.

[0753] Input: User's facial expressions, voice

[0754] Output: Emotion data (emotion labels and their intensities)

[0755] Step 3:

[0756] The operation log and emotion data collected by the device are sent to a server periodically or according to trigger conditions. For example, data is compiled and sent after a certain number of operations or at certain time intervals. The data sent to the server is used for subsequent analysis.

[0757] Input: Operation log, emotion data

[0758] Output: The dataset sent to the server

[0759] Step 4:

[0760] The server accumulates the received operation logs and emotion data in a large-scale data store and saves it in a database. This data is necessary for later high-speed query execution. Specifically, Amazon S3 or Google Cloud Storage can be used.

[0761] Input: The dataset sent to the server

[0762] Output: Accumulated data

[0763] Step 5:

[0764] The server preprocesses the accumulated data. Preprocessing involves cleansing and filtering the data to remove incomplete or duplicate data. It also normalizes and transforms the data, preparing it for machine learning algorithms to work properly.

[0765] Input: Accumulated data

[0766] Output: Preprocessed data

[0767] Step 6:

[0768] The server runs machine learning algorithms on the preprocessed data, such as using an LSTM model to analyze behavioral patterns and a BERT model to analyze sentiment data, to build a model of user behavioral characteristics.

[0769] Input: Preprocessed data

[0770] Output: Behavioral characteristics model

[0771] Step 7:

[0772] The server uses the constructed behavioral characteristic model to predict the user's next action. This prediction is made using past operation logs and emotional data. For example, it predicts what operation the user will perform next based on past patterns of performing specific operations while in a specific emotional state.

[0773] Input: Behavioral characteristics model, past operation logs and emotion data

[0774] Output: Next action prediction result

[0775] Step 8:

[0776] The server pre-reads the next required resource and delivers it to the device. Based on the predicted next action, it selects the articles, images, product information, etc. needed for that action and sends them to the device.

[0777] Input: Next action prediction result

[0778] Output: Prefetched resources

[0779] Step 9:

[0780] The device caches the delivered resources and displays them quickly the next time the user performs an action. This reduces loading time and improves the user experience. Specifically, data that has been cached in advance is displayed immediately.

[0781] Input: Prefetched resources

[0782] Output: Cached data, fast display

[0783] The above steps enable highly accurate behavioral prediction and resource preemption using user operation logs and emotional data.

[0784] (Application example 2)

[0785] 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."

[0786] Autonomous vehicle systems are required to predict the next action in real time using occupant operation logs and emotional data, and to provide optimal driving modes and routes while reducing occupant stress. However, conventional systems do not adequately collect and analyze emotional data, making it difficult to respond according to the occupant's emotional state. Therefore, more accurate behavioral prediction, including emotional data, is necessary to provide a comfortable and safe driving experience.

[0787] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user operation logs, means for identifying the user's behavior pattern by analyzing the collected operation logs and the user's emotional data, means for predicting the user's next behavior based on the identified behavior pattern and emotional data, means for prefetching resources corresponding to the predicted next behavior based on the emotional data, means for delivering the prefetched resources to the user terminal, and means for caching the delivered resources in the terminal. This enables behavior prediction that takes the user's emotional state into consideration in real time, making it possible to provide the optimal driving mode and route while reducing stress on passengers.

[0788] A "user operation log" is a series of operation records generated when a user operates a system or terminal.

[0789] "Emotional data" refers to data that indicates the user's emotional state, obtained from facial expressions, voice, heart rate, etc.

[0790] "Behavioral patterns" refer to the behavioral tendencies and characteristics obtained by analyzing a user's series of operation logs and emotional data.

[0791] "Predicting the next action" means predicting the next action the user will take based on past operation logs and emotional data.

[0792] "Pre-fetching resources" means preparing the data and content needed for the user's next predicted action in advance.

[0793] "User terminal" refers to a device or equipment that is directly operated by a user.

[0794] "Delivering" means that the server sends the prefetched resources to the user terminal.

[0795] "Caching on the device" means temporarily storing the delivered resource on the user's device.

[0796] This invention relates to a system that collects and analyzes the operation logs and emotional data of occupants in an autonomous vehicle, predicts their next actions based on the collected data, and provides the occupants with the optimal driving mode and route. This system is configured as follows.

[0797] When a user boards an autonomous vehicle, sensors (such as an operation panel, buttons, and screen) are used to collect operation log data, and an emotion engine (such as a camera, microphone, and heart rate sensor) is used to collect emotional data from the occupant. The operation log data records the operations performed by the occupant and the order in which they were performed. The emotional data also includes the occupant's facial expressions, voice, heart rate, and so on.

[0798] This data is collected in real time and recorded on the terminal (vehicle's computer). The terminal then sends the collected data to a server periodically or when certain conditions are met. The server then stores the data in a database and analyzes it using machine learning algorithms.

[0799] During the analysis process, the server identifies the user's behavioral patterns and emotional trends. Based on this, the server updates the behavioral characteristic model and predicts the next action to be taken. The behavioral characteristic model reflects past operation logs and emotional data, enabling highly accurate predictions. Based on the predicted behavior, resources that take emotional data into account (such as vehicle speed adjustment information, driving mode change instructions, relaxing music, etc.) are prepared in advance and distributed to the device.

[0800] The device caches the delivered resources and can immediately respond to the occupant's next action. For example, if the occupant's stress level rises while driving on the highway, the device can automatically adjust the atmosphere in the vehicle by selecting relaxing music or a gentle driving mode based on emotional data.

[0801] This series of processes enables behavior prediction and response that takes into account the user's emotional state in real time, improving passenger comfort and safety.

[0802] As a specific example, if a state of stress is detected while driving on a highway, the server analyzes this and operates the system to switch to an appropriate driving mode. An example of a prompt sentence to input to the generative AI model in this case could be, "Please explain a system that detects emotional data indicating a state of stress while driving on a highway and suggests driving modes or route changes." Using this prompt sentence, the system can acquire basic data for optimal processing.

[0803] As described above, this invention achieves more accurate behavior prediction and real-time response by integrating and analyzing operation logs and emotion data, thereby providing a comfortable driving experience for occupants of autonomous vehicles.

[0804] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0805] Step 1:

[0806] A user gets into an autonomous vehicle and operates the control panel and buttons, which generates an operation log. At the same time, the emotion engine uses cameras and microphones to sense the user's facial expressions, voice, heart rate, etc., and collects emotional data. The input is the user's operations and emotional state, and the output is the collected operation log and emotional data.

[0807] Step 2:

[0808] The terminal temporarily stores the collected operation log and emotion data and transmits them to the server at regular intervals (for example, in real time or at regular intervals). The input is the collected data, and the output is the data sent to the server. Specific operations include generating data packets and transmitting them via a communication network.

[0809] Step 3:

[0810] The server receives the operation log and emotion data and stores them in a database. Next, it uses machine learning algorithms to analyze this data and identify the user's behavioral patterns and emotional fluctuation trends. The input is the data sent to the server, and the output is the analysis results, which are specific data on behavioral patterns and emotional states. Specific operations include data cleaning, feature extraction, and model application.

[0811] Step 4:

[0812] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data, and contains information for predicting the user's next action. The input is the analysis results, and the output is the updated behavioral characteristic model. Specific operations include learning and updating the model.

[0813] Step 5:

[0814] The server predicts the next action based on the updated behavioral characteristic model. In particular, taking emotional data into consideration enables highly accurate predictions. Based on the prediction, the server looks ahead and prepares the required resources (e.g., driving mode adjustment information, relaxing music, etc.). The input is the updated behavioral characteristic model, and the output is the predicted next action and the required resources. Specific operations include applying the prediction algorithm and selecting resources.

[0815] Step 6:

[0816] The server delivers prefetched resources to the terminal. The terminal receives them and stores them in its cache. The input is the resource delivered from the server, and the output is the resource stored in the terminal's cache. The specific operations involve receiving data and writing it to the cache.

[0817] Step 7:

[0818] When the user takes the next action, the device immediately uses the cached resources and provides appropriate feedback. For example, if the user is feeling stressed, suggestions for relaxing music or driving modes may be displayed. The input is the user's action and the cached resources, and the output is feedback to the user. Specific operations include reading the resources and displaying them on the user interface.

[0819] In this way, each step is performed consecutively, enabling behavioral prediction and resource provision that takes into account the user's emotional state in real time.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] [Third embodiment]

[0824] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0825] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0826] 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).

[0827] 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.

[0828] 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.

[0829] 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).

[0830] 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. 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.

[0831] 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.

[0832] 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.

[0833] 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.

[0834] 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.

[0835] 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."

[0836] The present invention relates to a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience.

[0837] Program processing overview

[0838] This system works in cooperation with a server and a terminal. When a user operates an application, the operation log is collected and sent to the server. The server analyzes the collected operation log, identifies behavioral patterns, and predicts the next action based on the identified behavioral patterns. Based on the prediction results, the server distributes the next required resource to the terminal, and the terminal stores it in its cache.

[0839] 1. Collecting and sending operation logs

[0840] The device monitors user operations (clicks, scrolls, inputs, etc.) in real time and records the operation logs internally. The operation logs are sent to the server periodically or when certain conditions are met.

[0841] 2. Analyzing data and identifying behavioral patterns

[0842] The server analyzes the received operation logs to identify user behavior patterns. This analysis uses machine learning algorithms to extract trends, such as the time of day when users tend to use certain functions.

[0843] 3. Behavioral prediction and resource preemption

[0844] The server predicts the next action a user is likely to take based on their behavioral patterns. For example, in the case of a news app, it predicts that a user who frequently views articles in a particular category is likely to click on the next article in that category. The server then pre-fetches the next required resources (articles, images, product information, etc.) and delivers them to the device.

[0845] 4. Resource Delivery and Caching

[0846] The server instructs the device to deliver the predicted resource, which the device stores in a local cache and displays immediately when the user actually needs it.

[0847] Specific examples

[0848] Example 1: News app

[0849] A user opens a news app and browses to a specific category (e.g., sports).

[0850] The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[0851] The server analyzes the received operation log and identifies that the user tends to frequently view articles in the sports category.

[0852] Based on this behavioral pattern, the server predicts that the user is likely to view an article in the sports category next.

[0853] The server pre-reads the latest sports articles and delivers them to the device, which then stores the articles in its cache.

[0854] When a user clicks on an article in the sports category, the device immediately displays the cached article.

[0855] Example 2: Shopping app

[0856] A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[0857] The terminal records these operation logs and periodically sends them to the server.

[0858] The server analyzes the data and determines that the user frequently views products in the cosmetics category.

[0859] The server predicts the user's next action and pre-fetches the next required product data.

[0860] The server delivers pre-fetched product data to the terminal, which then stores the data in its cache.

[0861] When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[0862] In this way, the present invention realizes a system that improves user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and distributing resources, and caching.

[0863] The processing flow will be explained below.

[0864] Step 1:

[0865] The user interacts with the application: within the news app, the user scrolls and clicks to browse different articles or navigate to a particular category.

[0866] Step 2:

[0867] The device logs all user actions (clicks, scrolls, typing, etc.), including timestamps and specific actions.

[0868] Step 3:

[0869] When a device has been in operation for a certain period of time or when certain conditions are met, the collected operation logs are sent to a server, and security measures such as encryption are taken when sending the data.

[0870] Step 4:

[0871] The server stores the received operation logs in a database. The stored data includes user identification information, operation details, timestamps, etc.

[0872] Step 5:

[0873] The server analyzes the operation logs using machine learning algorithms at regular intervals or when new data arrives, which involves extracting user behavior patterns, such as how often a user visits articles in a particular category at a particular time of day.

[0874] Step 6:

[0875] The server updates the user's behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and is used to predict the user's next actions.

[0876] Step 7:

[0877] The server uses the behavioral trait model to predict the next action the user is likely to take, for example determining that the user is likely to view an article in the sports category next.

[0878] Step 8:

[0879] The server determines the next required resource (article, image, product information, etc.) based on the predicted behavior. If the prediction is highly accurate, it issues a command to prefetch that resource.

[0880] Step 9:

[0881] The server delivers prefetched resource data to the device, including specific article content and image files that will be needed next.

[0882] Step 10:

[0883] The terminal caches the received resource data, allowing the resource to be accessed quickly without network delays.

[0884] Step 11:

[0885] A user actually takes the following action: for example, clicks on an article in the sports category.

[0886] Step 12:

[0887] The device will instantly display the cached resources, resulting in a smoother and faster operating experience for users.

[0888] Step 13:

[0889] The server receives the user's real-time operation log again and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[0890] Through the above processing steps, this system predicts user behavior in real time and pre-fetches and caches necessary resources, significantly improving the user experience.

[0891] Example 1

[0892] 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."

[0893] Conventional systems simply collect user operation logs, but lack the ability to predict the next operation or prefetch resources based on the predicted operation. As a result, the next resource the user needs is not provided immediately, resulting in a poor user experience. Furthermore, the lack of consideration for network conditions and real-time environmental data makes efficient data transfer and cache management difficult.

[0894] 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.

[0895] In this invention, the server includes means for transmitting a user's operation log to the server, means for analyzing the received operation log to identify the user's behavior pattern, and means for predicting the user's next behavior based on the identified behavior pattern. This makes it possible to analyze the user's operation log in real time, predict the next required resources, and deliver the predicted resources to the terminal.

[0896] "User" refers to any individual or corporation that uses this system.

[0897] An "operation log" is a record of operations such as clicking, scrolling, and input that occur when a user operates an application.

[0898] "Server" refers to a computer system that analyzes collected operation logs, identifies behavioral patterns, and distributes resources.

[0899] "Terminal" refers to a device operated by a user, and is hardware that collects operation logs and receives and displays resources.

[0900] "Behavioral patterns" refer to behavioral tendencies or characteristics identified based on a series of actions a user has taken in the past.

[0901] "Behavior prediction" refers to estimating the operations a user is likely to perform in the future based on identified behavioral patterns.

[0902] "Resources" refers to data or information that a user might need next (e.g., news articles, product information, images, etc.).

[0903] "Looking ahead" means predicting and preparing the resources a user will need next.

[0904] "Delivery" refers to sending prefetched resources from the server to the terminal.

[0905] "Cache" is a mechanism that temporarily stores distributed resources inside the device and provides them quickly when the user needs them.

[0906] The present invention is a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience. This system operates in cooperation with a server and terminals.

[0907] When a user operates an application, an operation log is collected and temporarily saved on the device. For example, it is common for Android and iOS devices to save operation logs in an SQLite database. The device sends the operation log to the server based on certain conditions. Specifically, it sends a POST request to the server when the log size exceeds 1MB or every certain time (e.g., every hour).

[0908] The server analyzes the received operation logs and identifies user behavior patterns. This analysis uses machine learning algorithms such as Python's scikit-learn and TensorFlow. Libraries such as Pandas and NumPy are used for data processing. Specifically, the server converts the received operation logs into a Pandas data frame and fits it to a machine learning model to identify user behavior patterns.

[0909] Once a behavioral pattern is identified, the server predicts the next action the user is likely to take. For example, in a news app, if it determines that a user tends to frequently view articles in a particular category (e.g., sports), it predicts that the user is likely to click on an article in that category next.

[0910] The server then prefetches the resources needed based on the predicted behavior. These resources can include news articles, images, product information, etc. The server queries the database to retrieve the next required resource and prepares it. Specifically, the server executes a query to retrieve the latest news article and related image data.

[0911] The server delivers the prepared resources to the device. The device stores the received resources in a local cache. For caching, internal storage is used on Android devices, and NSCache is used on iOS devices. When the user actually needs the resource, the device immediately displays the cached resource.

[0912] Specific examples

[0913] Specific examples of news apps

[0914] 1. A user opens a news app and browses to a specific category (e.g., sports).

[0915] 2. The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[0916] 3. The server reads the received operation log as a Pandas data frame and analyzes the data.

[0917] 4. Based on the analysis results, the server identifies a tendency for users to frequently view articles in the sports category.

[0918] 5. The server predicts the user's next action and determines that they are likely to view articles in the sports category.

[0919] 6. The server prefetches the next sports category article and image it needs and prepares it.

[0920] 7. The server delivers the prepared article and image data to the terminal.

[0921] 8. The device stores the received articles and images in a local cache.

[0922] 9. When a user clicks on an article in the Sports category, the device immediately displays the cached article.

[0923] Example prompts for specific generative AI models:

[0924] "We provide an operation log showing users frequently viewing news articles in the sports category. Please predict their next action and deliver the necessary resources to the device in advance."

[0925] Example of a shopping app

[0926] 1. A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[0927] 2. The terminal records these operation logs and periodically sends them to the server.

[0928] 3. Analyze the operation log data received by the server using a Pandas data frame.

[0929] 4. The server clusters user behavior patterns and identifies a tendency to frequently browse products in the cosmetics category.

[0930] 5. The server uses the predictive model to predict which cosmetic product is likely to be viewed next.

[0931] 6. The server queries the database to retrieve the next required product data and prepares it.

[0932] 7. The server delivers the prepared product data to the terminal.

[0933] 8. The device caches the received product data in its internal storage.

[0934] 9. When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[0935] Example prompts for specific generative AI models:

[0936] "We will provide an operation log of users frequently browsing products in the cosmetics category. Please predict their next action, pre-read the necessary product data, and deliver it to the device."

[0937] In this way, this system can improve the user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and delivering resources, and caching.

[0938] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0939] Step 1: Collecting operational logs

[0940] The device monitors user operations (clicks, scrolling, input, etc.) in real time and records the operation log in its internal storage. Specifically, the device records the user's tap and scroll position and time.

[0941] Input: User actions (clicks, scrolling, typing, etc.)

[0942] Data processing: Log location, time, and type of operation

[0943] Output: Operation log saved in internal storage

[0944] Step 2: Send the operation log

[0945] The terminal sends the operation log to the server at regular intervals or when certain conditions are met, for example, when the log size exceeds 1MB.

[0946] Input: Operation log saved in internal storage

[0947] Data processing: Converts operation logs into a format that can be sent to the server

[0948] Output: Operation log sent to the server

[0949] Step 3: Analyze the data

[0950] The server analyzes the received operation logs using machine learning algorithms such as Python's scikit-learn and TensorFlow.

[0951] Input: Operation log sent to the server

[0952] Data processing: Convert to a Pandas data frame and fit to a machine learning model

[0953] Output: Analysis results (user behavior patterns)

[0954] Step 4: Identify patterns of behavior

[0955] The server uses the analysis results to identify user behavior patterns, such as the tendency to use certain functions at certain times of the day.

[0956] Input: Analysis results (user behavior patterns)

[0957] Data processing: Identifying behavioral patterns using clustering and pattern mining techniques

[0958] Output: Identified behavioral patterns

[0959] Step 5: Predict behavior

[0960] Based on the identified behavioral patterns, the server predicts the user's likely next action.

[0961] Input: Identified behavioral patterns

[0962] Data processing: Using predictive models to estimate the next steps

[0963] Output: Predicted next action

[0964] Step 6: Prefetching resources

[0965] Based on predicted behavior, the server pre-fetches resources that will be needed next, including, for example, news articles or product information that are likely to be viewed next.

[0966] Input: Predicted next action

[0967] Data processing: Query the relevant database to obtain the required resources

[0968] Output: Prefetched resources (news articles, product information, etc.)

[0969] Step 7: Deliver resources

[0970] The server delivers the prefetched resources to the terminal.

[0971] Input: Prefetched resources

[0972] Data processing: Convert resources into a distribution format and send them to the terminal

[0973] Output: Resources delivered to the device

[0974] Step 8: Saving the cache

[0975] The device stores the received resource in a local cache and displays it immediately when the user actually needs it.

[0976] Input: Resources delivered from the server

[0977] Data processing: Save resources to internal storage or NSCache

[0978] Output: Cached resources

[0979] Through these steps, user operation logs are collected, analyzed, and the next action is predicted, and necessary resources are pre-loaded and delivered quickly, improving the user experience.

[0980] (Application example 1)

[0981] 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."

[0982] Existing user experience improvement systems collect user operation logs and identify behavioral patterns to predict the next action and prefetch resources. However, there is a lack of systems that ensure immediate responsiveness to user actions while also having extensive caching capabilities. Furthermore, in scenarios such as virtual stores, optimization methods are required to instantly display the products and information that the user will be interested in next. Conventional systems lack sufficient resource prediction and caching capabilities for identified actions, making it difficult to provide end users with prompt services. Furthermore, the lack of optimization based on network conditions and user environments contributes to a decline in the quality of the user experience.

[0983] 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.

[0984] In this invention, the server includes means for collecting user operation logs, means for analyzing the collected operation logs to identify the user's behavioral patterns, means for predicting the user's next action based on the identified behavioral patterns, means for prefetching resources corresponding to the user's next action, means for delivering the prefetched resources to the user terminal, means for caching the delivered resources in the terminal, and means for instantly displaying the retained resources on the interface. This makes it possible to instantly provide products and information that the user will be interested in next, improving the shopping experience in virtual stores, etc. Furthermore, by providing optimized resources taking into account the user's network status, the quality of the user experience can be improved.

[0985] A "user operation log" is data that records the operation history of a user, such as clicking, scrolling, and inputting, when the user operates an application.

[0986] A "behavioral pattern" is a tendency or tendency for a particular operation or behavior to be repeated, extracted by analyzing a user's operation log.

[0987] "Next Action" refers to the action or choice a user is likely to take next, predicted based on identified behavioral patterns.

[0988] "Resources" refers to data or information that corresponds to the user's next action, specifically articles, images, product information, etc.

[0989] "Predictive load" means preparing and delivering the resources needed in advance for the next predicted action, so that they are available immediately when the user requests them.

[0990] "Cache" refers to a memory area or the act of temporarily storing data that has been retrieved, and is used to quickly provide that data when the user needs it again.

[0991] "Immediate display on the interface" means that cached resources are instantly displayed on the display screen or application in response to user actions.

[0992] "Network status" refers to the communication status between the user terminal and the server, specifically information about network performance such as communication speed, latency, and bandwidth.

[0993] "Optimization measures" are mechanisms for selecting and implementing optimal methods and techniques to improve user experience based on the user's network conditions and behavioral patterns.

[0994] "Predicted resources" are data or information that are prepared in advance by predicting the user's next action.

[0995] "Retained resources" refers to data or information that is cached on the device through prediction or prefetching, and is immediately available when needed.

[0996] This invention relates to a system that collects user operation logs, analyzes them to identify behavioral patterns, predicts future actions, and ultimately improves the user experience. This system mainly operates in cooperation with a server and a terminal. The server is responsible for analyzing operation logs, identifying behavioral patterns, predicting actions, and pre-fetching and distributing resources, while the terminal is responsible for recording, caching, and immediately displaying the collected operation logs.

[0997] The system works as follows: First, when a user operates an application, the device monitors the operation (clicking, scrolling, input, etc.) in real time and records the operation log internally. The collected operation log is then periodically sent to the server.

[0998] The server then analyzes the received operation logs using machine learning algorithms to identify user behavior patterns, such as the time of day when users tend to use certain features or frequently browse products in a particular category.

[0999] Next, based on the behavioral patterns, the server predicts the user's likely next action. Based on the prediction results, the server pre-fetches the next required resource (such as product information or article data) and delivers it to the device. The delivered resource is cached on the device and is displayed immediately when the user actually needs it.

[1000] As a concrete example, consider a shopping experience in a virtual store. When a user puts on a head-mounted display (HMD) and enters the virtual store, an operation log is recorded in real time. When the user browses multiple products (shoes, shirts, etc.), the operation log is sent to a server, which analyzes it and predicts that the user is likely to browse "pants" next. The server then pre-fetches the product information for "pants," which will be needed next, delivers it to the HMD, and stores it in a cache. When the user moves to the "pants" section, the cached product information is immediately displayed, providing a fast shopping experience.

[1001] (Example of a prompt)

[1002] User records: [view_shoes, view_shirt]

[1003] Next action prediction: recommend_pants

[1004] Cached resources: pants information

[1005] The system uses hardware and software such as a head-mounted display (HMD), a program implemented in Python, data exchange in JSON format, and the requests library for HTTP requests.

[1006] In this way, the present invention realizes a system that improves user experience by combining analysis of collected operation logs, identification of behavioral patterns, behavior prediction, resource prefetching and distribution, and caching functions.

[1007] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1008] Step 1:

[1009] The user interacts with the application.

[1010] At this time, the device monitors and records in real time the user's operations, such as clicking, scrolling, and input. This recorded operation log becomes a dataset that will be used for subsequent analysis.

[1011] Input: User actions

[1012] Output: Operation log recorded internally on the device

[1013] Step 2:

[1014] The recorded operation log is sent to the server periodically or when certain conditions are met.

[1015] The terminal sends operation logs to the server at specified intervals or based on specific trigger conditions, using HTTP requests to transfer data in JSON format.

[1016] Input: Operation log (internal to the terminal)

[1017] Output: Operation log (sent to server)

[1018] Step 3:

[1019] The server analyzes the received operation logs and identifies the user's behavior patterns.

[1020] The server inputs the operation logs into a machine learning algorithm to extract user operation trends. This analysis reveals behavioral patterns, such as frequent use of certain functions at certain times of the day.

[1021] Input: Operation log (data sent to the server)

[1022] Output: Identified behavioral patterns

[1023] Step 4:

[1024] The server predicts the user's next action based on the identified behavioral patterns.

[1025] Based on machine learning models, the server predicts what the user is likely to do next, and this prediction determines the next resource required.

[1026] Input: Behavioral pattern analysis results

[1027] Output: Predicted next action

[1028] Step 5:

[1029] The server pre-fetches the required resources based on the predicted next action and delivers them to the terminal.

[1030] The prefetched resources (e.g., product information, article data, etc.) are sent to the terminal via an HTTP request.

[1031] Input: Predicted next action

[1032] Output: Prefetched resources (delivered to device)

[1033] Step 6:

[1034] The device caches the delivered resources.

[1035] The device caches the received resources in local storage, making them immediately available when the user needs them.

[1036] Input: Prefetched resources

[1037] Output: Cached resources

[1038] Step 7:

[1039] The device will immediately display the cached resource when the user next performs an action.

[1040] For example, when a user browses for "pants" in a virtual store, the cached product information for "pants" is immediately displayed, providing a fast browsing experience for the user.

[1041] Input: User's next action

[1042] Output: View cached resources

[1043] In this way, the system of the present invention provides end users with fast and efficient services through a series of processes, from analyzing collected operation logs to predicting behavior, pre-reading and distributing resources, caching, and displaying them immediately. This entire process significantly improves the user experience.

[1044] 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.

[1045] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[1046] Program processing overview

[1047] This system works in cooperation with a server, a device, and an emotion engine. When a user operates an application, an operation log is collected, along with the user's emotional data. This data is sent to the server, where it is analyzed. The server identifies the user's behavioral patterns, predicts the user's next action, and pre-reads and delivers the next required resource to the device. The device stores this data in a cache and responds immediately to the user's operations.

[1048] 1. Collection and transmission of operation logs and emotion data

[1049] When a user operates an application, the operation details are recorded as an operation log. In addition, the emotion engine analyzes the user's facial expressions and voice to obtain emotional data. This is done in real time.

[1050] 2. Data collection and transmission

[1051] The device collects the user's operation log and emotional data and sends it to the server periodically or when certain conditions are met.

[1052] 3. Analyzing data and identifying behavioral patterns

[1053] The server stores the received operation logs and emotion data in a database and analyzes them using machine learning algorithms, thereby identifying user behavior patterns and emotional trends.

[1054] 4. Behavioral prediction and emotion-based compensation

[1055] The server updates the behavioral model based on the analysis results. The behavioral model reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Emotional data is an important factor in correcting behavioral predictions.

[1056] 5. Resource Prefetching and Delivery

[1057] The server uses a behavioral characteristic model to predict the user's next action, and pre-fetches and delivers the required resources (articles, images, product information, etc.) to the device.

[1058] 6. Resource caching and fast display

[1059] The device can cache the received resources and display them immediately when the user takes their next action.

[1060] Specific examples

[1061] Example 1: News app

[1062] 1. A user interacts with a news app and frequently browses political news. Facial expression analysis detects that the user is feeling stressed while reading a specific article.

[1063] 2. The device records these operation logs and facial expression data and periodically transmits them to the server.

[1064] 3. The server analyzes the operation log to identify patterns in which users frequently view political news, and at the same time, detects content that is likely to cause stress to users.

[1065] 4. Based on the analysis results, the server predicts the next news article that is likely to be displayed and predicts content that takes the user's emotions into consideration.

[1066] 5. The server delivers the next required resource to the device, which stores it in its cache.

[1067] 6. The next time a user clicks on a political news item, the device will instantly display the cached article, along with relaxing content if the article is expected to cause stress.

[1068] Example 2: Shopping app

[1069] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative emotions are detected when a user cancels a purchase just before it is due to be made.

[1070] 2. The device records these operation logs and emotion data and sends them to the server.

[1071] 3. The server analyzes the data and determines that users frequently browse the cosmetics category and tend to be hesitant to make a purchase.

[1072] 4. The server uses the behavioral characteristic model to predict the next required product data and resources to increase the user's purchasing intent.

[1073] 5. The server delivers this data to the device, which stores it in its cache.

[1074] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[1075] In this way, the present invention provides a system that uses operation logs and emotion data to achieve more accurate behavior prediction and resource preemption, thereby improving the user experience.

[1076] The processing flow will be explained below.

[1077] Step 1:

[1078] The user interacts with the application: The user opens the news app, scrolls, and clicks to read a particular article.

[1079] Step 2:

[1080] The device records user operations (clicks, scrolls, timestamps, etc.) as a log, while also collecting facial expression and voice data using the device's built-in camera and microphone.

[1081] Step 3:

[1082] The emotion engine analyzes the collected facial and voice data to recognize the user's emotions. The emotion recognition results (e.g., "excitement," "stress," "joy," etc.) are recorded along with the operation log.

[1083] Step 4:

[1084] The terminal transmits operation logs and emotion data to the server for a certain period of time or under certain conditions (for example, a specific time or page transition).

[1085] Step 5:

[1086] The server stores the received operation logs and emotion data in a database, including the user's identification information, operation details, timestamps, and emotion recognition results.

[1087] Step 6:

[1088] The server uses machine learning algorithms to analyze past operation logs and emotional data stored in a database, thereby identifying user behavior patterns and emotional trends.

[1089] Step 7:

[1090] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data and is used to predict the next action.

[1091] Step 8:

[1092] The server uses the behavioral trait model to predict what a user is likely to do next, for example determining that they are likely to view an article in a particular category next.

[1093] Step 9:

[1094] The server identifies the required resources (specific articles, images, product information, etc.) based on predicted behavior. Based on the emotional data, the server adjusts the resources according to the user's current emotional state. For example, if the emotional state is positive, relevant articles will be selected, and if the emotional state is negative, stress-reducing content will be selected.

[1095] Step 10:

[1096] The server delivers the predicted resource data to the device, including the specific article content and image files that will be needed next.

[1097] Step 11:

[1098] The device caches the received resource data, allowing it to be accessed quickly the next time the user takes action.

[1099] Step 12:

[1100] The user actually takes the following action: for example, clicks on an article in a particular category.

[1101] Step 13:

[1102] The device will instantly display cached resources, providing a better user experience by presenting content that takes into account the user's emotional state.

[1103] Step 14:

[1104] The server receives the user's real-time operation log and emotion data again, and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[1105] Through the above processing steps, the system predicts user behavior and emotions in real time, pre-fetching and caching necessary resources, thereby significantly improving the user experience.

[1106] Example 2

[1107] 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."

[1108] Conventional user behavior prediction systems identify behavioral patterns based on user operation logs and predict the user's next actions. However, the behavioral predictions obtained in this way are often inaccurate because they do not take into account the user's internal emotional state. In particular, when a user's psychological state significantly influences their behavior, conventional systems are unable to properly capture these fluctuations, and an improvement in the user experience cannot be expected. For this reason, more accurate behavior predictions and the provision of appropriate resources are needed.

[1109] 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.

[1110] In this invention, the server includes means for collecting user operation logs, means for acquiring user emotion data in real time, and means for periodically transmitting the collected operation logs and emotion data to the server, thereby enabling behavior prediction that takes into account both the user operation logs and emotion data, thereby improving the user experience.

[1111] A "user operation log" is data that records information such as a timestamp, operation content, and operation target when a user operates an application.

[1112] "Emotion data" is data that includes emotion labels such as joy, sadness, anger, and surprise, and their intensities, obtained by analyzing the user's facial expressions and voice.

[1113] The "server" is a central processing unit that receives, stores, and analyzes data sent from the terminal to identify trends in user behavior patterns and emotional fluctuations.

[1114] A "terminal" is a device that is directly operated by the user and is responsible for collecting operation logs and emotional data and transmitting them to a server.

[1115] A "machine learning algorithm" is a computational method for analyzing large data sets and identifying patterns, a technique used to build models of user behavioral characteristics.

[1116] A "behavioral characteristic model" is a mathematical model that learns a user's past operation patterns and emotional data and predicts future behavior.

[1117] "Resources" refers to digital content such as articles, images, and product information that are needed for the user's next action.

[1118] A "cache" is a high-speed storage area that stores frequently accessed data in advance, and is used to respond quickly to user operations.

[1119] "Preprocessing" is the process of preparing data for analysis, and includes data cleansing and filtering.

[1120] "User emotion fluctuation tendency" is an index that shows how a user's emotions change over time and depending on specific operations.

[1121] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[1122] The system consists of a device operated by the user, a server that analyzes data, and an emotion engine that acquires emotional data. The device is a user device such as a smartphone, tablet, or PC, and the emotion engine is a software module that uses facial expression and voice recognition. The server includes a cloud server and a dedicated database server for large-scale data processing.

[1123] When a user operates the application, the operation details are recorded as an operation log with a timestamp. In addition, the emotion engine analyzes the user's facial expressions and voice in real time and records them as emotion data. This data is sent to the server periodically or when a specific trigger event occurs.

[1124] The server accumulates the received operation logs and emotion data in a data store and saves it in a database. Next, preprocessing such as data cleansing and filtering is performed, and machine learning algorithms are applied. Specifically, a time series analysis model (e.g., LSTM) is used to identify behavioral patterns, and a natural language processing model (e.g., BERT) is used to analyze emotion data.

[1125] The behavioral model built based on the analysis results reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Using this behavioral model, the server prefetches the required resources (such as news articles or product information that are likely to be viewed next) and sends them to the device.

[1126] The device can cache the received resources and respond quickly to the user's next operation, thereby shortening loading times and improving the user experience.

[1127] Specific examples

[1128] Example 1: News app

[1129] 1. A user frequently browses political news on a news app. The timestamp and operation details are recorded as an operation log, and stress is detected by facial expression analysis.

[1130] 2. The device periodically sends operation logs and emotion data to the server. The trigger conditions for sending are a certain viewing time or number of operations.

[1131] 3. The server accumulates this data, performs preprocessing, and then analyzes it using a machine learning algorithm. From past data, the system learns behavioral patterns when the user feels stressed.

[1132] 4. The server predicts which article is likely to be displayed next and adjusts it based on sentiment data, including selecting relaxing content to reduce stress.

[1133] 5. The server pre-fetches the required resources and delivers them to the device.

[1134] 6. The device will cache it and display it immediately the next time the user clicks on a political news item, resulting in faster loading times.

[1135] Example 2: Shopping app

[1136] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative sentiment is detected when a user cancels a purchase just before it is due to be made.

[1137] 2. The device records the operation log and emotion data and sends them to the server.

[1138] 3. The server stores this data, preprocesses it, and then analyzes it using machine learning algorithms to identify trends that could cause users to hesitate before making a purchase.

[1139] 4. The server uses the behavioral characteristic model to predict the next product data needed and resources that will increase purchasing motivation.

[1140] 5. The server delivers this data to the device, which stores it in its cache.

[1141] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[1142] Example prompts to input to the generative AI model

[1143] "Please provide us with the operation logs and sentiment data necessary to predict the next user action."

[1144] In this way, the present invention provides a system that uses operation logs and emotion data to predict user behavior and predict optimal resources, thereby improving the user experience.

[1145] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1146] Step 1:

[1147] When a user interacts with an application, the details of the interaction are recorded in an operation log along with a timestamp. Specifically, every time a user clicks a button or navigates to a different page, the interaction is saved as a log. This operation log is part of the data used for subsequent analysis.

[1148] Input: User actions (clicks, page transitions, etc.)

[1149] Output: Operation log (timestamp and operation details)

[1150] Step 2:

[1151] The device analyzes the user's facial expressions and voice in real time using an emotion engine to obtain emotional data. The emotion engine is a software module that captures the user's facial expressions and voice using a camera and microphone and analyzes the data. The analysis results are output as a label and intensity value that indicates the user's emotional state.

[1152] Input: User's facial expressions, voice

[1153] Output: Emotion data (emotion labels and their intensities)

[1154] Step 3:

[1155] The operation log and emotion data collected by the device are sent to a server periodically or according to trigger conditions. For example, data is compiled and sent after a certain number of operations or at certain time intervals. The data sent to the server is used for subsequent analysis.

[1156] Input: Operation log, emotion data

[1157] Output: The dataset sent to the server

[1158] Step 4:

[1159] The server accumulates the received operation logs and emotion data in a large-scale data store and saves it in a database. This data is necessary for later high-speed query execution. Specifically, Amazon S3 or Google Cloud Storage can be used.

[1160] Input: The dataset sent to the server

[1161] Output: Accumulated data

[1162] Step 5:

[1163] The server preprocesses the accumulated data. Preprocessing involves cleansing and filtering the data to remove incomplete or duplicate data. It also normalizes and transforms the data, preparing it for machine learning algorithms to work properly.

[1164] Input: Accumulated data

[1165] Output: Preprocessed data

[1166] Step 6:

[1167] The server runs machine learning algorithms on the preprocessed data, such as using an LSTM model to analyze behavioral patterns and a BERT model to analyze sentiment data, to build a model of user behavioral characteristics.

[1168] Input: Preprocessed data

[1169] Output: Behavioral characteristics model

[1170] Step 7:

[1171] The server uses the constructed behavioral characteristic model to predict the user's next action. This prediction is made using past operation logs and emotional data. For example, it predicts what operation the user will perform next based on past patterns of performing specific operations while in a specific emotional state.

[1172] Input: Behavioral characteristics model, past operation logs and emotion data

[1173] Output: Next action prediction result

[1174] Step 8:

[1175] The server pre-reads the next required resource and delivers it to the device. Based on the predicted next action, it selects the articles, images, product information, etc. needed for that action and sends them to the device.

[1176] Input: Next action prediction result

[1177] Output: Prefetched resources

[1178] Step 9:

[1179] The device caches the delivered resources and displays them quickly the next time the user performs an action. This reduces loading time and improves the user experience. Specifically, data that has been cached in advance is displayed immediately.

[1180] Input: Prefetched resources

[1181] Output: Cached data, fast display

[1182] The above steps enable highly accurate behavioral prediction and resource preemption using user operation logs and emotional data.

[1183] (Application example 2)

[1184] 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."

[1185] Autonomous vehicle systems are required to predict the next action in real time using occupant operation logs and emotional data, and to provide optimal driving modes and routes while reducing occupant stress. However, conventional systems do not adequately collect and analyze emotional data, making it difficult to respond according to the occupant's emotional state. Therefore, more accurate behavioral prediction, including emotional data, is necessary to provide a comfortable and safe driving experience.

[1186] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user operation logs, means for identifying the user's behavior pattern by analyzing the collected operation logs and the user's emotional data, means for predicting the user's next behavior based on the identified behavior pattern and emotional data, means for prefetching resources corresponding to the predicted next behavior based on the emotional data, means for delivering the prefetched resources to the user terminal, and means for caching the delivered resources in the terminal. This enables behavior prediction that takes the user's emotional state into consideration in real time, making it possible to provide the optimal driving mode and route while reducing stress on passengers.

[1187] A "user operation log" is a series of operation records generated when a user operates a system or terminal.

[1188] "Emotional data" refers to data that indicates the user's emotional state, obtained from facial expressions, voice, heart rate, etc.

[1189] "Behavioral patterns" refer to the behavioral tendencies and characteristics obtained by analyzing a user's series of operation logs and emotional data.

[1190] "Predicting the next action" means predicting the next action the user will take based on past operation logs and emotional data.

[1191] "Pre-fetching resources" means preparing the data and content needed for the user's next predicted action in advance.

[1192] "User terminal" refers to a device or equipment that is directly operated by a user.

[1193] "Delivering" means that the server sends the prefetched resources to the user terminal.

[1194] "Caching on the device" means temporarily storing the delivered resource on the user's device.

[1195] This invention relates to a system that collects and analyzes the operation logs and emotional data of occupants in an autonomous vehicle, predicts their next actions based on the collected data, and provides the occupants with the optimal driving mode and route. This system is configured as follows.

[1196] When a user boards an autonomous vehicle, sensors (such as an operation panel, buttons, and screen) are used to collect operation log data, and an emotion engine (such as a camera, microphone, and heart rate sensor) is used to collect emotional data from the occupant. The operation log data records the operations performed by the occupant and the order in which they were performed. The emotional data also includes the occupant's facial expressions, voice, heart rate, and so on.

[1197] This data is collected in real time and recorded on the terminal (vehicle's computer). The terminal then sends the collected data to a server periodically or when certain conditions are met. The server then stores the data in a database and analyzes it using machine learning algorithms.

[1198] During the analysis process, the server identifies the user's behavioral patterns and emotional trends. Based on this, the server updates the behavioral characteristic model and predicts the next action to be taken. The behavioral characteristic model reflects past operation logs and emotional data, enabling highly accurate predictions. Based on the predicted behavior, resources that take emotional data into account (such as vehicle speed adjustment information, driving mode change instructions, relaxing music, etc.) are prepared in advance and distributed to the device.

[1199] The device caches the delivered resources and can immediately respond to the occupant's next action. For example, if the occupant's stress level rises while driving on the highway, the device can automatically adjust the atmosphere in the vehicle by selecting relaxing music or a gentle driving mode based on emotional data.

[1200] This series of processes enables behavior prediction and response that takes into account the user's emotional state in real time, improving passenger comfort and safety.

[1201] As a specific example, if a state of stress is detected while driving on a highway, the server analyzes this and operates the system to switch to an appropriate driving mode. An example of a prompt sentence to input to the generative AI model in this case could be, "Please explain a system that detects emotional data indicating a state of stress while driving on a highway and suggests driving modes or route changes." Using this prompt sentence, the system can acquire basic data for optimal processing.

[1202] As described above, this invention achieves more accurate behavior prediction and real-time response by integrating and analyzing operation logs and emotion data, thereby providing a comfortable driving experience for occupants of autonomous vehicles.

[1203] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1204] Step 1:

[1205] A user gets into an autonomous vehicle and operates the control panel and buttons, which generates an operation log. At the same time, the emotion engine uses cameras and microphones to sense the user's facial expressions, voice, heart rate, etc., and collects emotional data. The input is the user's operations and emotional state, and the output is the collected operation log and emotional data.

[1206] Step 2:

[1207] The terminal temporarily stores the collected operation log and emotion data and transmits them to the server at regular intervals (for example, in real time or at regular intervals). The input is the collected data, and the output is the data sent to the server. Specific operations include generating data packets and transmitting them via a communication network.

[1208] Step 3:

[1209] The server receives the operation log and emotion data and stores them in a database. Next, it uses machine learning algorithms to analyze this data and identify the user's behavioral patterns and emotional fluctuation trends. The input is the data sent to the server, and the output is the analysis results, which are specific data on behavioral patterns and emotional states. Specific operations include data cleaning, feature extraction, and model application.

[1210] Step 4:

[1211] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data, and contains information for predicting the user's next action. The input is the analysis results, and the output is the updated behavioral characteristic model. Specific operations include learning and updating the model.

[1212] Step 5:

[1213] The server predicts the next action based on the updated behavioral characteristic model. In particular, taking emotional data into consideration enables highly accurate predictions. Based on the prediction, the server looks ahead and prepares the required resources (e.g., driving mode adjustment information, relaxing music, etc.). The input is the updated behavioral characteristic model, and the output is the predicted next action and the required resources. Specific operations include applying the prediction algorithm and selecting resources.

[1214] Step 6:

[1215] The server delivers prefetched resources to the terminal. The terminal receives them and stores them in its cache. The input is the resource delivered from the server, and the output is the resource stored in the terminal's cache. The specific operations involve receiving data and writing it to the cache.

[1216] Step 7:

[1217] When the user takes the next action, the device immediately uses the cached resources and provides appropriate feedback. For example, if the user is feeling stressed, suggestions for relaxing music or driving modes may be displayed. The input is the user's action and the cached resources, and the output is feedback to the user. Specific operations include reading the resources and displaying them on the user interface.

[1218] In this way, each step is performed consecutively, enabling behavioral prediction and resource provision that takes into account the user's emotional state in real time.

[1219] 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.

[1220] 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.

[1221] 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.

[1222] [Fourth embodiment]

[1223] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1224] 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.

[1225] 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).

[1226] 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.

[1227] 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.

[1228] 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).

[1229] 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. 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.

[1230] 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.

[1231] 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.

[1232] 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.

[1233] 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.

[1234] 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.

[1235] 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."

[1236] The present invention relates to a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience.

[1237] Program processing overview

[1238] This system works in cooperation with a server and a terminal. When a user operates an application, the operation log is collected and sent to the server. The server analyzes the collected operation log, identifies behavioral patterns, and predicts the next action based on the identified behavioral patterns. Based on the prediction results, the server distributes the next required resource to the terminal, and the terminal stores it in its cache.

[1239] 1. Collecting and sending operation logs

[1240] The device monitors user operations (clicks, scrolls, inputs, etc.) in real time and records the operation logs internally. The operation logs are sent to the server periodically or when certain conditions are met.

[1241] 2. Analyzing data and identifying behavioral patterns

[1242] The server analyzes the received operation logs to identify user behavior patterns. This analysis uses machine learning algorithms to extract trends, such as the time of day when users tend to use certain functions.

[1243] 3. Behavioral prediction and resource preemption

[1244] The server predicts the next action a user is likely to take based on their behavioral patterns. For example, in the case of a news app, it predicts that a user who frequently views articles in a particular category is likely to click on the next article in that category. The server then pre-fetches the next required resources (articles, images, product information, etc.) and delivers them to the device.

[1245] 4. Resource Delivery and Caching

[1246] The server instructs the device to deliver the predicted resource, which the device stores in a local cache and displays immediately when the user actually needs it.

[1247] Specific examples

[1248] Example 1: News app

[1249] A user opens a news app and browses to a specific category (e.g., sports).

[1250] The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[1251] The server analyzes the received operation log and identifies that the user tends to frequently view articles in the sports category.

[1252] Based on this behavioral pattern, the server predicts that the user is likely to view an article in the sports category next.

[1253] The server pre-reads the latest sports articles and delivers them to the device, which then stores the articles in its cache.

[1254] When a user clicks on an article in the sports category, the device immediately displays the cached article.

[1255] Example 2: Shopping app

[1256] A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[1257] The terminal records these operation logs and periodically sends them to the server.

[1258] The server analyzes the data and determines that the user frequently views products in the cosmetics category.

[1259] The server predicts the user's next action and pre-fetches the next required product data.

[1260] The server delivers pre-fetched product data to the terminal, which then stores the data in its cache.

[1261] When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[1262] In this way, the present invention realizes a system that improves user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and distributing resources, and caching.

[1263] The processing flow will be explained below.

[1264] Step 1:

[1265] The user interacts with the application: within the news app, the user scrolls and clicks to browse different articles or navigate to a particular category.

[1266] Step 2:

[1267] The device logs all user actions (clicks, scrolls, typing, etc.), including timestamps and specific actions.

[1268] Step 3:

[1269] When a device has been in operation for a certain period of time or when certain conditions are met, the collected operation logs are sent to a server, and security measures such as encryption are taken when sending the data.

[1270] Step 4:

[1271] The server stores the received operation logs in a database. The stored data includes user identification information, operation details, timestamps, etc.

[1272] Step 5:

[1273] The server analyzes the operation logs using machine learning algorithms at regular intervals or when new data arrives, which involves extracting user behavior patterns, such as how often a user visits articles in a particular category at a particular time of day.

[1274] Step 6:

[1275] The server updates the user's behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and is used to predict the user's next actions.

[1276] Step 7:

[1277] The server uses the behavioral trait model to predict the next action the user is likely to take, for example determining that the user is likely to view an article in the sports category next.

[1278] Step 8:

[1279] The server determines the next required resource (article, image, product information, etc.) based on the predicted behavior. If the prediction is highly accurate, it issues a command to prefetch that resource.

[1280] Step 9:

[1281] The server delivers prefetched resource data to the device, including specific article content and image files that will be needed next.

[1282] Step 10:

[1283] The terminal caches the received resource data, allowing the resource to be accessed quickly without network delays.

[1284] Step 11:

[1285] A user actually takes the following action: for example, clicks on an article in the sports category.

[1286] Step 12:

[1287] The device will instantly display the cached resources, resulting in a smoother and faster operating experience for users.

[1288] Step 13:

[1289] The server receives the user's real-time operation log again and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[1290] Through the above processing steps, this system predicts user behavior in real time and pre-fetches and caches necessary resources, significantly improving the user experience.

[1291] Example 1

[1292] 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."

[1293] Conventional systems simply collect user operation logs, but lack the ability to predict the next operation or prefetch resources based on the predicted operation. As a result, the next resource the user needs is not provided immediately, resulting in a poor user experience. Furthermore, the lack of consideration for network conditions and real-time environmental data makes efficient data transfer and cache management difficult.

[1294] 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.

[1295] In this invention, the server includes means for transmitting a user's operation log to the server, means for analyzing the received operation log to identify the user's behavior pattern, and means for predicting the user's next behavior based on the identified behavior pattern. This makes it possible to analyze the user's operation log in real time, predict the next required resources, and deliver the predicted resources to the terminal.

[1296] "User" refers to any individual or corporation that uses this system.

[1297] An "operation log" is a record of operations such as clicking, scrolling, and input that occur when a user operates an application.

[1298] "Server" refers to a computer system that analyzes collected operation logs, identifies behavioral patterns, and distributes resources.

[1299] "Terminal" refers to a device operated by a user, and is hardware that collects operation logs and receives and displays resources.

[1300] "Behavioral patterns" refer to behavioral tendencies or characteristics identified based on a series of actions a user has taken in the past.

[1301] "Behavior prediction" refers to estimating the operations a user is likely to perform in the future based on identified behavioral patterns.

[1302] "Resources" refers to data or information that a user might need next (e.g., news articles, product information, images, etc.).

[1303] "Looking ahead" means predicting and preparing the resources a user will need next.

[1304] "Delivery" refers to sending prefetched resources from the server to the terminal.

[1305] "Cache" is a mechanism that temporarily stores distributed resources inside the device and provides them quickly when the user needs them.

[1306] The present invention is a system that collects and analyzes user operation logs to identify behavioral patterns, predict next actions, and ultimately improve the user experience. This system operates in cooperation with a server and terminals.

[1307] When a user operates an application, an operation log is collected and temporarily saved on the device. For example, it is common for Android and iOS devices to save operation logs in an SQLite database. The device sends the operation log to the server based on certain conditions. Specifically, it sends a POST request to the server when the log size exceeds 1MB or every certain time (e.g., every hour).

[1308] The server analyzes the received operation logs and identifies user behavior patterns. This analysis uses machine learning algorithms such as Python's scikit-learn and TensorFlow. Libraries such as Pandas and NumPy are used for data processing. Specifically, the server converts the received operation logs into a Pandas data frame and fits it to a machine learning model to identify user behavior patterns.

[1309] Once a behavioral pattern is identified, the server predicts the next action the user is likely to take. For example, in a news app, if it determines that a user tends to frequently view articles in a particular category (e.g., sports), it predicts that the user is likely to click on an article in that category next.

[1310] The server then prefetches the resources needed based on the predicted behavior. These resources can include news articles, images, product information, etc. The server queries the database to retrieve the next required resource and prepares it. Specifically, the server executes a query to retrieve the latest news article and related image data.

[1311] The server delivers the prepared resources to the device. The device stores the received resources in a local cache. For caching, internal storage is used on Android devices, and NSCache is used on iOS devices. When the user actually needs the resource, the device immediately displays the cached resource.

[1312] Specific examples

[1313] Specific examples of news apps

[1314] 1. A user opens a news app and browses to a specific category (e.g., sports).

[1315] 2. The device records the user's actions such as clicking and scrolling as a log and periodically sends it to the server.

[1316] 3. The server reads the received operation log as a Pandas data frame and analyzes the data.

[1317] 4. Based on the analysis results, the server identifies a tendency for users to frequently view articles in the sports category.

[1318] 5. The server predicts the user's next action and determines that they are likely to view articles in the sports category.

[1319] 6. The server prefetches the next sports category article and image it needs and prepares it.

[1320] 7. The server delivers the prepared article and image data to the terminal.

[1321] 8. The device stores the received articles and images in a local cache.

[1322] 9. When a user clicks on an article in the Sports category, the device immediately displays the cached article.

[1323] Example prompts for specific generative AI models:

[1324] "We provide an operation log showing users frequently viewing news articles in the sports category. Please predict their next action and deliver the necessary resources to the device in advance."

[1325] Example of a shopping app

[1326] 1. A user navigates through a shopping app and frequently browses a particular category (e.g., cosmetics).

[1327] 2. The terminal records these operation logs and periodically sends them to the server.

[1328] 3. Analyze the operation log data received by the server using a Pandas data frame.

[1329] 4. The server clusters user behavior patterns and identifies a tendency to frequently browse products in the cosmetics category.

[1330] 5. The server uses the predictive model to predict which cosmetic product is likely to be viewed next.

[1331] 6. The server queries the database to retrieve the next required product data and prepares it.

[1332] 7. The server delivers the prepared product data to the terminal.

[1333] 8. The device caches the received product data in its internal storage.

[1334] 9. When a user browses for a product in the cosmetics category, the device immediately displays the cached product data.

[1335] Example prompts for specific generative AI models:

[1336] "We will provide an operation log of users frequently browsing products in the cosmetics category. Please predict their next action, pre-read the necessary product data, and deliver it to the device."

[1337] In this way, this system can improve the user experience by collecting operation logs, analyzing behavioral patterns, predicting behavior, pre-reading and delivering resources, and caching.

[1338] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1339] Step 1: Collecting operational logs

[1340] The device monitors user operations (clicks, scrolling, input, etc.) in real time and records the operation log in its internal storage. Specifically, the device records the user's tap and scroll position and time.

[1341] Input: User actions (clicks, scrolling, typing, etc.)

[1342] Data processing: Log location, time, and type of operation

[1343] Output: Operation log saved in internal storage

[1344] Step 2: Send the operation log

[1345] The terminal sends the operation log to the server at regular intervals or when certain conditions are met, for example, when the log size exceeds 1MB.

[1346] Input: Operation log saved in internal storage

[1347] Data processing: Converts operation logs into a format that can be sent to the server

[1348] Output: Operation log sent to the server

[1349] Step 3: Analyze the data

[1350] The server analyzes the received operation logs using machine learning algorithms such as Python's scikit-learn and TensorFlow.

[1351] Input: Operation log sent to the server

[1352] Data processing: Convert to a Pandas data frame and fit to a machine learning model

[1353] Output: Analysis results (user behavior patterns)

[1354] Step 4: Identify patterns of behavior

[1355] The server uses the analysis results to identify user behavior patterns, such as the tendency to use certain functions at certain times of the day.

[1356] Input: Analysis results (user behavior patterns)

[1357] Data processing: Identifying behavioral patterns using clustering and pattern mining techniques

[1358] Output: Identified behavioral patterns

[1359] Step 5: Predict behavior

[1360] Based on the identified behavioral patterns, the server predicts the user's likely next action.

[1361] Input: Identified behavioral patterns

[1362] Data processing: Using predictive models to estimate the next steps

[1363] Output: Predicted next action

[1364] Step 6: Prefetching resources

[1365] Based on predicted behavior, the server pre-fetches resources that will be needed next, including, for example, news articles or product information that are likely to be viewed next.

[1366] Input: Predicted next action

[1367] Data processing: Query the relevant database to obtain the required resources

[1368] Output: Prefetched resources (news articles, product information, etc.)

[1369] Step 7: Deliver resources

[1370] The server delivers the prefetched resources to the terminal.

[1371] Input: Prefetched resources

[1372] Data processing: Convert resources into a distribution format and send them to the terminal

[1373] Output: Resources delivered to the device

[1374] Step 8: Saving the cache

[1375] The device stores the received resource in a local cache and displays it immediately when the user actually needs it.

[1376] Input: Resources delivered from the server

[1377] Data processing: Save resources to internal storage or NSCache

[1378] Output: Cached resources

[1379] Through these steps, user operation logs are collected, analyzed, and the next action is predicted, and necessary resources are pre-loaded and delivered quickly, improving the user experience.

[1380] (Application example 1)

[1381] 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."

[1382] Existing user experience improvement systems collect user operation logs and identify behavioral patterns to predict the next action and prefetch resources. However, there is a lack of systems that ensure immediate responsiveness to user actions while also having extensive caching capabilities. Furthermore, in scenarios such as virtual stores, optimization methods are required to instantly display the products and information that the user will be interested in next. Conventional systems lack sufficient resource prediction and caching capabilities for identified actions, making it difficult to provide end users with prompt services. Furthermore, the lack of optimization based on network conditions and user environments contributes to a decline in the quality of the user experience.

[1383] 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.

[1384] In this invention, the server includes means for collecting user operation logs, means for analyzing the collected operation logs to identify the user's behavioral patterns, means for predicting the user's next action based on the identified behavioral patterns, means for prefetching resources corresponding to the user's next action, means for delivering the prefetched resources to the user terminal, means for caching the delivered resources in the terminal, and means for instantly displaying the retained resources on the interface. This makes it possible to instantly provide products and information that the user will be interested in next, improving the shopping experience in virtual stores, etc. Furthermore, by providing optimized resources taking into account the user's network status, the quality of the user experience can be improved.

[1385] A "user operation log" is data that records the operation history of a user, such as clicking, scrolling, and inputting, when the user operates an application.

[1386] A "behavioral pattern" is a tendency or tendency for a particular operation or behavior to be repeated, extracted by analyzing a user's operation log.

[1387] "Next Action" refers to the action or choice a user is likely to take next, predicted based on identified behavioral patterns.

[1388] "Resources" refers to data or information that corresponds to the user's next action, specifically articles, images, product information, etc.

[1389] "Predictive load" means preparing and delivering the resources needed in advance for the next predicted action, so that they are available immediately when the user requests them.

[1390] "Cache" refers to a memory area or the act of temporarily storing data that has been retrieved, and is used to quickly provide that data when the user needs it again.

[1391] "Immediate display on the interface" means that cached resources are instantly displayed on the display screen or application in response to user actions.

[1392] "Network status" refers to the communication status between the user terminal and the server, specifically information about network performance such as communication speed, latency, and bandwidth.

[1393] "Optimization measures" are mechanisms for selecting and implementing optimal methods and techniques to improve user experience based on the user's network conditions and behavioral patterns.

[1394] "Predicted resources" are data or information that are prepared in advance by predicting the user's next action.

[1395] "Retained resources" refers to data or information that is cached on the device through prediction or prefetching, and is immediately available when needed.

[1396] This invention relates to a system that collects user operation logs, analyzes them to identify behavioral patterns, predicts future actions, and ultimately improves the user experience. This system mainly operates in cooperation with a server and a terminal. The server is responsible for analyzing operation logs, identifying behavioral patterns, predicting actions, and pre-fetching and distributing resources, while the terminal is responsible for recording, caching, and immediately displaying the collected operation logs.

[1397] The system works as follows: First, when a user operates an application, the device monitors the operation (clicking, scrolling, input, etc.) in real time and records the operation log internally. The collected operation log is then periodically sent to the server.

[1398] The server then analyzes the received operation logs using machine learning algorithms to identify user behavior patterns, such as the time of day when users tend to use certain features or frequently browse products in a particular category.

[1399] Next, based on the behavioral patterns, the server predicts the user's likely next action. Based on the prediction results, the server pre-fetches the next required resource (such as product information or article data) and delivers it to the device. The delivered resource is cached on the device and is displayed immediately when the user actually needs it.

[1400] As a concrete example, consider a shopping experience in a virtual store. When a user puts on a head-mounted display (HMD) and enters the virtual store, an operation log is recorded in real time. When the user browses multiple products (shoes, shirts, etc.), the operation log is sent to a server, which analyzes it and predicts that the user is likely to browse "pants" next. The server then pre-fetches the product information for "pants," which will be needed next, delivers it to the HMD, and stores it in a cache. When the user moves to the "pants" section, the cached product information is immediately displayed, providing a fast shopping experience.

[1401] (Example of a prompt)

[1402] User records: [view_shoes, view_shirt]

[1403] Next action prediction: recommend_pants

[1404] Cached resources: pants information

[1405] The system uses hardware and software such as a head-mounted display (HMD), a program implemented in Python, data exchange in JSON format, and the requests library for HTTP requests.

[1406] In this way, the present invention realizes a system that improves user experience by combining analysis of collected operation logs, identification of behavioral patterns, behavior prediction, resource prefetching and distribution, and caching functions.

[1407] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1408] Step 1:

[1409] The user interacts with the application.

[1410] At this time, the device monitors and records in real time the user's operations, such as clicking, scrolling, and input. This recorded operation log becomes a dataset that will be used for subsequent analysis.

[1411] Input: User actions

[1412] Output: Operation log recorded internally on the device

[1413] Step 2:

[1414] The recorded operation log is sent to the server periodically or when certain conditions are met.

[1415] The terminal sends operation logs to the server at specified intervals or based on specific trigger conditions, using HTTP requests to transfer data in JSON format.

[1416] Input: Operation log (internal to the terminal)

[1417] Output: Operation log (sent to server)

[1418] Step 3:

[1419] The server analyzes the received operation logs and identifies the user's behavior patterns.

[1420] The server inputs the operation logs into a machine learning algorithm to extract user operation trends. This analysis reveals behavioral patterns, such as frequent use of certain functions at certain times of the day.

[1421] Input: Operation log (data sent to the server)

[1422] Output: Identified behavioral patterns

[1423] Step 4:

[1424] The server predicts the user's next action based on the identified behavioral patterns.

[1425] Based on machine learning models, the server predicts what the user is likely to do next, and this prediction determines the next resource required.

[1426] Input: Behavioral pattern analysis results

[1427] Output: Predicted next action

[1428] Step 5:

[1429] The server pre-fetches the required resources based on the predicted next action and delivers them to the terminal.

[1430] The prefetched resources (e.g., product information, article data, etc.) are sent to the terminal via an HTTP request.

[1431] Input: Predicted next action

[1432] Output: Prefetched resources (delivered to device)

[1433] Step 6:

[1434] The device caches the delivered resources.

[1435] The device caches the received resources in local storage, making them immediately available when the user needs them.

[1436] Input: Prefetched resources

[1437] Output: Cached resources

[1438] Step 7:

[1439] The device will immediately display the cached resource when the user next performs an action.

[1440] For example, when a user browses for "pants" in a virtual store, the cached product information for "pants" is immediately displayed, providing a fast browsing experience for the user.

[1441] Input: User's next action

[1442] Output: View cached resources

[1443] In this way, the system of the present invention provides end users with fast and efficient services through a series of processes, from analyzing collected operation logs to predicting behavior, pre-reading and distributing resources, caching, and displaying them immediately. This entire process significantly improves the user experience.

[1444] 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.

[1445] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[1446] Program processing overview

[1447] This system works in cooperation with a server, a device, and an emotion engine. When a user operates an application, an operation log is collected, along with the user's emotional data. This data is sent to the server, where it is analyzed. The server identifies the user's behavioral patterns, predicts the user's next action, and pre-reads and delivers the next required resource to the device. The device stores this data in a cache and responds immediately to the user's operations.

[1448] 1. Collection and transmission of operation logs and emotion data

[1449] When a user operates an application, the operation details are recorded as an operation log. In addition, the emotion engine analyzes the user's facial expressions and voice to obtain emotional data. This is done in real time.

[1450] 2. Data collection and transmission

[1451] The device collects the user's operation log and emotional data and sends it to the server periodically or when certain conditions are met.

[1452] 3. Analyzing data and identifying behavioral patterns

[1453] The server stores the received operation logs and emotion data in a database and analyzes them using machine learning algorithms, thereby identifying user behavior patterns and emotional trends.

[1454] 4. Behavioral prediction and emotion-based compensation

[1455] The server updates the behavioral model based on the analysis results. The behavioral model reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Emotional data is an important factor in correcting behavioral predictions.

[1456] 5. Resource Prefetching and Delivery

[1457] The server uses a behavioral characteristic model to predict the user's next action, and pre-fetches and delivers the required resources (articles, images, product information, etc.) to the device.

[1458] 6. Resource caching and fast display

[1459] The device can cache the received resources and display them immediately when the user takes their next action.

[1460] Specific examples

[1461] Example 1: News app

[1462] 1. A user interacts with a news app and frequently browses political news. Facial expression analysis detects that the user is feeling stressed while reading a specific article.

[1463] 2. The device records these operation logs and facial expression data and periodically transmits them to the server.

[1464] 3. The server analyzes the operation log to identify patterns in which users frequently view political news, and at the same time, detects content that is likely to cause stress to users.

[1465] 4. Based on the analysis results, the server predicts the next news article that is likely to be displayed and predicts content that takes the user's emotions into consideration.

[1466] 5. The server delivers the next required resource to the device, which stores it in its cache.

[1467] 6. The next time a user clicks on a political news item, the device will instantly display the cached article, along with relaxing content if the article is expected to cause stress.

[1468] Example 2: Shopping app

[1469] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative emotions are detected when a user cancels a purchase just before it is due to be made.

[1470] 2. The device records these operation logs and emotion data and sends them to the server.

[1471] 3. The server analyzes the data and determines that users frequently browse the cosmetics category and tend to be hesitant to make a purchase.

[1472] 4. The server uses the behavioral characteristic model to predict the next required product data and resources to increase the user's purchasing intent.

[1473] 5. The server delivers this data to the device, which stores it in its cache.

[1474] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[1475] In this way, the present invention provides a system that uses operation logs and emotion data to achieve more accurate behavior prediction and resource preemption, thereby improving the user experience.

[1476] The processing flow will be explained below.

[1477] Step 1:

[1478] The user interacts with the application: The user opens the news app, scrolls, and clicks to read a particular article.

[1479] Step 2:

[1480] The device records user operations (clicks, scrolls, timestamps, etc.) as a log, while also collecting facial expression and voice data using the device's built-in camera and microphone.

[1481] Step 3:

[1482] The emotion engine analyzes the collected facial and voice data to recognize the user's emotions. The emotion recognition results (e.g., "excitement," "stress," "joy," etc.) are recorded along with the operation log.

[1483] Step 4:

[1484] The terminal transmits operation logs and emotion data to the server for a certain period of time or under certain conditions (for example, a specific time or page transition).

[1485] Step 5:

[1486] The server stores the received operation logs and emotion data in a database, including the user's identification information, operation details, timestamps, and emotion recognition results.

[1487] Step 6:

[1488] The server uses machine learning algorithms to analyze past operation logs and emotional data stored in a database, thereby identifying user behavior patterns and emotional trends.

[1489] Step 7:

[1490] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data and is used to predict the next action.

[1491] Step 8:

[1492] The server uses the behavioral trait model to predict what a user is likely to do next, for example determining that they are likely to view an article in a particular category next.

[1493] Step 9:

[1494] The server identifies the required resources (specific articles, images, product information, etc.) based on predicted behavior. Based on the emotional data, the server adjusts the resources according to the user's current emotional state. For example, if the emotional state is positive, relevant articles will be selected, and if the emotional state is negative, stress-reducing content will be selected.

[1495] Step 10:

[1496] The server delivers the predicted resource data to the device, including the specific article content and image files that will be needed next.

[1497] Step 11:

[1498] The device caches the received resource data, allowing it to be accessed quickly the next time the user takes action.

[1499] Step 12:

[1500] The user actually takes the following action: for example, clicks on an article in a particular category.

[1501] Step 13:

[1502] The device will instantly display cached resources, providing a better user experience by presenting content that takes into account the user's emotional state.

[1503] Step 14:

[1504] The server receives the user's real-time operation log and emotion data again, and starts the next analysis cycle based on the operation log, thus continuously carrying out the dynamic optimization process.

[1505] Through the above processing steps, the system predicts user behavior and emotions in real time, pre-fetching and caching necessary resources, thereby significantly improving the user experience.

[1506] Example 2

[1507] 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."

[1508] Conventional user behavior prediction systems identify behavioral patterns based on user operation logs and predict the user's next actions. However, the behavioral predictions obtained in this way are often inaccurate because they do not take into account the user's internal emotional state. In particular, when a user's psychological state significantly influences their behavior, conventional systems are unable to properly capture these fluctuations, and an improvement in the user experience cannot be expected. For this reason, more accurate behavior predictions and the provision of appropriate resources are needed.

[1509] 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.

[1510] In this invention, the server includes means for collecting user operation logs, means for acquiring user emotion data in real time, and means for periodically transmitting the collected operation logs and emotion data to the server, thereby enabling behavior prediction that takes into account both the user operation logs and emotion data, thereby improving the user experience.

[1511] A "user operation log" is data that records information such as a timestamp, operation content, and operation target when a user operates an application.

[1512] "Emotion data" is data that includes emotion labels such as joy, sadness, anger, and surprise, and their intensities, obtained by analyzing the user's facial expressions and voice.

[1513] The "server" is a central processing unit that receives, stores, and analyzes data sent from the terminal to identify trends in user behavior patterns and emotional fluctuations.

[1514] A "terminal" is a device that is directly operated by the user and is responsible for collecting operation logs and emotional data and transmitting them to a server.

[1515] A "machine learning algorithm" is a computational method for analyzing large data sets and identifying patterns, a technique used to build models of user behavioral characteristics.

[1516] A "behavioral characteristic model" is a mathematical model that learns a user's past operation patterns and emotional data and predicts future behavior.

[1517] "Resources" refers to digital content such as articles, images, and product information that are needed for the user's next action.

[1518] A "cache" is a high-speed storage area that stores frequently accessed data in advance, and is used to respond quickly to user operations.

[1519] "Preprocessing" is the process of preparing data for analysis, and includes data cleansing and filtering.

[1520] "User emotion fluctuation tendency" is an index that shows how a user's emotions change over time and depending on specific operations.

[1521] This invention relates to a system that collects and analyzes user operation logs and emotion data to identify behavioral patterns and predict next actions. This system achieves more accurate behavioral prediction by combining it with an emotion engine that recognizes the user's emotions.

[1522] The system consists of a device operated by the user, a server that analyzes data, and an emotion engine that acquires emotional data. The device is a user device such as a smartphone, tablet, or PC, and the emotion engine is a software module that uses facial expression and voice recognition. The server includes a cloud server and a dedicated database server for large-scale data processing.

[1523] When a user operates the application, the operation details are recorded as an operation log with a timestamp. In addition, the emotion engine analyzes the user's facial expressions and voice in real time and records them as emotion data. This data is sent to the server periodically or when a specific trigger event occurs.

[1524] The server accumulates the received operation logs and emotion data in a data store and saves it in a database. Next, preprocessing such as data cleansing and filtering is performed, and machine learning algorithms are applied. Specifically, a time series analysis model (e.g., LSTM) is used to identify behavioral patterns, and a natural language processing model (e.g., BERT) is used to analyze emotion data.

[1525] The behavioral model built based on the analysis results reflects the user's past operation patterns and emotional data and is used to predict the next action to be taken. Using this behavioral model, the server prefetches the required resources (such as news articles or product information that are likely to be viewed next) and sends them to the device.

[1526] The device can cache the received resources and respond quickly to the user's next operation, thereby shortening loading times and improving the user experience.

[1527] Specific examples

[1528] Example 1: News app

[1529] 1. A user frequently browses political news on a news app. The timestamp and operation details are recorded as an operation log, and stress is detected by facial expression analysis.

[1530] 2. The device periodically sends operation logs and emotion data to the server. The trigger conditions for sending are a certain viewing time or number of operations.

[1531] 3. The server accumulates this data, performs preprocessing, and then analyzes it using a machine learning algorithm. From past data, the system learns behavioral patterns when the user feels stressed.

[1532] 4. The server predicts which article is likely to be displayed next and adjusts it based on sentiment data, including selecting relaxing content to reduce stress.

[1533] 5. The server pre-fetches the required resources and delivers them to the device.

[1534] 6. The device will cache it and display it immediately the next time the user clicks on a political news item, resulting in faster loading times.

[1535] Example 2: Shopping app

[1536] 1. A user interacts with a shopping app and frequently browses products in the cosmetics category. Negative sentiment is detected when a user cancels a purchase just before it is due to be made.

[1537] 2. The device records the operation log and emotion data and sends them to the server.

[1538] 3. The server stores this data, preprocesses it, and then analyzes it using machine learning algorithms to identify trends that could cause users to hesitate before making a purchase.

[1539] 4. The server uses the behavioral characteristic model to predict the next product data needed and resources that will increase purchasing motivation.

[1540] 5. The server delivers this data to the device, which stores it in its cache.

[1541] 6. When the user views products in the cosmetics category again, the device quickly displays the cached product data and offers incentives (e.g., special offers and reviews) to encourage purchases.

[1542] Example prompts to input to the generative AI model

[1543] "Please provide us with the operation logs and sentiment data necessary to predict the next user action."

[1544] In this way, the present invention provides a system that uses operation logs and emotion data to predict user behavior and predict optimal resources, thereby improving the user experience.

[1545] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1546] Step 1:

[1547] When a user interacts with an application, the details of the interaction are recorded in an operation log along with a timestamp. Specifically, every time a user clicks a button or navigates to a different page, the interaction is saved as a log. This operation log is part of the data used for subsequent analysis.

[1548] Input: User actions (clicks, page transitions, etc.)

[1549] Output: Operation log (timestamp and operation details)

[1550] Step 2:

[1551] The device analyzes the user's facial expressions and voice in real time using an emotion engine to obtain emotional data. The emotion engine is a software module that captures the user's facial expressions and voice using a camera and microphone and analyzes the data. The analysis results are output as a label and intensity value that indicates the user's emotional state.

[1552] Input: User's facial expressions, voice

[1553] Output: Emotion data (emotion labels and their intensities)

[1554] Step 3:

[1555] The operation log and emotion data collected by the device are sent to a server periodically or according to trigger conditions. For example, data is compiled and sent after a certain number of operations or at certain time intervals. The data sent to the server is used for subsequent analysis.

[1556] Input: Operation log, emotion data

[1557] Output: The dataset sent to the server

[1558] Step 4:

[1559] The server accumulates the received operation logs and emotion data in a large-scale data store and saves it in a database. This data is necessary for later high-speed query execution. Specifically, Amazon S3 or Google Cloud Storage can be used.

[1560] Input: The dataset sent to the server

[1561] Output: Accumulated data

[1562] Step 5:

[1563] The server preprocesses the accumulated data. Preprocessing involves cleansing and filtering the data to remove incomplete or duplicate data. It also normalizes and transforms the data, preparing it for machine learning algorithms to work properly.

[1564] Input: Accumulated data

[1565] Output: Preprocessed data

[1566] Step 6:

[1567] The server runs machine learning algorithms on the preprocessed data, such as using an LSTM model to analyze behavioral patterns and a BERT model to analyze sentiment data, to build a model of user behavioral characteristics.

[1568] Input: Preprocessed data

[1569] Output: Behavioral characteristics model

[1570] Step 7:

[1571] The server uses the constructed behavioral characteristic model to predict the user's next action. This prediction is made using past operation logs and emotional data. For example, it predicts what operation the user will perform next based on past patterns of performing specific operations while in a specific emotional state.

[1572] Input: Behavioral characteristics model, past operation logs and emotion data

[1573] Output: Next action prediction result

[1574] Step 8:

[1575] The server pre-reads the next required resource and delivers it to the device. Based on the predicted next action, it selects the articles, images, product information, etc. needed for that action and sends them to the device.

[1576] Input: Next action prediction result

[1577] Output: Prefetched resources

[1578] Step 9:

[1579] The device caches the delivered resources and displays them quickly the next time the user performs an action. This reduces loading time and improves the user experience. Specifically, data that has been cached in advance is displayed immediately.

[1580] Input: Prefetched resources

[1581] Output: Cached data, fast display

[1582] The above steps enable highly accurate behavioral prediction and resource preemption using user operation logs and emotional data.

[1583] (Application example 2)

[1584] 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."

[1585] Autonomous vehicle systems are required to predict the next action in real time using occupant operation logs and emotional data, and to provide optimal driving modes and routes while reducing occupant stress. However, conventional systems do not adequately collect and analyze emotional data, making it difficult to respond according to the occupant's emotional state. Therefore, more accurate behavioral prediction, including emotional data, is necessary to provide a comfortable and safe driving experience.

[1586] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user operation logs, means for identifying the user's behavior pattern by analyzing the collected operation logs and the user's emotional data, means for predicting the user's next behavior based on the identified behavior pattern and emotional data, means for prefetching resources corresponding to the predicted next behavior based on the emotional data, means for delivering the prefetched resources to the user terminal, and means for caching the delivered resources in the terminal. This enables behavior prediction that takes the user's emotional state into consideration in real time, making it possible to provide the optimal driving mode and route while reducing stress on passengers.

[1587] A "user operation log" is a series of operation records generated when a user operates a system or terminal.

[1588] "Emotional data" refers to data that indicates the user's emotional state, obtained from facial expressions, voice, heart rate, etc.

[1589] "Behavioral patterns" refer to the behavioral tendencies and characteristics obtained by analyzing a user's series of operation logs and emotional data.

[1590] "Predicting the next action" means predicting the next action the user will take based on past operation logs and emotional data.

[1591] "Pre-fetching resources" means preparing the data and content needed for the user's next predicted action in advance.

[1592] "User terminal" refers to a device or equipment that is directly operated by a user.

[1593] "Delivering" means that the server sends the prefetched resources to the user terminal.

[1594] "Caching on the device" means temporarily storing the delivered resource on the user's device.

[1595] This invention relates to a system that collects and analyzes the operation logs and emotional data of occupants in an autonomous vehicle, predicts their next actions based on the collected data, and provides the occupants with the optimal driving mode and route. This system is configured as follows.

[1596] When a user boards an autonomous vehicle, sensors (such as an operation panel, buttons, and screen) are used to collect operation log data, and an emotion engine (such as a camera, microphone, and heart rate sensor) is used to collect emotional data from the occupant. The operation log data records the operations performed by the occupant and the order in which they were performed. The emotional data also includes the occupant's facial expressions, voice, heart rate, and so on.

[1597] This data is collected in real time and recorded on the terminal (vehicle's computer). The terminal then sends the collected data to a server periodically or when certain conditions are met. The server then stores the data in a database and analyzes it using machine learning algorithms.

[1598] During the analysis process, the server identifies the user's behavioral patterns and emotional trends. Based on this, the server updates the behavioral characteristic model and predicts the next action to be taken. The behavioral characteristic model reflects past operation logs and emotional data, enabling highly accurate predictions. Based on the predicted behavior, resources that take emotional data into account (such as vehicle speed adjustment information, driving mode change instructions, relaxing music, etc.) are prepared in advance and distributed to the device.

[1599] The device caches the delivered resources and can immediately respond to the occupant's next action. For example, if the occupant's stress level rises while driving on the highway, the device can automatically adjust the atmosphere in the vehicle by selecting relaxing music or a gentle driving mode based on emotional data.

[1600] This series of processes enables behavior prediction and response that takes into account the user's emotional state in real time, improving passenger comfort and safety.

[1601] As a specific example, if a state of stress is detected while driving on a highway, the server analyzes this and operates the system to switch to an appropriate driving mode. An example of a prompt sentence to input to the generative AI model in this case could be, "Please explain a system that detects emotional data indicating a state of stress while driving on a highway and suggests driving modes or route changes." Using this prompt sentence, the system can acquire basic data for optimal processing.

[1602] As described above, this invention achieves more accurate behavior prediction and real-time response by integrating and analyzing operation logs and emotion data, thereby providing a comfortable driving experience for occupants of autonomous vehicles.

[1603] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1604] Step 1:

[1605] A user gets into an autonomous vehicle and operates the control panel and buttons, which generates an operation log. At the same time, the emotion engine uses cameras and microphones to sense the user's facial expressions, voice, heart rate, etc., and collects emotional data. The input is the user's operations and emotional state, and the output is the collected operation log and emotional data.

[1606] Step 2:

[1607] The terminal temporarily stores the collected operation log and emotion data and transmits them to the server at regular intervals (for example, in real time or at regular intervals). The input is the collected data, and the output is the data sent to the server. Specific operations include generating data packets and transmitting them via a communication network.

[1608] Step 3:

[1609] The server receives the operation log and emotion data and stores them in a database. Next, it uses machine learning algorithms to analyze this data and identify the user's behavioral patterns and emotional fluctuation trends. The input is the data sent to the server, and the output is the analysis results, which are specific data on behavioral patterns and emotional states. Specific operations include data cleaning, feature extraction, and model application.

[1610] Step 4:

[1611] The server updates the behavioral characteristic model based on the analysis results. This model reflects the user's past operation patterns and emotional data, and contains information for predicting the user's next action. The input is the analysis results, and the output is the updated behavioral characteristic model. Specific operations include learning and updating the model.

[1612] Step 5:

[1613] The server predicts the next action based on the updated behavioral characteristic model. In particular, taking emotional data into consideration enables highly accurate predictions. Based on the prediction, the server looks ahead and prepares the required resources (e.g., driving mode adjustment information, relaxing music, etc.). The input is the updated behavioral characteristic model, and the output is the predicted next action and the required resources. Specific operations include applying the prediction algorithm and selecting resources.

[1614] Step 6:

[1615] The server delivers prefetched resources to the terminal. The terminal receives them and stores them in its cache. The input is the resource delivered from the server, and the output is the resource stored in the terminal's cache. The specific operations involve receiving data and writing it to the cache.

[1616] Step 7:

[1617] When the user takes the next action, the device immediately uses the cached resources and provides appropriate feedback. For example, if the user is feeling stressed, suggestions for relaxing music or driving modes may be displayed. The input is the user's action and the cached resources, and the output is feedback to the user. Specific operations include reading the resources and displaying them on the user interface.

[1618] In this way, each step is performed consecutively, enabling behavioral prediction and resource provision that takes into account the user's emotional state in real time.

[1619] 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.

[1620] 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.

[1621] 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.

[1622] 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.

[1623] 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.

[1624] 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.

[1625] 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).

[1626] 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.

[1627] 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."

[1628] 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.

[1629] 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).

[1630] 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.

[1631] 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.

[1632] 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.

[1633] 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.

[1634] 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.

[1635] 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.

[1636] 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.

[1637] 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.

[1638] 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.

[1639] 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.

[1640] The following is further disclosed regarding the above embodiment.

[1641] (Claim 1)

[1642] A means for collecting user operation logs;

[1643] A means of analyzing the collected operation logs to identify user behavior patterns;

[1644] a means for predicting a user's next action based on the identified behavioral patterns;

[1645] A means of pre-fetching application resources that correspond to the user's next action; and

[1646] a means for delivering the prefetched resources to the user device;

[1647] means for caching the delivered resources on the device;

[1648] A system including:

[1649] (Claim 2)

[1650] 2. The system of claim 1, further comprising: means for determining an optimization measure based on the analyzed behavioral patterns and taking into account the user's network status.

[1651] (Claim 3)

[1652] The system according to claim 1, further comprising means for collecting user operation logs and environmental data in real time to create a data set for analysis.

[1653] "Example 1"

[1654] (Claim 1)

[1655] A means for collecting user operation logs;

[1656] A means for transmitting the collected operation logs to a server;

[1657] A means for analyzing the received operation log to identify user behavior patterns;

[1658] a means for predicting a user's next action based on the identified behavioral patterns;

[1659] A means for pre-reading data corresponding to the user's next action;

[1660] A means for delivering the prefetched data to a user terminal;

[1661] A means for caching the delivered data on the terminal;

[1662] A system including:

[1663] (Claim 2)

[1664] 2. The system of claim 1, further comprising: means for determining an optimization measure based on the analyzed behavioral patterns and taking into account the user's network status.

[1665] (Claim 3)

[1666] The system according to claim 1, further comprising means for collecting user operation logs and environmental data in real time to create a data set for analysis.

[1667] "Application Example 1"

[1668] (Claim 1)

[1669] A means for collecting user operation logs;

[1670] A means of analyzing the collected operation logs to identify user behavior patterns;

[1671] a means for predicting a user's next action based on the identified behavioral patterns;

[1672] A means of prefetching resources that correspond to the user's next action; and

[1673] a means for delivering the prefetched resources to the user device;

[1674] means for caching the delivered resources on the device;

[1675] a means for immediately displaying the retained resources in the interface;

[1676] A system including:

[1677] (Claim 2)

[1678] The system of claim 1, further comprising: means for determining an optimization measure based on the analyzed behavioral pattern and taking into account the user's network status; and means for including the next required information in the resource.

[1679] (Claim 3)

[1680] The system of claim 1, further comprising means for collecting user operation logs and environmental data in real time to create a dataset for analysis, and means for maintaining consistency between predicted resources and cached resources.

[1681] "Example 2: Combining Emotion Engines"

[1682] (Claim 1)

[1683] A means for collecting user operation logs;

[1684] A means of obtaining user emotional data in real time;

[1685] a means for periodically transmitting the collected operation log and emotion data to a server;

[1686] a means for storing the received operation log and emotion data;

[1687] A means for preprocessing the accumulated data and applying machine learning algorithms to identify user behavior patterns and sentiment trends;

[1688] a means for predicting a user's next action based on the identified behavioral patterns and emotional data;

[1689] A means for pre-reading required resources and delivering them to the user device;

[1690] means for caching the delivered resources on the device;

[1691] A system including:

[1692] (Claim 2)

[1693] 10. The system of claim 1, further comprising: means for determining an optimization measure based on the analyzed behavioral patterns and emotion data, taking into account the user's network status.

[1694] (Claim 3)

[1695] The system of claim 1, further comprising means for collecting user operation logs and emotion data in real time to create a data set for analysis.

[1696] "Application example 2 when combining emotion engines"

[1697] (Claim 1)

[1698] A means for collecting user operation logs;

[1699] A means for identifying user behavior patterns by analyzing the collected operation logs and user emotion data;

[1700] a means for predicting a user's next action based on the identified behavioral patterns and emotional data;

[1701] a means for pre-reading resources corresponding to a next action predicted based on emotion data;

[1702] a means for delivering the prefetched resources to the user device;

[1703] means for caching the delivered resources on the device;

[1704] A system including:

[1705] (Claim 2)

[1706] 10. The system of claim 1, further comprising: means for determining an optimization measure based on the analyzed behavioral patterns and emotion data, taking into account the user's network status.

[1707] (Claim 3)

[1708] The system according to claim 1, further comprising means for collecting user operation logs, emotional data, and environmental data in real time to create a data set for analysis. [Explanation of symbols]

[1709] 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 user operation logs; A means of analyzing the collected operation logs to identify user behavior patterns; a means for predicting a user's next action based on the identified behavioral patterns; A means of pre-fetching application resources that correspond to the user's next action; and a means for delivering the prefetched resources to the user device; means for caching the delivered resources on the device; A system including:

2. The system according to claim 1 , further comprising: means for determining an optimization measure based on the analyzed behavioral patterns and taking into account the user's network status.

3. The system according to claim 1 , further comprising means for collecting user operation logs and environmental data in real time to create a data set for analysis.

Images