system

Abstract

We provide the system. [Solution] A means for collecting and storing information on a user's past behavior, Means for preprocessing the stored information and converting it into an analyzable format, A means for constructing a learning model to predict the user's future actions based on the converted information, A means for generating an optimal suggestion in accordance with the actions predicted by the aforementioned learning model, A means for notifying the user of the generated proposal, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventionally, there have been systems that aim to smooth the purchase process and reduce the cancellation rate in e-commerce sites and subscription services by making appropriate proposals based on the past actions of users. However, in conventional systems, the response to users is uniform, and it has been difficult to automatically make specific and effective proposals according to the needs of individual users. Also, although it has been required to quickly and accurately predict the behavior patterns of users and provide appropriate information in a timely manner, the conventional technologies have not been able to efficiently achieve this.

Means for Solving the Problems

[0005] By providing a means to collect and store information on users' past behavior, and to preprocess this information into an analyzable format, efficient data utilization becomes possible. Furthermore, the system provides a means to build a learning model for predicting users' future behavior based on this information, and to generate optimal suggestions based on the user's behavior patterns and predicted actions. In addition, by including a means to notify the user of these suggestions, it becomes possible to quickly implement specific responses tailored to individual needs, dramatically improving the user experience.

[0006] "User past behavior information" refers to data about various actions a user has taken in the past, including purchase history, website visit history, click history, and inquiry details.

[0007] "Storage" refers to the process of saving data so that it can be easily accessed later, and this involves using databases and storage systems.

[0008] "Preprocessing" refers to the preparatory stage for converting raw data into a format that can be efficiently used by machine learning models, and includes processes such as data cleansing and formatting standardization.

[0009] A "learning model" refers to an algorithm or statistical framework that learns a user's behavioral patterns and uses that to predict the next action they are likely to take.

[0010] An "optimal suggestion" refers to an item that is generated based on the user's predicted behavior and is a recommendation or offer of the most beneficial and effective product, service, or information for the user.

[0011] "Notification" is the process of communicating information from a system to a user, and this includes forms such as email, push notifications, and pop-ups on websites. [Brief explanation of the drawing]

[0012] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0013] An example of an embodiment of the system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0016] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0018] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0020] [First Embodiment]

[0021] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0022] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0023] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0025] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0026] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0027] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0028] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0033] This invention relates to a system that predicts future user behavior based on past user behavior information and provides users with optimal suggestions. This system consists of multiple servers and user terminals and aims to improve the user experience.

[0034] First, the server stores the user's past behavior information in a database. This includes purchase history, click history, and inquiry details. Next, the server preprocesses the stored data, performing data cleansing and formatting standardization to convert it into a format suitable for analysis.

[0035] The server trains a learning model using machine learning techniques based on pre-processed data. This model is designed to analyze user behavior patterns and predict their next actions. For example, in an e-commerce scenario, the server generates personalized product suggestions based on the user's purchase history, and in a subscription service scenario, it prepares special offers for users at risk of churn.

[0036] When suggestions are generated, they are notified to the user's device. When the user visits an e-commerce site, the device displays recommended products on the screen based on information from the server. The user can easily purchase the displayed products. Similarly, when a special offer is presented while using a subscription service, the user receives a notification about the offer and can check the available benefits.

[0037] For example, if a user has previously purchased multiple ebooks of a particular type, the server will automatically suggest new releases or products with special offers in a similar genre. Furthermore, when a user attempts to cancel a subscription service, the server will offer tailored offers to encourage continued use.

[0038] In this way, this system analyzes user behavior in real time and provides personalized suggestions quickly, enabling responses that meet the individual needs of each user. This allows companies to allocate resources efficiently and improve customer satisfaction.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The server collects data on users' past behavior from various data sources and stores it in a database. This includes purchase history, website browsing history, inquiry history, and so on.

[0042] Step 2:

[0043] The server preprocesses user behavior data stored in the database. This preprocessing includes imputing missing values, removing outliers, and standardizing the data format.

[0044] Step 3:

[0045] The server uses pre-processed data to train a machine learning model. The model includes an algorithm that learns patterns in user behavior and predicts the next action.

[0046] Step 4:

[0047] The server uses a trained model to predict the user's future behavior in real time. Based on the prediction results, it generates optimal product suggestions and special offers.

[0048] Step 5:

[0049] The server sends generated suggestions and offers to the terminal. The terminal then displays this information appropriately on the user screen.

[0050] Step 6:

[0051] When users see suggestions and offers displayed on their devices, the system helps them select products they are interested in and take advantage of benefits.

[0052] Step 7:

[0053] The user's response is fed back to the server. The server stores this feedback and uses it to improve the accuracy of the model in subsequent uses.

[0054] (Example 1)

[0055] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0056] In today's information society, where a vast amount of information flows, users face the challenge of difficulty in accessing information and products that meet their specific needs. Furthermore, while companies are required to respond quickly to the diverse needs of their customers, the systems and methods for providing personalized service are currently insufficient.

[0057] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0058] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for processing the stored information and converting it into an analyzable format; and means for training a predictive model to predict the user's future behavior based on the converted information. This makes it possible to predict the user's behavior and provide personalized suggestions quickly and accurately.

[0059] "User's past behavior information" refers to historical data about the user's past actions, including purchase history, browsing history, and inquiries.

[0060] A "means of storage" refers to a mechanism for saving collected data and maintaining it in a format that can be used for subsequent processing.

[0061] An "analyzable format" is a format in which data has the necessary structure for effective analysis and processing.

[0062] A "predictive model" is an algorithm or mathematical model used to predict future actions or outcomes based on given data.

[0063] "Personalized recommendations" refer to suggestions that present the most suitable options and products based on each user's past behavioral patterns and current circumstances.

[0064] A "user device" is a device used by a user for direct operation, such as a computer or a smartphone.

[0065] This invention is a system that provides personalized suggestions based on the user's past behavior information. This system mainly consists of a server and a user terminal, and aims to improve the user experience.

[0066] First, the server collects and stores information about the user's past behavior in a database. This information includes purchase history, click history, and inquiry details. At this stage, a common RDBMS (Relational Database Management System) is often used as the database management system.

[0067] Next, the server processes the accumulated information and converts it into an analyzable format. Preprocessing such as data cleansing and formatting is performed using programming languages ​​such as Python and R. This makes the data suitable as input for machine learning.

[0068] The server then trains a predictive model based on the pre-processed data. This model is generated using machine learning algorithms (e.g., TENSORFLOW® or PyTorch) to predict user behavior patterns. The trained model then generates personalized suggestions for each user.

[0069] The generated suggestions are notified to the user's terminal. The terminal displays the suggestions received from the server on the user interface, making it easy for the user to review their contents. Communication methods such as HTTP requests and WebSockets are used, enabling real-time suggestions.

[0070] For example, if a user has previously purchased multiple ebooks of a specific genre, the server will suggest new releases in that genre. Furthermore, by inputting a prompt such as, "Based on user A's purchase history over the past year, suggest products they are most likely to purchase next," appropriate results can be obtained.

[0071] This system enables companies to achieve efficient and accurate digital marketing and customer relationships, contributing to improved user satisfaction.

[0072] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0073] Step 1:

[0074] The server collects and stores past user behavior information in a database. Specifically, it obtains purchase history, click history, and inquiry details via APIs and stores them in a relational database. The input data is the user's operation history, and by saving this in the database, it is made usable for subsequent analysis and recommendations. The output is a well-organized dataset for use in subsequent processing.

[0075] Step 2:

[0076] The server processes the collected data and converts it into an analyzable format. This process uses Python to perform data cleansing, imputation of missing values, removal of outliers, and normalization for format consistency. The raw data accumulated in step 1 is used as input, and the output is organized, analyzable data.

[0077] Step 3:

[0078] The server trains a predictive model based on the data, which has been converted into an analyzable format. Using the machine learning library TensorFlow, it builds a model that learns user behavior patterns from the data. In this process, the processed data from step 2 is used as input to train the model, and the trained predictive model is obtained as output.

[0079] Step 4:

[0080] The server uses a trained predictive model to generate personalized recommendations for each user. In this step, the model analyzes user patterns and creates optimal product and service recommendations by providing instructions through prompts on what to suggest next. The input is the predictive model and the latest user data, and the output is a personalized list of recommendations.

[0081] Step 5:

[0082] The server notifies the terminal of the generated suggestions. The terminal displays the received suggestions on the user interface, allowing the user to easily review the suggestions. The input is suggestion data sent from the server, and the output is the suggested content displayed on the screen to the user. Based on this, the user can make choices such as purchasing products or using services.

[0083] (Application Example 1)

[0084] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0085] In recent years, accurately understanding user interests and providing relevant information in real time has become crucial for online platforms. However, traditional methods have been unable to fully utilize users' diverse past behavioral data, making it difficult to effectively deliver personalized advertising. This has created challenges in improving the user experience and ensuring platform profitability.

[0086] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0087] In this invention, the server includes means for collecting and storing information on a user's past behavior, means for preprocessing the stored information and converting it into an analyzable format, and means for constructing a learning model for predicting the user's future behavior based on the converted information. This enables the real-time and automated delivery of targeted advertisements tailored to the individual interests of each user.

[0088] "User's past behavior information" refers to data about actions a user has taken in the past, including browsing history, purchase history, and click history.

[0089] "Preprocessing" refers to the methods used to prepare raw data into an analyzable format, including processes such as data cleansing and formatting standardization.

[0090] A "learning model" is a predictive model built using machine learning techniques, which is trained to predict future behavior based on the user's past actions.

[0091] "Optimal suggestions" refer to information and products that are tailored to the user's interests and needs, generated based on predictions made by the aforementioned learning model.

[0092] "Notification" refers to a means of transmitting generated suggestions to the user's terminal, and is a method of providing information to the user in real time.

[0093] "Targeted advertising" refers to advertisements that are personalized based on a user's interests and behavioral history and delivered to specific users.

[0094] The system for realizing this invention collects and stores information on the user's past behavior and performs processing to build a learning model that predicts the user's future behavior based on this information. The server uses a database for this purpose and preprocesses the collected behavioral data to convert it into an analyzable format. In particular, preprocessing such as data cleansing and formatting standardization prepares the data for machine learning. Software libraries such as Pandas and Scikit-learn are used for this purpose.

[0095] Subsequently, the server builds a machine learning model and generates targeted ads based on user behavior patterns. The model is trained and stored on the server using algorithms such as random forest. This model is continuously updated to improve prediction accuracy.

[0096] The device notifies the user of targeted advertisements sent from the server and displays personalized ads in real time. This allows users to instantly find products and services that match their interests. For example, if a user frequently shows interest in health foods, their smartphone will receive notifications about new promotions and discounts.

[0097] Examples of specific prompt messages are as follows:

[0098] "Based on users' past behavioral data, predict the product categories they are most likely to be interested in next, and generate relevant ads."

[0099] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0100] Step 1:

[0101] The server collects user activity history data and stores it in a database. The input is user behavior data, and the output is historical behavior information in an organized format within the database. This ensures that all behavioral records necessary for subsequent processing are stored.

[0102] Step 2:

[0103] The server performs preprocessing on the stored data. Specifically, it performs data cleansing and formatting standardization, with the input being the raw data in the database. Subsequent data processing transforms the output into a dataset in an analyzable format.

[0104] Step 3:

[0105] The server trains a machine learning model using pre-processed data. The input is a dataset suitable for analysis, and the output is a learning model that predicts user behavior. Specifically, it uses a random forest to learn the data and improve prediction accuracy.

[0106] Step 4:

[0107] The server uses the built-in learning model to predict the user's next actions and generates optimally targeted advertisements based on those predictions. The input is the user's current data and the learning model, and the output is the advertising information presented to the user. The server utilizes the generative AI model to create advertisements that accurately reflect the user's interests.

[0108] Step 5:

[0109] The device notifies the user of targeted advertisements sent from the server. The input is advertising information from the server, and the output is personalized advertisements displayed on the user's device. Users can passively view advertisements related to their interests.

[0110] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0111] This invention relates to a system that predicts future user behavior based on past user behavior information, recognizes the user's emotions using an emotion engine, and reflects these emotions in the generated suggestions. The aim of this system is to improve the user experience by providing suggestions that are more suitable to the user's needs through cooperation between the server and the user terminal.

[0112] The server first collects and stores the user's past behavior data in a database. This data includes past purchase history, website usage history, and inquiry history. The collected data is then preprocessed, including imputing missing values ​​and removing outliers. Finally, a machine learning model is trained using this preprocessed data. This model analyzes the user's behavior patterns and predicts their next actions.

[0113] The emotion engine analyzes user text and behavioral data to recognize user emotions. This emotion information is used to tailor suggestions based on predicted user behavior. For example, by incorporating natural language processing technology, emotions can be estimated from user reviews and feedback and used as feedback for product recommendations.

[0114] Suggestions, taking into account sentiment information generated on the server, are notified to the user's device. When the user views the suggestion on their device, the displayed products and benefits are shown on the screen, and the user can select or use them.

[0115] For example, if a user writes a positive review for a particular product, the sentiment engine can detect that sentiment and suggest other products in the same category or related accessories. It can also offer users special discounts or improved product offers based on negative feedback.

[0116] The introduction of this system enables highly personalized suggestions that respond to each user's individual emotions and behaviors, allowing companies to improve customer satisfaction and provide services more efficiently.

[0117] The following describes the processing flow.

[0118] Step 1:

[0119] The server collects data on the user's past behavior from a database. This data includes purchase history, website activity logs, and past inquiries.

[0120] Step 2:

[0121] The server preprocesses the collected data. Specifically, it imputes missing values, filters outliers, and standardizes the data format to create an easy-to-understand dataset for analysis.

[0122] Step 3:

[0123] The server uses pre-processed data to train a learning model that predicts user behavior. This model uses machine learning algorithms to predict the user's purchasing intent and potential next actions.

[0124] Step 4:

[0125] The server uses an emotion engine to analyze user emotions from text data (reviews and inquiries). For example, it uses natural language processing to classify emotions into positive, negative, and neutral.

[0126] Step 5:

[0127] Based on the output of the emotion engine, the server incorporates emotional information into the behavior predicted by the learning model to generate suggestions tailored to the user. For example, it might suggest new products to users with positive emotions and discount offers to users with negative emotions.

[0128] Step 6:

[0129] The server generates suggestions and sends them to the terminal. The terminal then visually presents these to the user, providing an interface that allows them to understand the details of the products and services.

[0130] Step 7:

[0131] The user checks the presented suggestions and purchases products or uses services as needed. The device reports the user's selections to the server and collects the feedback as data.

[0132] Step 8:

[0133] Based on the feedback, the server retrains the model and emotion engine with data to improve their accuracy, and uses this to inform future recommendations. This process results in a more personalized user experience.

[0134] (Example 2)

[0135] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0136] Conventional systems typically made suggestions based on the user's past behavior, making it difficult to provide optimized suggestions that fully considered the user's emotions and intentions. This limited the potential for improving the user experience and the accuracy of the suggestions.

[0137] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0138] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning model for predicting the user's future behavior based on the converted information; means for performing sentiment analysis on the learning model and recognizing the user's emotions; means for generating suggestions based on the user's behavior prediction, taking into account the results of the sentiment analysis; and means for notifying the user of the generated suggestions. This makes it possible to provide highly accurate suggestions that take the user's emotions into consideration.

[0139] "User's past behavior information" refers to data such as the user's past purchase history, website usage history, and inquiry history.

[0140] "Data preprocessing" refers to the process of converting data into an analyzable format by imputing missing values ​​and removing outliers.

[0141] A "learning model" refers to an algorithm or structure that analyzes user behavior patterns based on collected data and predicts future behavior.

[0142] "Sentiment analysis" refers to the process of recognizing emotions from user text and behavioral data using natural language processing technology and generating a numerical emotion score.

[0143] "Means for generating suggestions" refers to the logic and mechanisms for generating personalized product and service suggestions based on user behavior predictions and sentiment analysis results.

[0144] "Means of notifying of proposals" refers to methods and technologies for notifying users of generated proposals and displaying them on the user's device.

[0145] This invention relates to a system that links a server and a terminal to provide optimized suggestions tailored to user needs. The main components of this system include a data collection module, a data preprocessing module, a machine learning model, a sentiment analysis engine, a suggestion generation module, and a notification module.

[0146] The server collects and stores users' past behavior information in a database. For example, it collects purchase history and browsing history using online store APIs and web tracking technologies. This information is preprocessed using data analysis software such as Python or R to impute missing values ​​and remove outliers.

[0147] The pre-processed data is then fed into a machine learning model. This model is built using libraries such as TensorFlow and PyTorch, and analyzes user behavior patterns to predict future actions. This makes it possible to predict which products and services users are likely to be interested in next.

[0148] Furthermore, the sentiment analysis engine applies natural language processing techniques to user text and behavioral data to identify user emotions. This process incorporates generative AI models such as OpenAI® to extract positive or negative emotions from review content and other data.

[0149] Based on sentiment analysis results and behavioral predictions, the suggestion generation module creates personalized suggestions. These suggestions are then sent from the server to the user's device. For example, if a user writes a positive review of a specific product, related products and accessories will be suggested. Conversely, if negative feedback is received, notifications of improved product offers or special discounts will be sent.

[0150] As a concrete example, by inputting a prompt message such as "Create a product suggestion best suited to user A based on their purchase history and reviews for the past month" into the AI ​​model, it becomes possible to create personalized suggestions for each user. In this way, the system of the present invention can provide highly accurate suggestions that take into account the user's emotions and behavior.

[0151] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0152] Step 1:

[0153] The server collects and stores users' past behavioral information in a database. It receives data such as purchase history and browsing history obtained from online store APIs and web tracking systems as input. This data is converted into a structured database format (e.g., CSV or SQL database format) and stored as output. Specifically, the server creates a data entry for each user, accumulating information in a consistent manner.

[0154] Step 2:

[0155] The server preprocesses the stored data and converts it into an analyzable format. The input is raw data stored in the database. Specific data processing involves using the Python Pandas library to impute missing values ​​(mean imputation and regression imputation) and remove outliers (statistical outlier detection). The output is a clean, analyzable dataset. This minimizes data noise and enables highly accurate analysis.

[0156] Step 3:

[0157] The server builds a machine learning model using preprocessed data and performs behavioral predictions. A cleaned dataset is used as input. Specifically, it trains a model using TensorFlow or PyTorch to learn user behavior patterns and predict the user's next action. The output is either the predicted user action or a list of products of interest. The model's accuracy is evaluated using a test dataset for validation.

[0158] Step 4:

[0159] The server uses an emotion analysis engine to recognize the user's emotions. The input data is text data such as user reviews and comments. For specific analysis, natural language processing techniques are utilized, for example, to calculate an emotion score from the text using a generative AI model. The output is a positive or negative emotion score based on the text written by the user. This allows the user's emotion-based judgment to be reflected in the recommendations.

[0160] Step 5:

[0161] The server generates suggestions based on behavioral predictions and sentiment scores. It receives predicted behavioral data and sentiment scores as input. Specifically, the suggestion model processes this data and uses prompts (e.g., "If user A's sentiment score is 8 or higher, suggest related products") to generate appropriate product and service combinations. The output is a customized suggestion tailored to the user. This enables more personalized suggestions based on the user's state.

[0162] Step 6:

[0163] The server notifies the user's device of the generated suggestions. The input is the customized suggestion content, which is the output of the suggestion generation module. Specifically, the server uses a notification API to send a push notification to the user's device. The output on the device is a list of suggested products and benefits. The user receives this suggestion and can immediately check the products that interest them. This series of actions improves user convenience and satisfaction.

[0164] (Application Example 2)

[0165] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0166] Traditional user behavior prediction systems have struggled to provide suggestions that reflect the emotional aspects of individual users. Furthermore, real-time product recommendations and special service offerings that respond to changes in user emotions are also difficult to achieve. This limits the improvement of the user experience and the achievement of higher customer satisfaction.

[0167] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0168] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning algorithm for predicting the user's future behavior based on the converted information; means for recognizing the user's emotional state using emotion analysis technology; and means for generating optimal suggestions based on the learning algorithm and the emotional state. This makes it possible to provide personalized suggestions and special discount information that take into account the emotional state and behavioral patterns of individual users.

[0169] "User's past behavior information" refers to data about the user's past actions, including purchase history, browsing history, review history, and other similar information.

[0170] "Storage" refers to saving collected data and making it available for later processing and analysis.

[0171] "Preprocessing" refers to a series of processes performed to convert raw data into an analyzable format, including imputing missing values ​​and removing outliers.

[0172] "Analyzable format" refers to a state where data is structured and can be effectively used with machine learning algorithms and analytical tools.

[0173] A "learning algorithm" refers to a computational method and program used to predict a user's future behavior based on collected data.

[0174] "Emotion analysis technology" refers to technologies for identifying emotions from user text data and behavioral data, and includes natural language processing and behavioral analysis.

[0175] "User emotional state" refers to information that indicates the type and intensity of emotions a user is experiencing at a given point in time.

[0176] "Optimal suggestions" are recommendations that are considered most beneficial or of interest to the user, generated based on the user's emotional state and predicted behavior.

[0177] "Special discount information" refers to information provided to increase user purchasing intent, offering goods or services at a lower price than the regular price.

[0178] The system for implementing this invention is configured as follows: First, the server collects information on the user's past behavior and stores it in a database. At this stage, the collected data includes the user's purchase history, browsing history, and review history. The stored data is then preprocessed to impute missing values ​​and remove outliers. Pandas, a Python library for data analysis, is used for this preprocessing.

[0179] The server uses machine learning algorithms to predict user behavior based on preprocessed data. This learning algorithm utilizes TensorFlow to build a model that accurately predicts the user's next action. Furthermore, the server employs sentiment analysis techniques, leveraging the NLTK natural language processing library to analyze collected user text data and recognize the user's emotional state.

[0180] Based on generated sentiment information and behavioral predictions, the server creates optimal suggestions for the user and notifies them via smartphone or web apps. These suggestions may include special discount information or recommendations for similar products. On the device, users can select from the information displayed on the screen and purchase or use it.

[0181] For example, when a user gives a positive review of a product, the server detects that emotion and suggests related products and accessories. It can also offer special discounts for negative feedback. Examples of prompts to input into the generative AI model include, "Please tell me how to suggest new related products based on products that have been highly rated in the past," and "Please write a procedure to detect positive emotions from user reviews and issue coupons based on them."

[0182] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0183] Step 1:

[0184] The server collects information about users' past behavior and stores it in a database. Inputs include user purchase history, browsing history, and review history. Storing this information in the database makes it available for later analysis.

[0185] Step 2:

[0186] The server preprocesses the stored data. The input is the raw data stored in the database, and the output is analyzable data with missing values ​​imputed and outliers removed. Specifically, Pandas is used to imputate missing values ​​and remove outliers.

[0187] Step 3:

[0188] The server uses machine learning algorithms based on preprocessed data to predict the user's future actions. The input for this step is preprocessed data, and the output is a model for predicting user behavior. TensorFlow is used to extract features from the data and build the learning model.

[0189] Step 4:

[0190] The server uses natural language processing (NLTK) to recognize the user's emotional state. Input is text data from user reviews and feedback, and output is the user's emotional state. Specifically, NLTK is used to analyze the text data and identify the emotions.

[0191] Step 5:

[0192] The server generates optimal suggestions based on a learning model and emotional states. The inputs are behavioral prediction data and emotional information, and the output is user-optimized suggestions. These suggestions may include product recommendations and special discount information.

[0193] Step 6:

[0194] The server notifies the user's terminal of the generated suggestions. In this step, the terminal displays the suggestions, and the user makes selections or purchases based on them. The input is the suggestion data sent from the server, and the output is the suggestion information that the user can see on their terminal.

[0195] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0196] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0197] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0198] [Second Embodiment]

[0199] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0200] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0201] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0202] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0203] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0204] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0205] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0206] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0207] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0208] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0209] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0210] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0211] This invention relates to a system that predicts future user behavior based on past user behavior information and provides users with optimal suggestions. This system consists of multiple servers and user terminals and aims to improve the user experience.

[0212] First, the server stores the user's past behavior information in a database. This includes purchase history, click history, and inquiry details. Next, the server preprocesses the stored data, performing data cleansing and formatting standardization to convert it into a format suitable for analysis.

[0213] The server trains a learning model using machine learning techniques based on pre-processed data. This model is designed to analyze user behavior patterns and predict their next actions. For example, in an e-commerce scenario, the server generates personalized product suggestions based on the user's purchase history, and in a subscription service scenario, it prepares special offers for users at risk of churn.

[0214] When suggestions are generated, they are notified to the user's device. When the user visits an e-commerce site, the device displays recommended products on the screen based on information from the server. The user can easily purchase the displayed products. Similarly, when a special offer is presented while using a subscription service, the user receives a notification about the offer and can check the available benefits.

[0215] For example, if a user has previously purchased multiple ebooks of a particular type, the server will automatically suggest new releases or products with special offers in a similar genre. Furthermore, when a user attempts to cancel a subscription service, the server will offer tailored offers to encourage continued use.

[0216] In this way, this system analyzes user behavior in real time and provides personalized suggestions quickly, enabling responses that meet the individual needs of each user. This allows companies to allocate resources efficiently and improve customer satisfaction.

[0217] The following describes the processing flow.

[0218] Step 1:

[0219] The server collects data on users' past behavior from various data sources and stores it in a database. This includes purchase history, website browsing history, inquiry history, and so on.

[0220] Step 2:

[0221] The server preprocesses user behavior data stored in the database. This preprocessing includes imputing missing values, removing outliers, and standardizing the data format.

[0222] Step 3:

[0223] The server uses pre-processed data to train a machine learning model. The model includes an algorithm that learns patterns in user behavior and predicts the next action.

[0224] Step 4:

[0225] The server uses a trained model to predict the user's future behavior in real time. Based on the prediction results, it generates optimal product suggestions and special offers.

[0226] Step 5:

[0227] The server sends generated suggestions and offers to the terminal. The terminal then displays this information appropriately on the user screen.

[0228] Step 6:

[0229] When users see suggestions and offers displayed on their devices, the system helps them select products they are interested in and take advantage of benefits.

[0230] Step 7:

[0231] The user's response is fed back to the server. The server stores this feedback and uses it to improve the accuracy of the model in subsequent uses.

[0232] (Example 1)

[0233] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0234] In today's information society, where a vast amount of information flows, users face the challenge of difficulty in accessing information and products that meet their specific needs. Furthermore, while companies are required to respond quickly to the diverse needs of their customers, the systems and methods for providing personalized service are currently insufficient.

[0235] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0236] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for processing the stored information and converting it into an analyzable format; and means for training a predictive model to predict the user's future behavior based on the converted information. This makes it possible to predict the user's behavior and provide personalized suggestions quickly and accurately.

[0237] "User's past behavior information" refers to historical data about the user's past actions, including purchase history, browsing history, and inquiries.

[0238] A "means of storage" refers to a mechanism for saving collected data and maintaining it in a format that can be used for subsequent processing.

[0239] An "analyzable format" is a format in which data has the necessary structure for effective analysis and processing.

[0240] A "predictive model" is an algorithm or mathematical model used to predict future actions or outcomes based on given data.

[0241] "Personalized recommendations" refer to suggestions that present the most suitable options and products based on each user's past behavioral patterns and current circumstances.

[0242] A "user device" is a device used by a user for direct operation, such as a computer or a smartphone.

[0243] This invention is a system that provides personalized suggestions based on the user's past behavior information. This system mainly consists of a server and a user terminal, and aims to improve the user experience.

[0244] First, the server collects and stores information about the user's past behavior in a database. This information includes purchase history, click history, and inquiry details. At this stage, a common RDBMS (Relational Database Management System) is often used as the database management system.

[0245] Next, the server processes the accumulated information and converts it into an analyzable format. Preprocessing such as data cleansing and formatting is performed using programming languages ​​such as Python and R. This makes the data suitable as input for machine learning.

[0246] The server then trains a predictive model based on the pre-processed data. This model is generated using machine learning algorithms (e.g., TensorFlow or PyTorch) to predict user behavior patterns. The trained model then generates individual suggestions for each user.

[0247] The generated suggestions are notified to the user's terminal. The terminal displays the suggestions received from the server on the user interface, making it easy for the user to review their contents. Communication methods such as HTTP requests and WebSockets are used, enabling real-time suggestions.

[0248] For example, if a user has previously purchased multiple ebooks of a specific genre, the server will suggest new releases in that genre. Furthermore, by inputting a prompt such as, "Based on user A's purchase history over the past year, suggest products they are most likely to purchase next," appropriate results can be obtained.

[0249] This system enables companies to achieve efficient and accurate digital marketing and customer relationships, contributing to improved user satisfaction.

[0250] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0251] Step 1:

[0252] The server collects and stores past user behavior information in a database. Specifically, it obtains purchase history, click history, and inquiry details via APIs and stores them in a relational database. The input data is the user's operation history, and by saving this in the database, it is made usable for subsequent analysis and recommendations. The output is a well-organized dataset for use in subsequent processing.

[0253] Step 2:

[0254] The server processes the collected data and converts it into an analyzable format. This process uses Python to perform data cleansing, imputation of missing values, removal of outliers, and normalization for format consistency. The raw data accumulated in step 1 is used as input, and the output is organized, analyzable data.

[0255] Step 3:

[0256] The server trains a predictive model based on the data, which has been converted into an analyzable format. Using the machine learning library TensorFlow, it builds a model that learns user behavior patterns from the data. In this process, the processed data from step 2 is used as input to train the model, and the trained predictive model is obtained as output.

[0257] Step 4:

[0258] The server uses a trained predictive model to generate personalized recommendations for each user. In this step, the model analyzes user patterns and creates optimal product and service recommendations by providing instructions through prompts on what to suggest next. The input is the predictive model and the latest user data, and the output is a personalized list of recommendations.

[0259] Step 5:

[0260] The server notifies the terminal of the generated suggestions. The terminal displays the received suggestions on the user interface, allowing the user to easily review the suggestions. The input is suggestion data sent from the server, and the output is the suggested content displayed on the screen to the user. Based on this, the user can make choices such as purchasing products or using services.

[0261] (Application Example 1)

[0262] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0263] In recent years, accurately understanding user interests and providing relevant information in real time has become crucial for online platforms. However, traditional methods have been unable to fully utilize users' diverse past behavioral data, making it difficult to effectively deliver personalized advertising. This has created challenges in improving the user experience and ensuring platform profitability.

[0264] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0265] In this invention, the server includes means for collecting and storing information on a user's past behavior, means for preprocessing the stored information and converting it into an analyzable format, and means for constructing a learning model for predicting the user's future behavior based on the converted information. This enables the real-time and automated delivery of targeted advertisements tailored to the individual interests of each user.

[0266] "User's past behavior information" refers to data about actions a user has taken in the past, including browsing history, purchase history, and click history.

[0267] "Preprocessing" refers to the methods used to prepare raw data into an analyzable format, including processes such as data cleansing and formatting standardization.

[0268] A "learning model" is a predictive model built using machine learning techniques, which is trained to predict future behavior based on the user's past actions.

[0269] "Optimal suggestions" refer to information and products that are tailored to the user's interests and needs, generated based on predictions made by the aforementioned learning model.

[0270] "Notification" refers to a means of transmitting generated suggestions to the user's terminal, and is a method of providing information to the user in real time.

[0271] "Targeted advertising" refers to advertisements that are personalized based on a user's interests and behavioral history and delivered to specific users.

[0272] The system for realizing this invention collects and stores information on the user's past behavior and performs processing to build a learning model that predicts the user's future behavior based on this information. The server uses a database for this purpose and preprocesses the collected behavioral data to convert it into an analyzable format. In particular, preprocessing such as data cleansing and formatting standardization prepares the data for machine learning. Software libraries such as Pandas and Scikit-learn are used for this purpose.

[0273] Subsequently, the server builds a machine learning model and generates targeted ads based on user behavior patterns. The model is trained and stored on the server using algorithms such as random forest. This model is continuously updated to improve prediction accuracy.

[0274] The device notifies the user of targeted advertisements sent from the server and displays personalized ads in real time. This allows users to instantly find products and services that match their interests. For example, if a user frequently shows interest in health foods, their smartphone will receive notifications about new promotions and discounts.

[0275] Examples of specific prompt messages are as follows:

[0276] "Based on users' past behavioral data, predict the product categories they are most likely to be interested in next, and generate relevant ads."

[0277] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0278] Step 1:

[0279] The server collects the user's activity history data and stores it in a database. The input is the user's behavior data, and the output is the past behavior information in a format organized in the database. This ensures that all the action records necessary for subsequent processing are stored.

[0280] Step 2:

[0281] The server performs preprocessing on the stored data. Specifically, it conducts data cleansing and format unification. The input is the raw data in the database. Through subsequent data processing, the output becomes a dataset converted into an analyzable format.

[0282] Step 3:

[0283] The server trains a machine learning model using the preprocessed data. The input is a dataset suitable for analysis, and the output is a learning model that predicts the user's behavior. As specific computational processing, it uses random forest to learn the data and improve the prediction accuracy.

[0284] Step 4:

[0285] The server uses the constructed learning model to predict the next action and generates the optimal targeted advertisement based on it. The input is the user's current data and the learning model, and the output is the promotional information presented to the user. The server utilizes a generation AI model to create advertisements that accurately reflect the user's interests.

[0286] Step 5:

[0287] The terminal notifies the user of the targeted advertisement sent from the server. The input is the advertisement information from the server, and the output is the personalized advertisement displayed on the user terminal. The user can passively view advertisements related to their interests.

[0288] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0289] This invention relates to a system that predicts future user behavior based on past user behavior information, recognizes the user's emotions using an emotion engine, and reflects these emotions in the generated suggestions. The aim of this system is to improve the user experience by providing suggestions that are more suitable to the user's needs through cooperation between the server and the user terminal.

[0290] The server first collects and stores the user's past behavior data in a database. This data includes past purchase history, website usage history, and inquiry history. The collected data is then preprocessed, including imputing missing values ​​and removing outliers. Finally, a machine learning model is trained using this preprocessed data. This model analyzes the user's behavior patterns and predicts their next actions.

[0291] The emotion engine analyzes user text and behavioral data to recognize user emotions. This emotion information is used to tailor suggestions based on predicted user behavior. For example, by incorporating natural language processing technology, emotions can be estimated from user reviews and feedback and used as feedback for product recommendations.

[0292] Suggestions, taking into account sentiment information generated on the server, are notified to the user's device. When the user views the suggestion on their device, the displayed products and benefits are shown on the screen, and the user can select or use them.

[0293] For example, if a user writes a positive review for a particular product, the sentiment engine can detect that sentiment and suggest other products in the same category or related accessories. It can also offer users special discounts or improved product offers based on negative feedback.

[0294] The introduction of this system enables highly personalized suggestions that respond to each user's individual emotions and behaviors, allowing companies to improve customer satisfaction and provide services more efficiently.

[0295] The following describes the processing flow.

[0296] Step 1:

[0297] The server collects data on the user's past behavior from a database. This data includes purchase history, website activity logs, and past inquiries.

[0298] Step 2:

[0299] The server preprocesses the collected data. Specifically, it imputes missing values, filters outliers, and standardizes the data format to create an easy-to-understand dataset for analysis.

[0300] Step 3:

[0301] The server uses pre-processed data to train a learning model that predicts user behavior. This model uses machine learning algorithms to predict the user's purchasing intent and potential next actions.

[0302] Step 4:

[0303] The server uses an emotion engine to analyze user emotions from text data (reviews and inquiries). For example, it uses natural language processing to classify emotions into positive, negative, and neutral.

[0304] Step 5:

[0305] Based on the output of the emotion engine, the server takes into account the emotion information in the actions predicted by the learning model and generates a proposal suitable for the user. For example, it proposes new products to users with positive emotions and discount offers to users with negative emotions.

[0306] Step 6:

[0307] The server transmits the proposal generated by it to the terminal. The terminal visually presents this to the user and provides an interface through which details of products or services can be understood.

[0308] Step 7:

[0309] The user checks the presented proposal and, if necessary, purchases a product or uses a service. The terminal reports the user's selection to the server and collects the feedback as data.

[0310] Step 8:

[0311] Based on the feedback, the server re-learns the data to improve the accuracy of the model and the emotion engine, and utilizes this for the next proposal. Through this process, a more personalized user experience is realized.

[0312] (Example 2)

[0313] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0314] In a conventional system, proposals based on the user's past behavior information are common, and it is difficult to make optimized proposals that fully consider the user's emotions and intentions. Therefore, there are limitations in improving the user experience and the accuracy of proposals.

[0315] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following respective means.

[0316] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning model for predicting the user's future behavior based on the converted information; means for performing sentiment analysis on the learning model and recognizing the user's emotions; means for generating suggestions based on the user's behavior prediction, taking into account the results of the sentiment analysis; and means for notifying the user of the generated suggestions. This makes it possible to provide highly accurate suggestions that take the user's emotions into consideration.

[0317] "User's past behavior information" refers to data such as the user's past purchase history, website usage history, and inquiry history.

[0318] "Data preprocessing" refers to the process of converting data into an analyzable format by imputing missing values ​​and removing outliers.

[0319] A "learning model" refers to an algorithm or structure that analyzes user behavior patterns based on collected data and predicts future behavior.

[0320] "Sentiment analysis" refers to the process of recognizing emotions from user text and behavioral data using natural language processing technology and generating a numerical emotion score.

[0321] "Means for generating suggestions" refers to the logic and mechanisms for generating personalized product and service suggestions based on user behavior predictions and sentiment analysis results.

[0322] "Means of notifying of proposals" refers to methods and technologies for notifying users of generated proposals and displaying them on the user's device.

[0323] This invention relates to a system that links a server and a terminal to provide optimized suggestions tailored to user needs. The main components of this system include a data collection module, a data preprocessing module, a machine learning model, a sentiment analysis engine, a suggestion generation module, and a notification module.

[0324] The server collects and stores users' past behavior information in a database. For example, it collects purchase history and browsing history using online store APIs and web tracking technologies. This information is preprocessed using data analysis software such as Python or R to impute missing values ​​and remove outliers.

[0325] The pre-processed data is then fed into a machine learning model. This model is built using libraries such as TensorFlow and PyTorch, and analyzes user behavior patterns to predict future actions. This makes it possible to predict which products and services users are likely to be interested in next.

[0326] Furthermore, the sentiment analysis engine applies natural language processing techniques to user text and behavioral data to identify user emotions. This process incorporates generative AI models such as OpenAI to extract positive or negative emotions from review content and other data.

[0327] Based on sentiment analysis results and behavioral predictions, the suggestion generation module creates personalized suggestions. These suggestions are then sent from the server to the user's device. For example, if a user writes a positive review of a specific product, related products and accessories will be suggested. Conversely, if negative feedback is received, notifications of improved product offers or special discounts will be sent.

[0328] As a concrete example, by inputting a prompt message such as "Create a product suggestion best suited to user A based on their purchase history and reviews for the past month" into the AI ​​model, it becomes possible to create personalized suggestions for each user. In this way, the system of the present invention can provide highly accurate suggestions that take into account the user's emotions and behavior.

[0329] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0330] Step 1:

[0331] The server collects and stores users' past behavioral information in a database. It receives data such as purchase history and browsing history obtained from online store APIs and web tracking systems as input. This data is converted into a structured database format (e.g., CSV or SQL database format) and stored as output. Specifically, the server creates a data entry for each user, accumulating information in a consistent manner.

[0332] Step 2:

[0333] The server preprocesses the stored data and converts it into an analyzable format. The input is raw data stored in the database. Specific data processing involves using the Python Pandas library to impute missing values ​​(mean imputation and regression imputation) and remove outliers (statistical outlier detection). The output is a clean, analyzable dataset. This minimizes data noise and enables highly accurate analysis.

[0334] Step 3:

[0335] The server builds a machine learning model using preprocessed data and performs behavioral predictions. A cleaned dataset is used as input. Specifically, it trains a model using TensorFlow or PyTorch to learn user behavior patterns and predict the user's next action. The output is either the predicted user action or a list of products of interest. The model's accuracy is evaluated using a test dataset for validation.

[0336] Step 4:

[0337] The server uses an emotion analysis engine to recognize the user's emotions. The input data is text data such as user reviews and comments. For specific analysis, natural language processing techniques are utilized, for example, to calculate an emotion score from the text using a generative AI model. The output is a positive or negative emotion score based on the text written by the user. This allows the user's emotion-based judgment to be reflected in the recommendations.

[0338] Step 5:

[0339] The server generates suggestions based on behavioral predictions and sentiment scores. It receives predicted behavioral data and sentiment scores as input. Specifically, the suggestion model processes this data and uses prompts (e.g., "If user A's sentiment score is 8 or higher, suggest related products") to generate appropriate product and service combinations. The output is a customized suggestion tailored to the user. This enables more personalized suggestions based on the user's state.

[0340] Step 6:

[0341] The server notifies the user's device of the generated suggestions. The input is the customized suggestion content, which is the output of the suggestion generation module. Specifically, the server uses a notification API to send a push notification to the user's device. The output on the device is a list of suggested products and benefits. The user receives this suggestion and can immediately check the products that interest them. This series of actions improves user convenience and satisfaction.

[0342] (Application Example 2)

[0343] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0344] Traditional user behavior prediction systems have struggled to provide suggestions that reflect the emotional aspects of individual users. Furthermore, real-time product recommendations and special service offerings that respond to changes in user emotions are also difficult to achieve. This limits the improvement of the user experience and the achievement of higher customer satisfaction.

[0345] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0346] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning algorithm for predicting the user's future behavior based on the converted information; means for recognizing the user's emotional state using emotion analysis technology; and means for generating optimal suggestions based on the learning algorithm and the emotional state. This makes it possible to provide personalized suggestions and special discount information that take into account the emotional state and behavioral patterns of individual users.

[0347] "User's past behavior information" refers to data about the user's past actions, including purchase history, browsing history, review history, and other similar information.

[0348] "Storage" refers to saving collected data and making it available for later processing and analysis.

[0349] "Preprocessing" refers to a series of processes performed to convert raw data into an analyzable format, including imputing missing values ​​and removing outliers.

[0350] "Analyzable format" refers to a state where data is structured and can be effectively used with machine learning algorithms and analytical tools.

[0351] A "learning algorithm" refers to a computational method and program used to predict a user's future behavior based on collected data.

[0352] "Emotion analysis technology" refers to technologies for identifying emotions from user text data and behavioral data, and includes natural language processing and behavioral analysis.

[0353] "User emotional state" refers to information that indicates the type and intensity of emotions a user is experiencing at a given point in time.

[0354] "Optimal suggestions" are recommendations that are considered most beneficial or of interest to the user, generated based on the user's emotional state and predicted behavior.

[0355] "Special discount information" refers to information provided to increase user purchasing intent, offering goods or services at a lower price than the regular price.

[0356] The system for implementing this invention is configured as follows: First, the server collects information on the user's past behavior and stores it in a database. At this stage, the collected data includes the user's purchase history, browsing history, and review history. The stored data is then preprocessed to impute missing values ​​and remove outliers. Pandas, a Python library for data analysis, is used for this preprocessing.

[0357] The server uses machine learning algorithms to predict user behavior based on preprocessed data. This learning algorithm utilizes TensorFlow to build a model that accurately predicts the user's next action. Furthermore, the server employs sentiment analysis techniques, leveraging the NLTK natural language processing library to analyze collected user text data and recognize the user's emotional state.

[0358] Based on generated sentiment information and behavioral predictions, the server creates optimal suggestions for the user and notifies them via smartphone or web apps. These suggestions may include special discount information or recommendations for similar products. On the device, users can select from the information displayed on the screen and purchase or use it.

[0359] For example, when a user gives a positive review of a product, the server detects that emotion and suggests related products and accessories. It can also offer special discounts for negative feedback. Examples of prompts to input into the generative AI model include, "Please tell me how to suggest new related products based on products that have been highly rated in the past," and "Please write a procedure to detect positive emotions from user reviews and issue coupons based on them."

[0360] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0361] Step 1:

[0362] The server collects information about users' past behavior and stores it in a database. Inputs include user purchase history, browsing history, and review history. Storing this information in the database makes it available for later analysis.

[0363] Step 2:

[0364] The server preprocesses the stored data. The input is the raw data stored in the database, and the output is analyzable data with missing values ​​imputed and outliers removed. Specifically, Pandas is used to imputate missing values ​​and remove outliers.

[0365] Step 3:

[0366] The server uses machine learning algorithms based on preprocessed data to predict the user's future actions. The input for this step is preprocessed data, and the output is a model for predicting user behavior. TensorFlow is used to extract features from the data and build the learning model.

[0367] Step 4:

[0368] The server uses natural language processing (NLTK) to recognize the user's emotional state. Input is text data from user reviews and feedback, and output is the user's emotional state. Specifically, NLTK is used to analyze the text data and identify the emotions.

[0369] Step 5:

[0370] The server generates optimal suggestions based on a learning model and emotional states. The inputs are behavioral prediction data and emotional information, and the output is user-optimized suggestions. These suggestions may include product recommendations and special discount information.

[0371] Step 6:

[0372] The server notifies the user's terminal of the generated suggestions. In this step, the terminal displays the suggestions, and the user makes selections or purchases based on them. The input is the suggestion data sent from the server, and the output is the suggestion information that the user can see on their terminal.

[0373] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0374] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0375] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0376] [Third Embodiment]

[0377] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0378] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0379] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0380] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0381] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0382] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0383] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0384] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0385] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0386] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0387] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0388] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0389] This invention relates to a system that predicts future user behavior based on past user behavior information and provides users with optimal suggestions. This system consists of multiple servers and user terminals and aims to improve the user experience.

[0390] First, the server stores the user's past behavior information in a database. This includes purchase history, click history, and inquiry details. Next, the server preprocesses the stored data, performing data cleansing and formatting standardization to convert it into a format suitable for analysis.

[0391] The server trains a learning model using machine learning techniques based on pre-processed data. This model is designed to analyze user behavior patterns and predict their next actions. For example, in an e-commerce scenario, the server generates personalized product suggestions based on the user's purchase history, and in a subscription service scenario, it prepares special offers for users at risk of churn.

[0392] When suggestions are generated, they are notified to the user's device. When the user visits an e-commerce site, the device displays recommended products on the screen based on information from the server. The user can easily purchase the displayed products. Similarly, when a special offer is presented while using a subscription service, the user receives a notification about the offer and can check the available benefits.

[0393] For example, if a user has previously purchased multiple ebooks of a particular type, the server will automatically suggest new releases or products with special offers in a similar genre. Furthermore, when a user attempts to cancel a subscription service, the server will offer tailored offers to encourage continued use.

[0394] In this way, this system analyzes user behavior in real time and provides personalized suggestions quickly, enabling responses that meet the individual needs of each user. This allows companies to allocate resources efficiently and improve customer satisfaction.

[0395] The following describes the processing flow.

[0396] Step 1:

[0397] The server collects data on users' past behavior from various data sources and stores it in a database. This includes purchase history, website browsing history, inquiry history, and so on.

[0398] Step 2:

[0399] The server preprocesses user behavior data stored in the database. This preprocessing includes imputing missing values, removing outliers, and standardizing the data format.

[0400] Step 3:

[0401] The server uses pre-processed data to train a machine learning model. The model includes an algorithm that learns patterns in user behavior and predicts the next action.

[0402] Step 4:

[0403] The server uses a trained model to predict the user's future behavior in real time. Based on the prediction results, it generates optimal product suggestions and special offers.

[0404] Step 5:

[0405] The server sends generated suggestions and offers to the terminal. The terminal then displays this information appropriately on the user screen.

[0406] Step 6:

[0407] When users see suggestions and offers displayed on their devices, the system helps them select products they are interested in and take advantage of benefits.

[0408] Step 7:

[0409] The user's response is fed back to the server. The server stores this feedback and uses it to improve the accuracy of the model in subsequent uses.

[0410] (Example 1)

[0411] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0412] In today's information society, where a vast amount of information flows, users face the challenge of difficulty in accessing information and products that meet their specific needs. Furthermore, while companies are required to respond quickly to the diverse needs of their customers, the systems and methods for providing personalized service are currently insufficient.

[0413] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0414] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for processing the stored information and converting it into an analyzable format; and means for training a predictive model to predict the user's future behavior based on the converted information. This makes it possible to predict the user's behavior and provide personalized suggestions quickly and accurately.

[0415] "User's past behavior information" refers to historical data about the user's past actions, including purchase history, browsing history, and inquiries.

[0416] A "means of storage" refers to a mechanism for saving collected data and maintaining it in a format that can be used for subsequent processing.

[0417] An "analyzable format" is a format in which data has the necessary structure for effective analysis and processing.

[0418] A "predictive model" is an algorithm or mathematical model used to predict future actions or outcomes based on given data.

[0419] "Personalized recommendations" refer to suggestions that present the most suitable options and products based on each user's past behavioral patterns and current circumstances.

[0420] A "user device" is a device used by a user for direct operation, such as a computer or a smartphone.

[0421] This invention is a system that provides personalized suggestions based on the user's past behavior information. This system mainly consists of a server and a user terminal, and aims to improve the user experience.

[0422] First, the server collects and stores information about the user's past behavior in a database. This information includes purchase history, click history, and inquiry details. At this stage, a common RDBMS (Relational Database Management System) is often used as the database management system.

[0423] Next, the server processes the accumulated information and converts it into an analyzable format. Preprocessing such as data cleansing and formatting is performed using programming languages ​​such as Python and R. This makes the data suitable as input for machine learning.

[0424] The server then trains a predictive model based on the pre-processed data. This model is generated using machine learning algorithms (e.g., TensorFlow or PyTorch) to predict user behavior patterns. The trained model then generates individual suggestions for each user.

[0425] The generated suggestions are notified to the user's terminal. The terminal displays the suggestions received from the server on the user interface, making it easy for the user to review their contents. Communication methods such as HTTP requests and WebSockets are used, enabling real-time suggestions.

[0426] For example, if a user has previously purchased multiple ebooks of a specific genre, the server will suggest new releases in that genre. Furthermore, by inputting a prompt such as, "Based on user A's purchase history over the past year, suggest products they are most likely to purchase next," appropriate results can be obtained.

[0427] This system enables companies to achieve efficient and accurate digital marketing and customer relationships, contributing to improved user satisfaction.

[0428] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0429] Step 1:

[0430] The server collects and stores past user behavior information in a database. Specifically, it obtains purchase history, click history, and inquiry details via APIs and stores them in a relational database. The input data is the user's operation history, and by saving this in the database, it is made usable for subsequent analysis and recommendations. The output is a well-organized dataset for use in subsequent processing.

[0431] Step 2:

[0432] The server processes the collected data and converts it into an analyzable format. This process uses Python to perform data cleansing, imputation of missing values, removal of outliers, and normalization for format consistency. The raw data accumulated in step 1 is used as input, and the output is organized, analyzable data.

[0433] Step 3:

[0434] The server trains a predictive model based on the data, which has been converted into an analyzable format. Using the machine learning library TensorFlow, it builds a model that learns user behavior patterns from the data. In this process, the processed data from step 2 is used as input to train the model, and the trained predictive model is obtained as output.

[0435] Step 4:

[0436] The server uses a trained predictive model to generate personalized recommendations for each user. In this step, the model analyzes user patterns and creates optimal product and service recommendations by providing instructions through prompts on what to suggest next. The input is the predictive model and the latest user data, and the output is a personalized list of recommendations.

[0437] Step 5:

[0438] The server notifies the terminal of the generated suggestions. The terminal displays the received suggestions on the user interface, allowing the user to easily review the suggestions. The input is suggestion data sent from the server, and the output is the suggested content displayed on the screen to the user. Based on this, the user can make choices such as purchasing products or using services.

[0439] (Application Example 1)

[0440] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0441] In recent years, accurately understanding user interests and providing relevant information in real time has become crucial for online platforms. However, traditional methods have been unable to fully utilize users' diverse past behavioral data, making it difficult to effectively deliver personalized advertising. This has created challenges in improving the user experience and ensuring platform profitability.

[0442] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0443] In this invention, the server includes means for collecting and storing information on a user's past behavior, means for preprocessing the stored information and converting it into an analyzable format, and means for constructing a learning model for predicting the user's future behavior based on the converted information. This enables the real-time and automated delivery of targeted advertisements tailored to the individual interests of each user.

[0444] "User's past behavior information" refers to data about actions a user has taken in the past, including browsing history, purchase history, and click history.

[0445] "Preprocessing" refers to the methods used to prepare raw data into an analyzable format, including processes such as data cleansing and formatting standardization.

[0446] A "learning model" is a predictive model built using machine learning techniques, which is trained to predict future behavior based on the user's past actions.

[0447] "Optimal suggestions" refer to information and products that are tailored to the user's interests and needs, generated based on predictions made by the aforementioned learning model.

[0448] "Notification" refers to a means of transmitting generated suggestions to the user's terminal, and is a method of providing information to the user in real time.

[0449] "Targeted advertising" refers to advertisements that are personalized based on a user's interests and behavioral history and delivered to specific users.

[0450] The system for realizing this invention collects and stores information on the user's past behavior and performs processing to build a learning model that predicts the user's future behavior based on this information. The server uses a database for this purpose and preprocesses the collected behavioral data to convert it into an analyzable format. In particular, preprocessing such as data cleansing and formatting standardization prepares the data for machine learning. Software libraries such as Pandas and Scikit-learn are used for this purpose.

[0451] Subsequently, the server builds a machine learning model and generates targeted ads based on user behavior patterns. The model is trained and stored on the server using algorithms such as random forest. This model is continuously updated to improve prediction accuracy.

[0452] The device notifies the user of targeted advertisements sent from the server and displays personalized ads in real time. This allows users to instantly find products and services that match their interests. For example, if a user frequently shows interest in health foods, their smartphone will receive notifications about new promotions and discounts.

[0453] Examples of specific prompt messages are as follows:

[0454] "Based on users' past behavioral data, predict the product categories they are most likely to be interested in next, and generate relevant ads."

[0455] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0456] Step 1:

[0457] The server collects user activity history data and stores it in a database. The input is user behavior data, and the output is historical behavior information in an organized format within the database. This ensures that all behavioral records necessary for subsequent processing are stored.

[0458] Step 2:

[0459] The server performs preprocessing on the stored data. Specifically, it performs data cleansing and formatting standardization, with the input being the raw data in the database. Subsequent data processing transforms the output into a dataset in an analyzable format.

[0460] Step 3:

[0461] The server trains a machine learning model using pre-processed data. The input is a dataset suitable for analysis, and the output is a learning model that predicts user behavior. Specifically, it uses a random forest to learn the data and improve prediction accuracy.

[0462] Step 4:

[0463] The server uses the built-in learning model to predict the user's next actions and generates optimally targeted advertisements based on those predictions. The input is the user's current data and the learning model, and the output is the advertising information presented to the user. The server utilizes the generative AI model to create advertisements that accurately reflect the user's interests.

[0464] Step 5:

[0465] The device notifies the user of targeted advertisements sent from the server. The input is advertising information from the server, and the output is personalized advertisements displayed on the user's device. Users can passively view advertisements related to their interests.

[0466] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0467] This invention relates to a system that predicts future user behavior based on past user behavior information, recognizes the user's emotions using an emotion engine, and reflects these emotions in the generated suggestions. The aim of this system is to improve the user experience by providing suggestions that are more suitable to the user's needs through cooperation between the server and the user terminal.

[0468] The server first collects and stores the user's past behavior data in a database. This data includes past purchase history, website usage history, and inquiry history. The collected data is then preprocessed, including imputing missing values ​​and removing outliers. Finally, a machine learning model is trained using this preprocessed data. This model analyzes the user's behavior patterns and predicts their next actions.

[0469] The emotion engine analyzes user text and behavioral data to recognize user emotions. This emotion information is used to tailor suggestions based on predicted user behavior. For example, by incorporating natural language processing technology, emotions can be estimated from user reviews and feedback and used as feedback for product recommendations.

[0470] Suggestions, taking into account sentiment information generated on the server, are notified to the user's device. When the user views the suggestion on their device, the displayed products and benefits are shown on the screen, and the user can select or use them.

[0471] For example, if a user writes a positive review for a particular product, the sentiment engine can detect that sentiment and suggest other products in the same category or related accessories. It can also offer users special discounts or improved product offers based on negative feedback.

[0472] The introduction of this system enables highly personalized suggestions that respond to each user's individual emotions and behaviors, allowing companies to improve customer satisfaction and provide services more efficiently.

[0473] The following describes the processing flow.

[0474] Step 1:

[0475] The server collects data on the user's past behavior from a database. This data includes purchase history, website activity logs, and past inquiries.

[0476] Step 2:

[0477] The server preprocesses the collected data. Specifically, it imputes missing values, filters outliers, and standardizes the data format to create an easy-to-understand dataset for analysis.

[0478] Step 3:

[0479] The server uses pre-processed data to train a learning model that predicts user behavior. This model uses machine learning algorithms to predict the user's purchasing intent and potential next actions.

[0480] Step 4:

[0481] The server uses an emotion engine to analyze user emotions from text data (reviews and inquiries). For example, it uses natural language processing to classify emotions into positive, negative, and neutral.

[0482] Step 5:

[0483] Based on the output of the emotion engine, the server incorporates emotional information into the behavior predicted by the learning model to generate suggestions tailored to the user. For example, it might suggest new products to users with positive emotions and discount offers to users with negative emotions.

[0484] Step 6:

[0485] The server generates suggestions and sends them to the terminal. The terminal then visually presents these to the user, providing an interface that allows them to understand the details of the products and services.

[0486] Step 7:

[0487] The user checks the presented suggestions and purchases products or uses services as needed. The device reports the user's selections to the server and collects the feedback as data.

[0488] Step 8:

[0489] Based on the feedback, the server retrains the model and emotion engine with data to improve their accuracy, and uses this to inform future recommendations. This process results in a more personalized user experience.

[0490] (Example 2)

[0491] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0492] Conventional systems typically made suggestions based on the user's past behavior, making it difficult to provide optimized suggestions that fully considered the user's emotions and intentions. This limited the potential for improving the user experience and the accuracy of the suggestions.

[0493] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0494] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning model for predicting the user's future behavior based on the converted information; means for performing sentiment analysis on the learning model and recognizing the user's emotions; means for generating suggestions based on the user's behavior prediction, taking into account the results of the sentiment analysis; and means for notifying the user of the generated suggestions. This makes it possible to provide highly accurate suggestions that take the user's emotions into consideration.

[0495] "User's past behavior information" refers to data such as the user's past purchase history, website usage history, and inquiry history.

[0496] "Data preprocessing" refers to the process of converting data into an analyzable format by imputing missing values ​​and removing outliers.

[0497] A "learning model" refers to an algorithm or structure that analyzes user behavior patterns based on collected data and predicts future behavior.

[0498] "Sentiment analysis" refers to the process of recognizing emotions from user text and behavioral data using natural language processing technology and generating a numerical emotion score.

[0499] "Means for generating suggestions" refers to the logic and mechanisms for generating personalized product and service suggestions based on user behavior predictions and sentiment analysis results.

[0500] "Means of notifying of proposals" refers to methods and technologies for notifying users of generated proposals and displaying them on the user's device.

[0501] This invention relates to a system that links a server and a terminal to provide optimized suggestions tailored to user needs. The main components of this system include a data collection module, a data preprocessing module, a machine learning model, a sentiment analysis engine, a suggestion generation module, and a notification module.

[0502] The server collects and stores users' past behavior information in a database. For example, it collects purchase history and browsing history using online store APIs and web tracking technologies. This information is preprocessed using data analysis software such as Python or R to impute missing values ​​and remove outliers.

[0503] The pre-processed data is then fed into a machine learning model. This model is built using libraries such as TensorFlow and PyTorch, and analyzes user behavior patterns to predict future actions. This makes it possible to predict which products and services users are likely to be interested in next.

[0504] Furthermore, the sentiment analysis engine applies natural language processing techniques to user text and behavioral data to identify user emotions. This process incorporates generative AI models such as OpenAI to extract positive or negative emotions from review content and other data.

[0505] Based on sentiment analysis results and behavioral predictions, the suggestion generation module creates personalized suggestions. These suggestions are then sent from the server to the user's device. For example, if a user writes a positive review of a specific product, related products and accessories will be suggested. Conversely, if negative feedback is received, notifications of improved product offers or special discounts will be sent.

[0506] As a concrete example, by inputting a prompt message such as "Create a product suggestion best suited to user A based on their purchase history and reviews for the past month" into the AI ​​model, it becomes possible to create personalized suggestions for each user. In this way, the system of the present invention can provide highly accurate suggestions that take into account the user's emotions and behavior.

[0507] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0508] Step 1:

[0509] The server collects and stores users' past behavioral information in a database. It receives data such as purchase history and browsing history obtained from online store APIs and web tracking systems as input. This data is converted into a structured database format (e.g., CSV or SQL database format) and stored as output. Specifically, the server creates a data entry for each user, accumulating information in a consistent manner.

[0510] Step 2:

[0511] The server preprocesses the stored data and converts it into an analyzable format. The input is raw data stored in the database. Specific data processing involves using the Python Pandas library to impute missing values ​​(mean imputation and regression imputation) and remove outliers (statistical outlier detection). The output is a clean, analyzable dataset. This minimizes data noise and enables highly accurate analysis.

[0512] Step 3:

[0513] The server builds a machine learning model using preprocessed data and performs behavioral predictions. A cleaned dataset is used as input. Specifically, it trains a model using TensorFlow or PyTorch to learn user behavior patterns and predict the user's next action. The output is either the predicted user action or a list of products of interest. The model's accuracy is evaluated using a test dataset for validation.

[0514] Step 4:

[0515] The server uses an emotion analysis engine to recognize the user's emotions. The input data is text data such as user reviews and comments. For specific analysis, natural language processing techniques are utilized, for example, to calculate an emotion score from the text using a generative AI model. The output is a positive or negative emotion score based on the text written by the user. This allows the user's emotion-based judgment to be reflected in the recommendations.

[0516] Step 5:

[0517] The server generates suggestions based on behavioral predictions and sentiment scores. It receives predicted behavioral data and sentiment scores as input. Specifically, the suggestion model processes this data and uses prompts (e.g., "If user A's sentiment score is 8 or higher, suggest related products") to generate appropriate product and service combinations. The output is a customized suggestion tailored to the user. This enables more personalized suggestions based on the user's state.

[0518] Step 6:

[0519] The server notifies the user's device of the generated suggestions. The input is the customized suggestion content, which is the output of the suggestion generation module. Specifically, the server uses a notification API to send a push notification to the user's device. The output on the device is a list of suggested products and benefits. The user receives this suggestion and can immediately check the products that interest them. This series of actions improves user convenience and satisfaction.

[0520] (Application Example 2)

[0521] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0522] Traditional user behavior prediction systems have struggled to provide suggestions that reflect the emotional aspects of individual users. Furthermore, real-time product recommendations and special service offerings that respond to changes in user emotions are also difficult to achieve. This limits the improvement of the user experience and the achievement of higher customer satisfaction.

[0523] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0524] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning algorithm for predicting the user's future behavior based on the converted information; means for recognizing the user's emotional state using emotion analysis technology; and means for generating optimal suggestions based on the learning algorithm and the emotional state. This makes it possible to provide personalized suggestions and special discount information that take into account the emotional state and behavioral patterns of individual users.

[0525] "User's past behavior information" refers to data about the user's past actions, including purchase history, browsing history, review history, and other similar information.

[0526] "Storage" refers to saving collected data and making it available for later processing and analysis.

[0527] "Preprocessing" refers to a series of processes performed to convert raw data into an analyzable format, including imputing missing values ​​and removing outliers.

[0528] "Analyzable format" refers to a state where data is structured and can be effectively used with machine learning algorithms and analytical tools.

[0529] A "learning algorithm" refers to a computational method and program used to predict a user's future behavior based on collected data.

[0530] "Emotion analysis technology" refers to technologies for identifying emotions from user text data and behavioral data, and includes natural language processing and behavioral analysis.

[0531] "User emotional state" refers to information that indicates the type and intensity of emotions a user is experiencing at a given point in time.

[0532] "Optimal suggestions" are recommendations that are considered most beneficial or of interest to the user, generated based on the user's emotional state and predicted behavior.

[0533] "Special discount information" refers to information provided to increase user purchasing intent, offering goods or services at a lower price than the regular price.

[0534] The system for implementing this invention is configured as follows: First, the server collects information on the user's past behavior and stores it in a database. At this stage, the collected data includes the user's purchase history, browsing history, and review history. The stored data is then preprocessed to impute missing values ​​and remove outliers. Pandas, a Python library for data analysis, is used for this preprocessing.

[0535] The server uses machine learning algorithms to predict user behavior based on preprocessed data. This learning algorithm utilizes TensorFlow to build a model that accurately predicts the user's next action. Furthermore, the server employs sentiment analysis techniques, leveraging the NLTK natural language processing library to analyze collected user text data and recognize the user's emotional state.

[0536] Based on generated sentiment information and behavioral predictions, the server creates optimal suggestions for the user and notifies them via smartphone or web apps. These suggestions may include special discount information or recommendations for similar products. On the device, users can select from the information displayed on the screen and purchase or use it.

[0537] For example, when a user gives a positive review of a product, the server detects that emotion and suggests related products and accessories. It can also offer special discounts for negative feedback. Examples of prompts to input into the generative AI model include, "Please tell me how to suggest new related products based on products that have been highly rated in the past," and "Please write a procedure to detect positive emotions from user reviews and issue coupons based on them."

[0538] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0539] Step 1:

[0540] The server collects information about users' past behavior and stores it in a database. Inputs include user purchase history, browsing history, and review history. Storing this information in the database makes it available for later analysis.

[0541] Step 2:

[0542] The server preprocesses the stored data. The input is the raw data stored in the database, and the output is analyzable data with missing values ​​imputed and outliers removed. Specifically, Pandas is used to imputate missing values ​​and remove outliers.

[0543] Step 3:

[0544] The server uses machine learning algorithms based on preprocessed data to predict the user's future actions. The input for this step is preprocessed data, and the output is a model for predicting user behavior. TensorFlow is used to extract features from the data and build the learning model.

[0545] Step 4:

[0546] The server uses natural language processing (NLTK) to recognize the user's emotional state. Input is text data from user reviews and feedback, and output is the user's emotional state. Specifically, NLTK is used to analyze the text data and identify the emotions.

[0547] Step 5:

[0548] The server generates optimal suggestions based on a learning model and emotional states. The inputs are behavioral prediction data and emotional information, and the output is user-optimized suggestions. These suggestions may include product recommendations and special discount information.

[0549] Step 6:

[0550] The server notifies the user's terminal of the generated suggestions. In this step, the terminal displays the suggestions, and the user makes selections or purchases based on them. The input is the suggestion data sent from the server, and the output is the suggestion information that the user can see on their terminal.

[0551] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0552] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0553] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0554] [Fourth Embodiment]

[0555] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0556] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0557] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0558] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0559] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0560] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0561] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0562] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0563] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0564] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0565] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0566] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0567] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0568] This invention relates to a system that predicts future user behavior based on past user behavior information and provides users with optimal suggestions. This system consists of multiple servers and user terminals and aims to improve the user experience.

[0569] First, the server stores the user's past behavior information in a database. This includes purchase history, click history, and inquiry details. Next, the server preprocesses the stored data, performing data cleansing and formatting standardization to convert it into a format suitable for analysis.

[0570] The server trains a learning model using machine learning techniques based on pre-processed data. This model is designed to analyze user behavior patterns and predict their next actions. For example, in an e-commerce scenario, the server generates personalized product suggestions based on the user's purchase history, and in a subscription service scenario, it prepares special offers for users at risk of churn.

[0571] When suggestions are generated, they are notified to the user's device. When the user visits an e-commerce site, the device displays recommended products on the screen based on information from the server. The user can easily purchase the displayed products. Similarly, when a special offer is presented while using a subscription service, the user receives a notification about the offer and can check the available benefits.

[0572] For example, if a user has previously purchased multiple ebooks of a particular type, the server will automatically suggest new releases or products with special offers in a similar genre. Furthermore, when a user attempts to cancel a subscription service, the server will offer tailored offers to encourage continued use.

[0573] In this way, this system analyzes user behavior in real time and provides personalized suggestions quickly, enabling responses that meet the individual needs of each user. This allows companies to allocate resources efficiently and improve customer satisfaction.

[0574] The following describes the processing flow.

[0575] Step 1:

[0576] The server collects data on users' past behavior from various data sources and stores it in a database. This includes purchase history, website browsing history, inquiry history, and so on.

[0577] Step 2:

[0578] The server preprocesses user behavior data stored in the database. This preprocessing includes imputing missing values, removing outliers, and standardizing the data format.

[0579] Step 3:

[0580] The server uses pre-processed data to train a machine learning model. The model includes an algorithm that learns patterns in user behavior and predicts the next action.

[0581] Step 4:

[0582] The server uses a trained model to predict the user's future behavior in real time. Based on the prediction results, it generates optimal product suggestions and special offers.

[0583] Step 5:

[0584] The server sends generated suggestions and offers to the terminal. The terminal then displays this information appropriately on the user screen.

[0585] Step 6:

[0586] When users see suggestions and offers displayed on their devices, the system helps them select products they are interested in and take advantage of benefits.

[0587] Step 7:

[0588] The user's response is fed back to the server. The server stores this feedback and uses it to improve the accuracy of the model in subsequent uses.

[0589] (Example 1)

[0590] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0591] In today's information society, where a vast amount of information flows, users face the challenge of difficulty in accessing information and products that meet their specific needs. Furthermore, while companies are required to respond quickly to the diverse needs of their customers, the systems and methods for providing personalized service are currently insufficient.

[0592] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0593] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for processing the stored information and converting it into an analyzable format; and means for training a predictive model to predict the user's future behavior based on the converted information. This makes it possible to predict the user's behavior and provide personalized suggestions quickly and accurately.

[0594] "User's past behavior information" refers to historical data about the user's past actions, including purchase history, browsing history, and inquiries.

[0595] A "means of storage" refers to a mechanism for saving collected data and maintaining it in a format that can be used for subsequent processing.

[0596] An "analyzable format" is a format in which data has the necessary structure for effective analysis and processing.

[0597] A "predictive model" is an algorithm or mathematical model used to predict future actions or outcomes based on given data.

[0598] "Personalized recommendations" refer to suggestions that present the most suitable options and products based on each user's past behavioral patterns and current circumstances.

[0599] A "user device" is a device used by a user for direct operation, such as a computer or a smartphone.

[0600] This invention is a system that provides personalized suggestions based on the user's past behavior information. This system mainly consists of a server and a user terminal, and aims to improve the user experience.

[0601] First, the server collects and stores information about the user's past behavior in a database. This information includes purchase history, click history, and inquiry details. At this stage, a common RDBMS (Relational Database Management System) is often used as the database management system.

[0602] Next, the server processes the accumulated information and converts it into an analyzable format. Preprocessing such as data cleansing and formatting is performed using programming languages ​​such as Python and R. This makes the data suitable as input for machine learning.

[0603] The server then trains a predictive model based on the pre-processed data. This model is generated using machine learning algorithms (e.g., TensorFlow or PyTorch) to predict user behavior patterns. The trained model then generates individual suggestions for each user.

[0604] The generated suggestions are notified to the user's terminal. The terminal displays the suggestions received from the server on the user interface, making it easy for the user to review their contents. Communication methods such as HTTP requests and WebSockets are used, enabling real-time suggestions.

[0605] For example, if a user has previously purchased multiple ebooks of a specific genre, the server will suggest new releases in that genre. Furthermore, by inputting a prompt such as, "Based on user A's purchase history over the past year, suggest products they are most likely to purchase next," appropriate results can be obtained.

[0606] This system enables companies to achieve efficient and accurate digital marketing and customer relationships, contributing to improved user satisfaction.

[0607] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0608] Step 1:

[0609] The server collects and stores past user behavior information in a database. Specifically, it obtains purchase history, click history, and inquiry details via APIs and stores them in a relational database. The input data is the user's operation history, and by saving this in the database, it is made usable for subsequent analysis and recommendations. The output is a well-organized dataset for use in subsequent processing.

[0610] Step 2:

[0611] The server processes the collected data and converts it into an analyzable format. This process uses Python to perform data cleansing, imputation of missing values, removal of outliers, and normalization for format consistency. The raw data accumulated in step 1 is used as input, and the output is organized, analyzable data.

[0612] Step 3:

[0613] The server trains a predictive model based on the data, which has been converted into an analyzable format. Using the machine learning library TensorFlow, it builds a model that learns user behavior patterns from the data. In this process, the processed data from step 2 is used as input to train the model, and the trained predictive model is obtained as output.

[0614] Step 4:

[0615] The server uses a trained predictive model to generate personalized recommendations for each user. In this step, the model analyzes user patterns and creates optimal product and service recommendations by providing instructions through prompts on what to suggest next. The input is the predictive model and the latest user data, and the output is a personalized list of recommendations.

[0616] Step 5:

[0617] The server notifies the terminal of the generated suggestions. The terminal displays the received suggestions on the user interface, allowing the user to easily review the suggestions. The input is suggestion data sent from the server, and the output is the suggested content displayed on the screen to the user. Based on this, the user can make choices such as purchasing products or using services.

[0618] (Application Example 1)

[0619] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0620] In recent years, accurately understanding user interests and providing relevant information in real time has become crucial for online platforms. However, traditional methods have been unable to fully utilize users' diverse past behavioral data, making it difficult to effectively deliver personalized advertising. This has created challenges in improving the user experience and ensuring platform profitability.

[0621] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0622] In this invention, the server includes means for collecting and storing information on a user's past behavior, means for preprocessing the stored information and converting it into an analyzable format, and means for constructing a learning model for predicting the user's future behavior based on the converted information. This enables the real-time and automated delivery of targeted advertisements tailored to the individual interests of each user.

[0623] "User's past behavior information" refers to data about actions a user has taken in the past, including browsing history, purchase history, and click history.

[0624] "Preprocessing" refers to the methods used to prepare raw data into an analyzable format, including processes such as data cleansing and formatting standardization.

[0625] A "learning model" is a predictive model built using machine learning techniques, which is trained to predict future behavior based on the user's past actions.

[0626] "Optimal suggestions" refer to information and products that are tailored to the user's interests and needs, generated based on predictions made by the aforementioned learning model.

[0627] "Notification" refers to a means of transmitting generated suggestions to the user's terminal, and is a method of providing information to the user in real time.

[0628] "Targeted advertising" refers to advertisements that are personalized based on a user's interests and behavioral history and delivered to specific users.

[0629] The system for realizing this invention collects and stores information on the user's past behavior and performs processing to build a learning model that predicts the user's future behavior based on this information. The server uses a database for this purpose and preprocesses the collected behavioral data to convert it into an analyzable format. In particular, preprocessing such as data cleansing and formatting standardization prepares the data for machine learning. Software libraries such as Pandas and Scikit-learn are used for this purpose.

[0630] Subsequently, the server builds a machine learning model and generates targeted ads based on user behavior patterns. The model is trained and stored on the server using algorithms such as random forest. This model is continuously updated to improve prediction accuracy.

[0631] The device notifies the user of targeted advertisements sent from the server and displays personalized ads in real time. This allows users to instantly find products and services that match their interests. For example, if a user frequently shows interest in health foods, their smartphone will receive notifications about new promotions and discounts.

[0632] Examples of specific prompt messages are as follows:

[0633] "Based on users' past behavioral data, predict the product categories they are most likely to be interested in next, and generate relevant ads."

[0634] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0635] Step 1:

[0636] The server collects user activity history data and stores it in a database. The input is user behavior data, and the output is historical behavior information in an organized format within the database. This ensures that all behavioral records necessary for subsequent processing are stored.

[0637] Step 2:

[0638] The server performs preprocessing on the stored data. Specifically, it performs data cleansing and formatting standardization, with the input being the raw data in the database. Subsequent data processing transforms the output into a dataset in an analyzable format.

[0639] Step 3:

[0640] The server trains a machine learning model using pre-processed data. The input is a dataset suitable for analysis, and the output is a learning model that predicts user behavior. Specifically, it uses a random forest to learn the data and improve prediction accuracy.

[0641] Step 4:

[0642] The server uses the built-in learning model to predict the user's next actions and generates optimally targeted advertisements based on those predictions. The input is the user's current data and the learning model, and the output is the advertising information presented to the user. The server utilizes the generative AI model to create advertisements that accurately reflect the user's interests.

[0643] Step 5:

[0644] The device notifies the user of targeted advertisements sent from the server. The input is advertising information from the server, and the output is personalized advertisements displayed on the user's device. Users can passively view advertisements related to their interests.

[0645] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0646] This invention relates to a system that predicts future user behavior based on past user behavior information, recognizes the user's emotions using an emotion engine, and reflects these emotions in the generated suggestions. The aim of this system is to improve the user experience by providing suggestions that are more suitable to the user's needs through cooperation between the server and the user terminal.

[0647] The server first collects and stores the user's past behavior data in a database. This data includes past purchase history, website usage history, and inquiry history. The collected data is then preprocessed, including imputing missing values ​​and removing outliers. Finally, a machine learning model is trained using this preprocessed data. This model analyzes the user's behavior patterns and predicts their next actions.

[0648] The emotion engine analyzes user text and behavioral data to recognize user emotions. This emotion information is used to tailor suggestions based on predicted user behavior. For example, by incorporating natural language processing technology, emotions can be estimated from user reviews and feedback and used as feedback for product recommendations.

[0649] Suggestions, taking into account sentiment information generated on the server, are notified to the user's device. When the user views the suggestion on their device, the displayed products and benefits are shown on the screen, and the user can select or use them.

[0650] For example, if a user writes a positive review for a particular product, the sentiment engine can detect that sentiment and suggest other products in the same category or related accessories. It can also offer users special discounts or improved product offers based on negative feedback.

[0651] The introduction of this system enables highly personalized suggestions that respond to each user's individual emotions and behaviors, allowing companies to improve customer satisfaction and provide services more efficiently.

[0652] The following describes the processing flow.

[0653] Step 1:

[0654] The server collects data on the user's past behavior from a database. This data includes purchase history, website activity logs, and past inquiries.

[0655] Step 2:

[0656] The server preprocesses the collected data. Specifically, it imputes missing values, filters outliers, and standardizes the data format to create an easy-to-understand dataset for analysis.

[0657] Step 3:

[0658] The server uses pre-processed data to train a learning model that predicts user behavior. This model uses machine learning algorithms to predict the user's purchasing intent and potential next actions.

[0659] Step 4:

[0660] The server uses an emotion engine to analyze user emotions from text data (reviews and inquiries). For example, it uses natural language processing to classify emotions into positive, negative, and neutral.

[0661] Step 5:

[0662] Based on the output of the emotion engine, the server incorporates emotional information into the behavior predicted by the learning model to generate suggestions tailored to the user. For example, it might suggest new products to users with positive emotions and discount offers to users with negative emotions.

[0663] Step 6:

[0664] The server generates suggestions and sends them to the terminal. The terminal then visually presents these to the user, providing an interface that allows them to understand the details of the products and services.

[0665] Step 7:

[0666] The user checks the presented suggestions and purchases products or uses services as needed. The device reports the user's selections to the server and collects the feedback as data.

[0667] Step 8:

[0668] Based on the feedback, the server retrains the model and emotion engine with data to improve their accuracy, and uses this to inform future recommendations. This process results in a more personalized user experience.

[0669] (Example 2)

[0670] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0671] Conventional systems typically made suggestions based on the user's past behavior, making it difficult to provide optimized suggestions that fully considered the user's emotions and intentions. This limited the potential for improving the user experience and the accuracy of the suggestions.

[0672] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0673] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning model for predicting the user's future behavior based on the converted information; means for performing sentiment analysis on the learning model and recognizing the user's emotions; means for generating suggestions based on the user's behavior prediction, taking into account the results of the sentiment analysis; and means for notifying the user of the generated suggestions. This makes it possible to provide highly accurate suggestions that take the user's emotions into consideration.

[0674] "User's past behavior information" refers to data such as the user's past purchase history, website usage history, and inquiry history.

[0675] "Data preprocessing" refers to the process of converting data into an analyzable format by imputing missing values ​​and removing outliers.

[0676] A "learning model" refers to an algorithm or structure that analyzes user behavior patterns based on collected data and predicts future behavior.

[0677] "Sentiment analysis" refers to the process of recognizing emotions from user text and behavioral data using natural language processing technology and generating a numerical emotion score.

[0678] "Means for generating suggestions" refers to the logic and mechanisms for generating personalized product and service suggestions based on user behavior predictions and sentiment analysis results.

[0679] "Means of notifying of proposals" refers to methods and technologies for notifying users of generated proposals and displaying them on the user's device.

[0680] This invention relates to a system that links a server and a terminal to provide optimized suggestions tailored to user needs. The main components of this system include a data collection module, a data preprocessing module, a machine learning model, a sentiment analysis engine, a suggestion generation module, and a notification module.

[0681] The server collects and stores users' past behavior information in a database. For example, it collects purchase history and browsing history using online store APIs and web tracking technologies. This information is preprocessed using data analysis software such as Python or R to impute missing values ​​and remove outliers.

[0682] The pre-processed data is then fed into a machine learning model. This model is built using libraries such as TensorFlow and PyTorch, and analyzes user behavior patterns to predict future actions. This makes it possible to predict which products and services users are likely to be interested in next.

[0683] Furthermore, the sentiment analysis engine applies natural language processing techniques to user text and behavioral data to identify user emotions. This process incorporates generative AI models such as OpenAI to extract positive or negative emotions from review content and other data.

[0684] Based on sentiment analysis results and behavioral predictions, the suggestion generation module creates personalized suggestions. These suggestions are then sent from the server to the user's device. For example, if a user writes a positive review of a specific product, related products and accessories will be suggested. Conversely, if negative feedback is received, notifications of improved product offers or special discounts will be sent.

[0685] As a concrete example, by inputting a prompt message such as "Create a product suggestion best suited to user A based on their purchase history and reviews for the past month" into the AI ​​model, it becomes possible to create personalized suggestions for each user. In this way, the system of the present invention can provide highly accurate suggestions that take into account the user's emotions and behavior.

[0686] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0687] Step 1:

[0688] The server collects and stores users' past behavioral information in a database. It receives data such as purchase history and browsing history obtained from online store APIs and web tracking systems as input. This data is converted into a structured database format (e.g., CSV or SQL database format) and stored as output. Specifically, the server creates a data entry for each user, accumulating information in a consistent manner.

[0689] Step 2:

[0690] The server preprocesses the stored data and converts it into an analyzable format. The input is raw data stored in the database. Specific data processing involves using the Python Pandas library to impute missing values ​​(mean imputation and regression imputation) and remove outliers (statistical outlier detection). The output is a clean, analyzable dataset. This minimizes data noise and enables highly accurate analysis.

[0691] Step 3:

[0692] The server builds a machine learning model using preprocessed data and performs behavioral predictions. A cleaned dataset is used as input. Specifically, it trains a model using TensorFlow or PyTorch to learn user behavior patterns and predict the user's next action. The output is either the predicted user action or a list of products of interest. The model's accuracy is evaluated using a test dataset for validation.

[0693] Step 4:

[0694] The server uses an emotion analysis engine to recognize the user's emotions. The input data is text data such as user reviews and comments. For specific analysis, natural language processing techniques are utilized, for example, to calculate an emotion score from the text using a generative AI model. The output is a positive or negative emotion score based on the text written by the user. This allows the user's emotion-based judgment to be reflected in the recommendations.

[0695] Step 5:

[0696] The server generates suggestions based on behavioral predictions and sentiment scores. It receives predicted behavioral data and sentiment scores as input. Specifically, the suggestion model processes this data and uses prompts (e.g., "If user A's sentiment score is 8 or higher, suggest related products") to generate appropriate product and service combinations. The output is a customized suggestion tailored to the user. This enables more personalized suggestions based on the user's state.

[0697] Step 6:

[0698] The server notifies the user's device of the generated suggestions. The input is the customized suggestion content, which is the output of the suggestion generation module. Specifically, the server uses a notification API to send a push notification to the user's device. The output on the device is a list of suggested products and benefits. The user receives this suggestion and can immediately check the products that interest them. This series of actions improves user convenience and satisfaction.

[0699] (Application Example 2)

[0700] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0701] Traditional user behavior prediction systems have struggled to provide suggestions that reflect the emotional aspects of individual users. Furthermore, real-time product recommendations and special service offerings that respond to changes in user emotions are also difficult to achieve. This limits the improvement of the user experience and the achievement of higher customer satisfaction.

[0702] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0703] In this invention, the server includes means for collecting and storing information on the user's past behavior; means for preprocessing the stored information and converting it into an analyzable format; means for constructing a learning algorithm for predicting the user's future behavior based on the converted information; means for recognizing the user's emotional state using emotion analysis technology; and means for generating optimal suggestions based on the learning algorithm and the emotional state. This makes it possible to provide personalized suggestions and special discount information that take into account the emotional state and behavioral patterns of individual users.

[0704] "User's past behavior information" refers to data about the user's past actions, including purchase history, browsing history, review history, and other similar information.

[0705] "Storage" refers to saving collected data and making it available for later processing and analysis.

[0706] "Preprocessing" refers to a series of processes performed to convert raw data into an analyzable format, including imputing missing values ​​and removing outliers.

[0707] "Analyzable format" refers to a state where data is structured and can be effectively used with machine learning algorithms and analytical tools.

[0708] A "learning algorithm" refers to a computational method and program used to predict a user's future behavior based on collected data.

[0709] "Emotion analysis technology" refers to technologies for identifying emotions from user text data and behavioral data, and includes natural language processing and behavioral analysis.

[0710] "User emotional state" refers to information that indicates the type and intensity of emotions a user is experiencing at a given point in time.

[0711] "Optimal suggestions" are recommendations that are considered most beneficial or of interest to the user, generated based on the user's emotional state and predicted behavior.

[0712] "Special discount information" refers to information provided to increase user purchasing intent, offering goods or services at a lower price than the regular price.

[0713] The system for implementing this invention is configured as follows: First, the server collects information on the user's past behavior and stores it in a database. At this stage, the collected data includes the user's purchase history, browsing history, and review history. The stored data is then preprocessed to impute missing values ​​and remove outliers. Pandas, a Python library for data analysis, is used for this preprocessing.

[0714] The server uses machine learning algorithms to predict user behavior based on preprocessed data. This learning algorithm utilizes TensorFlow to build a model that accurately predicts the user's next action. Furthermore, the server employs sentiment analysis techniques, leveraging the NLTK natural language processing library to analyze collected user text data and recognize the user's emotional state.

[0715] Based on generated sentiment information and behavioral predictions, the server creates optimal suggestions for the user and notifies them via smartphone or web apps. These suggestions may include special discount information or recommendations for similar products. On the device, users can select from the information displayed on the screen and purchase or use it.

[0716] For example, when a user gives a positive review of a product, the server detects that emotion and suggests related products and accessories. It can also offer special discounts for negative feedback. Examples of prompts to input into the generative AI model include, "Please tell me how to suggest new related products based on products that have been highly rated in the past," and "Please write a procedure to detect positive emotions from user reviews and issue coupons based on them."

[0717] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0718] Step 1:

[0719] The server collects information about users' past behavior and stores it in a database. Inputs include user purchase history, browsing history, and review history. Storing this information in the database makes it available for later analysis.

[0720] Step 2:

[0721] The server preprocesses the stored data. The input is the raw data stored in the database, and the output is analyzable data with missing values ​​imputed and outliers removed. Specifically, Pandas is used to imputate missing values ​​and remove outliers.

[0722] Step 3:

[0723] The server uses machine learning algorithms based on preprocessed data to predict the user's future actions. The input for this step is preprocessed data, and the output is a model for predicting user behavior. TensorFlow is used to extract features from the data and build the learning model.

[0724] Step 4:

[0725] The server uses natural language processing (NLTK) to recognize the user's emotional state. Input is text data from user reviews and feedback, and output is the user's emotional state. Specifically, NLTK is used to analyze the text data and identify the emotions.

[0726] Step 5:

[0727] The server generates optimal suggestions based on a learning model and emotional states. The inputs are behavioral prediction data and emotional information, and the output is user-optimized suggestions. These suggestions may include product recommendations and special discount information.

[0728] Step 6:

[0729] The server notifies the user's terminal of the generated suggestions. In this step, the terminal displays the suggestions, and the user makes selections or purchases based on them. The input is the suggestion data sent from the server, and the output is the suggestion information that the user can see on their terminal.

[0730] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0731] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0732] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0733] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0734] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0735] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0736] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0737] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0738] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0739] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0740] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0741] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0742] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0743] 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.

[0744] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0745] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0746] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0747] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0748] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0749] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0750] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0751] The following is further disclosed regarding the embodiments described above.

[0752] (Claim 1)

[0753] A means for collecting and storing information on a user's past behavior,

[0754] Means for preprocessing the stored information and converting it into an analyzable format,

[0755] A means for constructing a learning model to predict the user's future actions based on the converted information,

[0756] A means for generating an optimal suggestion in accordance with the actions predicted by the aforementioned learning model,

[0757] A means for notifying the user of the generated proposal,

[0758] A system that includes this.

[0759] (Claim 2)

[0760] The system according to claim 1, characterized in that natural language processing technology is incorporated into the learning model.

[0761] (Claim 3)

[0762] The system according to claim 1, characterized in that the generated suggestions automate product recommendations.

[0763] "Example 1"

[0764] (Claim 1)

[0765] A means of collecting and storing information on a user's past behavior,

[0766] The means for processing the accumulated information and converting it into an analyzable format,

[0767] A means for training a predictive model to predict the user's future actions based on the converted information,

[0768] Means for generating personalized suggestions based on the actions predicted by the aforementioned predictive model,

[0769] Means for notifying the user device of the generated proposal,

[0770] A system that includes this.

[0771] (Claim 2)

[0772] The system according to claim 1, characterized in that natural language processing technology is incorporated into the prediction model.

[0773] (Claim 3)

[0774] The system according to claim 1, characterized in that the generated proposals automate product recommendations.

[0775] "Application Example 1"

[0776] (Claim 1)

[0777] A means for collecting and storing information on a user's past behavior,

[0778] Means for preprocessing the stored information and converting it into an analyzable format,

[0779] A means for constructing a learning model to predict the user's future actions based on the converted information,

[0780] A means for generating an optimal suggestion in accordance with the actions predicted by the aforementioned learning model,

[0781] A means for notifying the user of the generated proposal and displaying targeted advertisements,

[0782] A system that includes this.

[0783] (Claim 2)

[0784] The system according to claim 1, characterized in that natural language processing technology is incorporated into the learning model.

[0785] (Claim 3)

[0786] The system according to claim 1, characterized in that the generated proposal automates the display of advertisements.

[0787] "Example 2 of combining an emotion engine"

[0788] (Claim 1)

[0789] A means for collecting and storing information on a user's past behavior,

[0790] Means for preprocessing the stored information and converting it into an analyzable format,

[0791] A means for constructing a learning model to predict the user's future actions based on the converted information,

[0792] A means for performing sentiment analysis on the aforementioned learning model and recognizing the user's emotions,

[0793] A means for generating suggestions based on user behavior predictions, taking into account the results of the aforementioned sentiment analysis,

[0794] A means for notifying the user of the generated proposal,

[0795] A system that includes this.

[0796] (Claim 2)

[0797] The system according to claim 1, characterized in that natural language processing technology is incorporated into the learning model.

[0798] (Claim 3)

[0799] The system according to claim 1, characterized in that the generated suggestions automate product recommendations using sentiment analysis.

[0800] "Application example 2 when combining with an emotional engine"

[0801] (Claim 1)

[0802] A means for collecting and storing information on a user's past behavior,

[0803] Means for preprocessing the stored information and converting it into an analyzable format,

[0804] A means for constructing a learning algorithm to predict the user's future actions based on the converted information,

[0805] A means of recognizing a user's emotional state using emotion analysis technology,

[0806] A means for generating an optimal suggestion based on the learning algorithm and emotional state,

[0807] A means for notifying the user of the generated proposal,

[0808] A system that includes this.

[0809] (Claim 2)

[0810] The system according to claim 1, characterized in that natural language processing technology is incorporated into the learning algorithm.

[0811] (Claim 3)

[0812] The system according to claim 1, characterized in that the generated suggestions automate product presentation and further include special discount information based on the user's emotional state. [Explanation of Symbols]

[0813] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting and storing information on a user's past behavior, Means for preprocessing the stored information and converting it into an analyzable format, A means for constructing a learning model to predict the user's future actions based on the converted information, A means for generating an optimal suggestion in accordance with the actions predicted by the aforementioned learning model, A means for notifying the user of the generated proposal, A system that includes this.

2. The system according to claim 1, characterized in that natural language processing technology is incorporated into the learning model.

3. The system according to claim 1, characterized in that the generated suggestions automate product recommendations.

Images