System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting behavioral big-data; means for cleansing and pre-processing the collected behavioral big-data; means for using the pre-processed behavioral big-data to train a generative AI; means for using the generative AI to analyze customer needs and issues and generate optimal recommendations; and means for providing the generated recommendations to the customer.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Although the value of behavioral big data is increasing, many customers lack the expertise to utilize the data, preventing them from extracting its primary value. Small and medium-sized enterprises and sole proprietors, in particular, lack the resources and skills necessary for data analysis and strategic planning. Therefore, there is a need for methods that allow data to be utilized efficiently, even without specialized knowledge. Another issue that must be addressed is the lack of flexible data utilization proposal systems that can accommodate a variety of business situations. [Means for solving the problem]

[0005] The present invention solves these problems by providing a system that collects, cleans, and preprocesses behavioral big data, analyzes customer needs and issues using trained generative AI, and generates optimal proposals. This system includes a means for generating proposals using behavioral big data, and handles data such as search history, location information, and purchase history in particular. Furthermore, by using natural language processing to analyze customer needs, it is possible to efficiently utilize data and provide appropriate action plans even without specialized knowledge.

[0006] "Behavioral big data" refers to large datasets about the behavior of individuals or groups, including search history, location information, purchase history, and more.

[0007] "Generative AI" refers to artificial intelligence models that are trained to perform specific tasks using large amounts of data, including natural language processing and predictive models in particular.

[0008] "Cleansing" refers to the process of removing noise and unnecessary information from a dataset, making it suitable for analysis and model training.

[0009] "Preprocessing" refers to a series of processes for converting collected data into an analyzable format, specifically including normalizing the data format, filling in missing values, and correcting outliers.

[0010] "Natural language processing" refers to techniques that enable computers to understand, interpret, and generate human language, including text analysis and language model training.

[0011] "Proposals" refer to specific action plans and strategies provided to customers based on data analyzed by generative AI.

[0012] "Customers" refers to companies, organizations, universities, self-employed individuals, etc. that use this system and have specific needs or challenges regarding data utilization.

[0013] "API" stands for Application Program Interface, and refers to an interface that allows different software systems to interact with each other and exchange data and functions.

[0014] "Database query" refers to statements used to extract specific information from a database, including SQL and NoSQL-style query languages.

[0015] An "HTTP request" refers to a part of the communication protocol used by a client to request data from a server, and specifically includes request methods such as GET and POST. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[0038] System Overview

[0039] Data collection and preprocessing

[0040] 1. Data Collection

[0041] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history.

[0042] 2. Data cleansing and preprocessing

[0043] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[0044] Training generative AI

[0045] 3. Initializing the Model

[0046] The server loads the initial generative AI model and configures it for training.

[0047] 4. Training and optimization

[0048] The server uses the training data to train the generative AI, specifically by using deep learning algorithms to optimize the model's parameters.

[0049] Processing user inquiries

[0050] 5. Receiving User Input

[0051] Users input their issues and needs through a dialogue interface, such as, "I'd like to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic."

[0052] 6. Sending input data

[0053] The terminal transmits the user's input to the server.

[0054] Analysis and suggestions by generative AI

[0055] 7. Needs analysis

[0056] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[0057] 8. Data Reference and Proposal Generation

[0058] Based on the user's needs, the server refers to relevant behavioral big data and uses generative AI to generate optimal suggestions, such as extending business hours, improving the menu, or introducing delivery services.

[0059] 9. Submitting the proposal results

[0060] The server sends the generated proposal to the terminal.

[0061] Viewing and taking action on results

[0062] 10. Display of Offers

[0063] The terminal displays the proposals received from the server on a user interface.

[0064] 11. Taking Action

[0065] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0066] Specific examples

[0067] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[0068] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[0069] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers is increasing at night, but that the opportunity is being missed due to short business hours.

[0070] 3. Proposal Generation: Generate proposals such as extending business hours, improving the menu, or introducing new delivery services.

[0071] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[0072] Through the above process, an efficient data strategy utilizing behavioral big data and generative AI can be provided, making it possible to solve customer problems even without specialized knowledge.

[0073] The processing flow will be explained below.

[0074] Step 1:

[0075] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[0076] Step 2:

[0077] The server cleanses the collected data by filtering out unnecessary information and noise, filling in missing values, and standardizing the data format, thereby ensuring reliable data.

[0078] Step 3:

[0079] The server then preprocesses the cleansed data. Specifically, it converts the data into features and puts them in a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[0080] Step 4:

[0081] The server loads and configures the initial generative AI model, then trains the model using the preprocessed data, specifically by using deep learning algorithms to optimize the model parameters.

[0082] Step 5:

[0083] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[0084] Step 6:

[0085] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[0086] Step 7:

[0087] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[0088] Step 8:

[0089] The server references relevant behavioral big data based on the analyzed needs and retrieves the required data set using a database query.

[0090] Step 9:

[0091] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing delivery services.

[0092] Step 10:

[0093] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[0094] Step 11:

[0095] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[0096] Step 12:

[0097] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0098] Example 1

[0099] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0100] In recent years, the use of behavioral big data has been attracting attention in a wide range of fields. However, there are only a limited number of systems that can efficiently collect and cleanse massive amounts of data and then use generative AI to make useful suggestions. In particular, there are technical challenges in quickly and accurately providing appropriate suggestions that address users' specific needs and challenges.

[0101] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0102] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training the generative AI using the preprocessed data, means for receiving and analyzing customer input, means for referencing the received customer input based on a database and a generative AI model to obtain related data and generate optimal suggestions, and means for displaying the generated suggestions via a user interface, thereby enabling the server to quickly and accurately respond to specific needs and challenges of users and provide optimal suggestions.

[0103] "Behavioral big data" refers to large amounts of data about the behavior of individuals and groups, including, for example, search history, location information, and purchase history.

[0104] "Cleansing" is the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[0105] "Preprocessing" is the process of converting the cleansed data into a format that is easier to analyze and extracting features.

[0106] "Generative AI" refers to algorithms and models that use artificial intelligence technology to learn from data and generate new data.

[0107] "Customer needs and issues" refer to the specific problems and requests that customers have, which the system aims to solve and support.

[0108] "Analysis" is the process of analyzing data or information to understand its structure and trends.

[0109] "Proposals" refer to practical actions or measures derived based on the analysis results.

[0110] "User interface" is a general term for the screens and operating methods that allow users to interact with a system.

[0111] A "database" is a system for efficiently storing and searching data, and is used to manage behavioral big data.

[0112] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[0113] Data collection and preprocessing

[0114] The server collects behavioral big data through APIs. Specifically, it obtains data such as search history, location information, and purchase history from services such as Google Analytics and Foursquare. To do this, it uses tools such as Amazon Web Services (AWS) API Gateway. The collected data is stored in a local database.

[0115] Next, the server cleanses the collected data using Python's pandas library, completing missing values, removing duplicate data, filtering outliers, etc. After that, it uses Scikit-learn to standardize the data and perform feature transformation, including one-hot encoding.

[0116] Training generative AI

[0117] The server loads the initial generative AI model using TensorFlow or PyTorch and configures the training. Specifically, it sets hyperparameters such as the learning rate and batch size. Using the preprocessed data as input, it divides the training data into batches and feeds them into the generative AI, which then updates the model parameters using a deep learning algorithm. Amazon SageMaker can be used to train and optimize the model.

[0118] Processing user inquiries

[0119] Users input their needs and challenges through a dialogue interface. For example, they might enter "I want to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic" into an input field on a smartphone app. The device converts this input data into JSON format and sends it to the server via HTTPS. Specifically, it uses Amazon Lex or Google Dialogflow to accept user input in natural language.

[0120] Analysis and suggestions by generative AI

[0121] The server uses natural language processing to analyze the received user needs. It uses an NLP model, such as OpenAI GPT-3, to tokenize the user's input message and extract the subject and purpose of the sentence. It then queries relevant data from an Oracle database and uses a generative AI model to generate optimal suggestions. Examples include extending business hours, improving the menu, or introducing a new delivery service.

[0122] Sending and viewing proposal results

[0123] The generated suggestions are sent from the server to the device. The server converts the suggestions into JSON format and sends it to the device via HTTPS. The device analyzes the received data and displays it on the screen in a format that is easy for the user to understand. React Native is used to display the suggestions in a mobile application, providing specific instructions that the user can immediately follow.

[0124] Examples of concrete examples and prompts

[0125] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will process the following:

[0126] 1. Data collection: The server collects data such as the number of users at the nearest station and the usage status of restaurant delivery services in the surrounding area.

[0127] 2. Analysis: The server uses AI to analyze the data and understand user trends.

[0128] 3. Proposal Generation: The server generates proposals such as extending business hours, improving the menu, or introducing a new delivery service.

[0129] 4. Presentation: Display the suggestions on the device and provide specific instructions to the user.

[0130] Example prompts to be input to the generative AI model:

[0131] User input: Sales have been declining since the COVID-19 outbreak, so I would like to know specific actions to improve sales.

[0132] Prompt: Analyze the collected data and create effective proposals to increase user sales, such as extending business hours, improving the menu, or introducing a new delivery service.

[0133] In this way, the present invention is a system that utilizes behavioral big data and generative AI to quickly and accurately provide optimal suggestions for specific challenges faced by users.

[0134] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0135] Step 1:

[0136] The server collects the behavioral big data.

[0137] Input: API key and credentials obtained from Google Analytics, Foursquare, etc.

[0138] Data processing: Send API requests to retrieve data such as search history, location information, and purchase history.

[0139] Output: Behavioral big data stored in a local database.

[0140] Specifically, the server executes the API periodically and saves the acquired data in CSV or JSON format.

[0141] Step 2:

[0142] The server cleanses and pre-processes the collected data.

[0143] Input: Behavioral big data obtained in step 1.

[0144] Data processing: We use the Python pandas library to impute missing values, remove duplicates, and filter outliers. We then use Scikit-learn to standardize and one-hot encode the numerical data.

[0145] Output: A clean, pre-processed dataset.

[0146] Specifically, the server executes the processing script, loads the data into memory, cleanses and preprocesses it, and then saves it back to the database.

[0147] Step 3:

[0148] The server trains the generative AI using the preprocessed data.

[0149] Input: The cleaned dataset from step 2 and the initial generative AI model.

[0150] Data computation: Use TensorFlow or PyTorch to set model parameters, batch the training data, and apply deep learning algorithms.

[0151] Output: A trained generative AI model.

[0152] Specifically, the server sets the learning rate and batch size, monitors progress while training the model, and stops early or adjusts the learning rate as needed.

[0153] Step 4:

[0154] Users input their issues and needs through a dialogue interface.

[0155] Input: Text entered by the user into the dialogue interface (e.g., "I would like to know what measures can be taken to increase sales at my ramen shop after the COVID-19 outbreak").

[0156] Data processing: Convert input text into JSON format.

[0157] Output: The converted input data in JSON format.

[0158] Specifically, the user enters text into a smartphone app and taps the send button.

[0159] Step 5:

[0160] The terminal transmits the user's input to the server.

[0161] Input: The input data in JSON format obtained in step 4.

[0162] Data processing: JSON data is sent to the server via HTTPS protocol.

[0163] Output: The user's input data received by the server.

[0164] Specifically, the terminal generates an HTTPS request and sends data to a specific endpoint on the server.

[0165] Step 6:

[0166] The server analyzes the received user needs using natural language processing (NLP).

[0167] Input: The user input data received in step 5.

[0168] Data Computation: Uses NLP models such as OpenAI GPT-3 to tokenize input text and extract the subject and purpose of the sentence.

[0169] Output: Needs analysis results.

[0170] Specifically, the server runs an NLP analysis model to identify key keywords and context from the user's input data.

[0171] Step 7:

[0172] The server references relevant behavioral big data based on the user's needs and uses generative AI to generate optimal suggestions.

[0173] Input: Needs analysis results obtained in step 6 and behavioral big data in an Oracle database.

[0174] Data computation: Query relevant data and input it into generative AI models to generate optimal recommendations.

[0175] Output: The generated optimal proposal.

[0176] Specifically, the server runs a database query, inputs the retrieved data into a generative AI model, and creates recommendations.

[0177] Step 8:

[0178] The server sends the generated proposal to the terminal.

[0179] Input: The best proposal generated in step 7.

[0180] Data processing: The proposal content is converted into JSON format and sent to the terminal via HTTPS protocol.

[0181] Output: The proposal data received by the device.

[0182] Specifically, the server formats the proposal in JSON format, generates an HTTPS request, and sends it to the terminal.

[0183] Step 9:

[0184] The terminal displays the proposals received from the server on a user interface.

[0185] Input: Proposal data received in step 8.

[0186] Data processing: Analyze the received JSON data and display it on the user interface.

[0187] Output: The suggestions displayed in the user interface.

[0188] Specifically, the device analyzes the received data and displays the suggestions on the screen in list or graph format.

[0189] Step 10:

[0190] The user reviews the generative AI's suggestions and takes specific actions based on them.

[0191] Input: The suggestions shown in step 9.

[0192] Data processing: Implementing specific changes or new services based on your suggestions.

[0193] Output: The action taken (e.g., extending business hours, launching delivery service).

[0194] As a specific operation, the user can follow the guide within the app to perform specific actions and provide feedback of the results to the system.

[0195] (Application example 1)

[0196] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0197] Modern brick-and-mortar stores are required to effectively utilize customer behavior data to increase sales and customer satisfaction. However, the systems required to collect, analyze, and provide this data to customers are complex and require a great deal of effort and specialized knowledge. Effectively providing the information obtained through this process is also a challenge. While it is particularly important to provide effective product recommendations for the next time a customer visits the store based on their purchasing history and behavioral patterns, concrete methods for achieving this are still lacking.

[0198] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0199] In this invention, the server includes a means for collecting behavioral big data, a means for cleansing and preprocessing the collected data, a means for training a generating AI using the preprocessed data, a means for analyzing customer needs and issues using the generating AI and generating optimal proposals, and a means for providing the generated proposals to the customer via a smartphone application. This makes it possible to effectively analyze customer behavioral data and present recommended products for the customer's next visit to the store.

[0200] "Behavioral big data" refers to large amounts of data about customer behavior, including search history, location information, purchase history, and time spent in a store.

[0201] "Data cleansing" is the process of removing unnecessary information and noise from collected raw data and preparing it in a form suitable for analysis.

[0202] "Preprocessing" refers to a series of data transformations that convert raw data into a format suitable for analysis and model training.

[0203] "Generative AI" is a type of artificial intelligence that uses deep learning and machine learning techniques to generate optimal suggestions and predictions from data.

[0204] "Natural Language Processing (NLP)" is a technology that enables computers to understand, analyze, and generate human language.

[0205] A "smartphone application" is a software program that runs on a smartphone and provides specific functions and services to users.

[0206] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[0207] System Overview

[0208] The system has the following features:

[0209] 1. Data Collection

[0210] The server collects behavioral big data (e.g., search history, location information, purchase history, and store dwell time) through the API. This data includes customer behavior patterns and purchase history.

[0211] 2. Data cleansing and preprocessing

[0212] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[0213] 3. Training the generative AI

[0214] The server loads the initial generative AI model and configures it for learning, trains the generative AI using training data, and optimizes the model's parameters using deep learning algorithms.

[0215] 4. User Inquiry Processing

[0216] Users input their issues and needs through a dialogue interface on the smartphone application, such as, "What products should I recommend to you the next time I visit the store?"

[0217] 5. Analysis and suggestions by generative AI

[0218] The server uses natural language processing (NLP) to analyze the received user needs. This allows it to understand the specific challenges and requests the user faces. Next, it references the collected behavioral big data and uses generative AI to generate optimal suggestions. For example, it may suggest products based on the customer's purchasing history or the optimal product placement method within the store.

[0219] 6. Submitting the proposal results

[0220] The server sends the generated proposal to the device (smartphone), which displays the proposal to the user on the smartphone application.

[0221] Specific examples

[0222] For example, if a brick-and-mortar store owner types, "I want to know what products to recommend to customers the next time they visit," the system will do the following:

[0223] 1. Data collection: The server collects behavioral big data such as past purchase history, frequency of visits to the store, and length of stay in the store.

[0224] 2. Analysis: Generative AI analyzes the data and determines which products in a particular category are likely to be purchased during the customer's next visit.

[0225] 3. Suggestion generation: A suggestion is generated to recommend products in that category for the customer's next visit.

[0226] 4. Presentation: The smartphone application displays the generated proposal to the user and notifies the customer.

[0227] Prompt Sentence Examples

[0228] For example, the following prompts are used:

[0229] "Based on their recent purchase history, create a list of products to suggest to them the next time they visit."

[0230] "Please analyze in-store dwell time and traffic flow data and suggest optimal product placement."

[0231] This system enables efficient data analysis and proposals using behavioral big data and generative AI, even without specialized knowledge.

[0232] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0233] Step 1:

[0234] The server collects customer behavior big data (e.g., search history, location information, purchase history, and store stay time) through the API. This data includes customer behavior patterns and purchase history.

[0235] Input: Customer behavior data collected via API

[0236] Data processing: collecting and structuring data

[0237] Output: Behavioral big data in Pandas DataFrame format

[0238] Step 2:

[0239] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[0240] Input: Behavioral big data in Pandas DataFrame format

[0241] Data processing: Missing value handling, duplicate data removal, conversion to feature quantities

[0242] Output: Cleansed and preprocessed data

[0243] Step 3:

[0244] The server loads the initial generative AI model, trains it with the preprocessed data, and optimizes the model's parameters using deep learning algorithms.

[0245] Input: Cleansed and preprocessed data

[0246] Data Computation: Optimizing model parameters using deep learning

[0247] Output: A trained generative AI model

[0248] Step 4:

[0249] Users input their issues and needs through a dialogue interface on the smartphone application. For example, "What products should we recommend for you the next time you visit?"

[0250] Input: User needs and challenges (in natural language format)

[0251] Data processing: User input is sent to the server in text format

[0252] Output: User-entered data sent to the server

[0253] Step 5:

[0254] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[0255] Input: User needs and issues (text data)

[0256] Data Computing: Needs Analysis with Natural Language Processing

[0257] Output: Analyzed user needs data

[0258] Step 6:

[0259] The server uses generative AI to generate optimal suggestions based on the analyzed user needs, referencing the collected behavioral big data. For example, product suggestions based on a customer's purchasing history or optimal product placement methods in a store.

[0260] Input: Analyzed user needs data and behavioral big data

[0261] Data calculation: Proposal generation by generative AI

[0262] Output: Generated proposal data

[0263] Step 7:

[0264] The server sends the generated proposal to the terminal (smartphone), and displays the proposal to the user on the smartphone application.

[0265] Input: Generated proposal data

[0266] Data transmission: Sends the proposed data to a smartphone application.

[0267] Output: Suggestions displayed on the smartphone application

[0268] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0269] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user, and is specifically implemented in the following form.

[0270] System Overview

[0271] Data collection and preprocessing

[0272] 1. Data Collection

[0273] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[0274] 2. Data cleansing and preprocessing

[0275] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing. The preprocessed data is then stored in a database.

[0276] Training generative AI

[0277] 3. Initializing the Model

[0278] The server loads and configures the initial generative AI model.

[0279] 4. Training and optimization

[0280] The server uses the preprocessed data to train the generative AI, specifically by using deep learning algorithms to optimize model parameters.

[0281] Emotion engine integration

[0282] 5. Emotion engine integration

[0283] The server integrates an emotion engine and analyzes user input (text and voice) to recognize emotions.

[0284] Processing user inquiries

[0285] 6. Receiving User Input

[0286] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[0287] 7. Sending input data

[0288] The terminal sends the user's input to the server. The input is sent to the server in text format as an HTTP request.

[0289] Sentiment Analysis and Recommendations

[0290] 8. Needs and Sentiment Analysis

[0291] The server analyzes the received user input using natural language processing (NLP) and an emotion engine, thereby identifying the user's emotional state as well as their challenges and needs.

[0292] 9. Data Reference and Proposal Generation

[0293] The server references relevant behavioral big data based on the analyzed needs and emotions, executes database queries to retrieve the necessary data sets, and then uses generative AI to generate optimal suggestions. For example, it generates specific action plans such as extending business hours, improving the menu, or introducing delivery services.

[0294] 10. Adjusting suggestions based on emotions

[0295] The server adjusts the suggestions based on the user's emotional state as recognized by the emotion engine, for example, by changing the suggestions to be more detailed and comforting if the user is feeling anxious.

[0296] 11. Submitting the proposal results

[0297] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[0298] Viewing and taking action on results

[0299] 12. Display of Offers

[0300] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[0301] 13. Taking Action

[0302] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0303] Specific examples

[0304] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[0305] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[0306] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers at night is increasing, but that the business hours are short, meaning the business is missing out. At the same time, the emotion engine recognizes the owner's concerns.

[0307] 3. Proposal Generation: Generates proposals such as extending business hours, improving the menu, and introducing new delivery services. Based on the sentiment engine, the proposals also include wording that will ease customers' concerns.

[0308] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[0309] This concludes the description of the embodiment of the present invention. This system provides an efficient data strategy that utilizes behavioral big data, generative AI, and an emotion engine, making it possible to solve customer problems without requiring specialized knowledge.

[0310] The processing flow will be explained below.

[0311] Step 1:

[0312] The server collects behavioral big data (e.g., search history, location information, purchase history) through APIs, sends API requests, and stores the acquired data in storage.

[0313] Step 2:

[0314] The server cleanses the collected data by imputing missing values, correcting outliers, and filtering out unnecessary information and noise.

[0315] Step 3:

[0316] The server preprocesses the cleansed data, converting it into features and converting it into a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[0317] Step 4:

[0318] The server loads and configures the initial generative AI model, then trains the AI ​​model using the preprocessed data, and uses deep learning algorithms to optimize the model parameters.

[0319] Step 5:

[0320] The server integrates an emotion engine that analyzes user input (text and voice) and recognizes emotions.

[0321] Step 6:

[0322] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[0323] Step 7:

[0324] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[0325] Step 8:

[0326] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[0327] Step 9:

[0328] The server references relevant behavioral big data based on the analyzed needs, executes database queries, and retrieves the required data sets.

[0329] Step 10:

[0330] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing new delivery services.

[0331] Step 11:

[0332] The server tailors the suggestions based on the user's emotional state as recognized by the emotion engine, for example including reassuring words for a user who is feeling anxious.

[0333] Step 12:

[0334] The server sends the generated proposal to the device in a format such as JSON, as an HTTP response.

[0335] Step 13:

[0336] The terminal displays the proposals received from the server on the user interface, presenting the proposals in an easy-to-understand format.

[0337] Step 14:

[0338] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0339] Example 2

[0340] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0341] In today's data-driven society, companies are required to effectively utilize behavioral big data to gain a deep understanding of their customers' needs and challenges. However, conventional systems require time-consuming cleansing and preprocessing of collected data, and require significant effort to train generative AI models and analyze their needs. Furthermore, it is difficult to analyze customer sentiment and make recommendations based on it, which makes it difficult to improve customer satisfaction and build trusting relationships with customers in real-world business situations.

[0342] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0343] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the data, means for training a generation AI using the preprocessed data, means for analyzing user input and recognizing emotions, means for adjusting proposal content based on the recognized emotions, and means for providing generated proposals to customers. This enables efficient cleansing and preprocessing of collected data, enabling rapid model training and needs analysis. Furthermore, adjusting proposal content based on customer emotions enables more personalized proposals, improving customer satisfaction and building trust in business relationships.

[0344] "Behavioral big data" is a dataset that collects a large amount of information about user behavior, including search history, location information, purchase history, and so on.

[0345] "Data cleansing" refers to the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[0346] "Preprocessing" is the process of formatting data and converting it into features before analysis or model training.

[0347] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to generate new data and information from input data.

[0348] An "emotion engine" is a system that uses natural language processing technology to analyze and recognize emotions from user input data (text and voice).

[0349] "Needs analysis" is the process of identifying what customers want and what issues they face based on their statements and data.

[0350] "Adjusting proposals" is the process of optimizing proposals provided to customers based on the analysis results and recognized emotional information.

[0351] "Providing" refers to the act of presenting the generated proposal to the customer in an appropriate format.

[0352] The present invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user. Specific embodiments for implementing the present invention are described below.

[0353] Data collection and preprocessing

[0354] The server first collects behavioral big data using APIs (e.g., location information API, purchase history API). This includes user search history, location information, purchase history, etc. The collected data is temporarily stored in storage. The server then cleanses the data and removes unnecessary information and noise. Preprocessing is completed by formatting the data and converting it into the required features. The preprocessed data is then stored in a database.

[0355] Training generative AI

[0356] The server initializes and trains the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch). Using the preprocessed data, the deep learning algorithm optimizes the model's parameters. This training process enables the generative AI to efficiently analyze customer needs and challenges.

[0357] Emotion engine integration

[0358] The server also integrates an emotion engine (e.g., IBM Watson Tone Analyzer) that uses natural language processing technology to recognize emotions from user input data (text and voice). The emotion engine identifies the emotion contained in the user input and passes that emotion information to the generative AI.

[0359] Processing user inquiries

[0360] Users use a dialogue interface (e.g., a web form or chatbot) to input their issues and needs. For example, they might input, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic." The device then sends the user's input to the server. The input is sent as an HTTP request to the server, where it is analyzed.

[0361] Sentiment Analysis and Suggestion Generation

[0362] The server analyzes the received user input using natural language processing technology and an emotion engine. This allows it to identify the user's needs, challenges, and even their emotions. It then references the necessary behavioral big data based on the analysis results and uses generative AI to generate optimal proposals. For example, it creates specific action plans, such as extending business hours or introducing new delivery services.

[0363] Tailoring suggestions based on emotions

[0364] The server adjusts the suggestions based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the suggestions will be wrapped in reassuring language and include specific, actionable actions.

[0365] Sending the proposal results

[0366] Finally, the server sends the generated proposal in JSON format to the device, which displays the proposal in a user interface and presents it to the user in an easy-to-understand manner.

[0367] Specific examples

[0368] If a ramen shop owner types in, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will act as follows:

[0369] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[0370] 2. Analysis: Generative AI analyzes the data and determines that nighttime visitors are increasing but that short opening hours are a missed opportunity. The emotion engine also recognizes the owner's concerns.

[0371] 3. Proposal Generation: Propose extended hours, improved menu items, and new delivery services, along with language to ease concerns.

[0372] 4. Presentation: The device displays the generated proposal to the owner and prompts them to take specific action.

[0373] Example prompt sentence:

[0374] "Please suggest specific actions to improve sales for a ramen shop whose sales have declined due to the COVID-19 pandemic."

[0375] This system allows for personalized recommendations based on customer needs and emotions, contributing to business success.

[0376] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0377] Program processing flow

[0378] Step 1:

[0379] Data collection

[0380] The server collects behavioral big data using APIs (e.g., location information API, purchase history API). For example, the server obtains user search history, location information, and purchase history through API requests.

[0381] Input: Raw data obtained from API (search history, location information, purchase history)

[0382] Output: The collected raw data is stored in the server storage.

[0383] Step 2:

[0384] Data Cleansing and Preprocessing

[0385] The server cleanses the collected data by removing duplicates, filling in missing data, and removing noise, and then formats the data as features.

[0386] Input: Raw data collected in step 1

[0387] Output: The cleansed and preprocessed data is stored in a database.

[0388] Step 3:

[0389] Model initialization

[0390] The server initializes the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch), loading the initialization configuration file and setting the necessary parameters.

[0391] Input: Initialization config file for training

[0392] Output: Initialized generative AI model

[0393] Step 4:

[0394] Training and Optimization

[0395] The server uses the preprocessed data to train the generative AI, specifically optimizing the model parameters (weights, biases) using a deep learning algorithm.

[0396] Input: Preprocessed data, initialized generative AI model

[0397] Output: Optimized generative AI model after training

[0398] Step 5:

[0399] Emotion engine collaboration

[0400] The server uses an emotion engine (e.g., natural language processing technology) to analyze text and voice input from the user and recognize emotions. The analysis results are passed to the generative AI.

[0401] Input: User input data (text or voice)

[0402] Output: Emotion analysis results

[0403] Step 6:

[0404] Receiving User Input

[0405] Users input their issues and needs using a dialogue interface (e.g., web form, chatbot), and the input information is sent from the terminal to the server.

[0406] Input: User's question or need (e.g., "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic.")

[0407] Output: HTTP request to the server

[0408] Step 7:

[0409] Sending input data

[0410] The terminal sends the user's input to the server as an HTTP request. The input is sent in text format.

[0411] Input: User input data (text)

[0412] Output: HTTP request sent to the server

[0413] Step 8:

[0414] Needs and Sentiment Analysis

[0415] The server analyzes the received user input using natural language processing technology and an emotion engine, thereby identifying the user's needs, challenges, and emotions.

[0416] Input: HTTP request sent to the server, sentiment analysis results

[0417] Output: Analysis results of needs and emotions

[0418] Step 9:

[0419] Data lookup and proposal generation

[0420] The server then references relevant behavioral big data based on the analysis results and uses generative AI to generate optimal proposals, such as action plans to extend business hours or introduce new delivery services.

[0421] Input: Needs and emotion analysis results, behavioral big data

[0422] Output: Optimal suggestions from generative AI

[0423] Step 10:

[0424] Tailoring suggestions based on emotions

[0425] The server adjusts the content of the suggestions based on the analysis results of the emotion engine. For example, if a user is feeling anxious, the server will make the suggestions more detailed and change the wording to give a sense of security.

[0426] Input: Optimal suggestions by generative AI, emotion analysis results

[0427] Output: Adjusted proposal

[0428] Step 11:

[0429] Sending the proposal results

[0430] The server sends the generated proposal in JSON format to the device, which receives it and displays it in its user interface.

[0431] Input: Adjusted proposal

[0432] Output: HTTP response to the device

[0433] Step 12:

[0434] View Suggestions

[0435] The device displays the proposals received from the server on the user interface in a way that is easy for the user to understand and intuitively understand.

[0436] Input: HTTP response from the server

[0437] Output: The suggestions displayed to the user

[0438] Step 13:

[0439] Execute Action

[0440] The user can review the displayed suggestions made by the generated AI and take specific actions based on them, such as setting new business hours, improving the menu, or launching a new delivery service.

[0441] Input: Displayed suggestion

[0442] Output: The specific actions taken

[0443] (Application example 2)

[0444] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0445] Conventional online shopping sites have systems that recommend products by collecting and analyzing customer behavior data, but they lack a means to recognize the emotional state of the customer and make more appropriate and personalized suggestions. This means that suggestions cannot be made that fully take into account the anxiety and excitement that customers feel, making it difficult to improve customer satisfaction.

[0446] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training a generation AI using the preprocessed data, means for analyzing customer needs and issues using the generation AI and generating optimal proposals, means for detecting user emotions using an emotion recognition engine, means for adjusting the content of the proposal based on the user's emotional state, and means for providing the generated proposals to the customer. This enables personalized proposals that take customer emotions into consideration, thereby increasing customer satisfaction.

[0447] (definition statement)

[0448] "Behavioral big data" refers to large amounts of data based on user behavior, including search history, location information, purchase history, and more.

[0449] "Cleansing" refers to the process of removing unnecessary information and noise from acquired data to improve the quality of the data.

[0450] "Preprocessing" refers to the process of formatting data into a format suitable for analysis, and includes the extraction of features.

[0451] "Generative AI" refers to artificial intelligence models that are trained using deep learning algorithms to generate optimal outputs from given input data.

[0452] An "emotion recognition engine" refers to technology that analyzes a user's input data (text or voice) and identifies their emotional state.

[0453] "Natural language processing (NLP)" is a technology for understanding, analyzing, and generating natural language, and refers to algorithms that interpret the meaning of sentences.

[0454] "Customer needs" refers to the products and services that users desire, or the related issues and demands.

[0455] "Suggestion tailoring" refers to optimizing generated suggestions based on the user's emotional state, which can include changing the tone and content of the suggestions.

[0456] "User's emotional state" refers to the psychological emotions (e.g., anxiety, excitement, sadness, joy) that a user is feeling at a particular moment.

[0457] "Personalization" refers to customization to meet the specific needs and preferences of individual users.

[0458] "Customer satisfaction" refers to an indicator that shows how satisfied customers are with the products and services provided.

[0459] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. The system is implemented in the following form.

[0460] Data collection and preprocessing

[0461] Collecting behavioral big data

[0462] The server collects big data on user behavior through the API. For example, it collects data such as the user's search history on an online shopping site, location information, and purchase history. This allows it to understand the user's behavioral patterns. Upon an API request, the data is sent to the server and stored in storage.

[0463] Data Cleansing and Preprocessing

[0464] The server cleanses the collected data, removing unnecessary information and noise, and formats the data to convert it into features. The preprocessed data is stored in a database for later analysis.

[0465] Training generative AI

[0466] Model initialization

[0467] The server loads and configures the initial generative AI model, which is then trained on the deep learning algorithms that will be used later.

[0468] Training and Optimization

[0469] The server uses the preprocessed data to train the generative AI, which optimizes the model parameters and enables it to generate optimal suggestions based on the user's behavioral data.

[0470] Emotion engine integration

[0471] emotion recognition

[0472] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The emotion recognition engine detects emotions contained in the text entered by the user and sends the results to the server.

[0473] Processing user inquiries

[0474] Receiving User Input

[0475] Users input their problems and needs through the dialogue interface of the smartphone application, for example, by entering a prompt sentence such as, "I'm looking for new running shoes, but I'm not sure which ones to get."

[0476] Sending input data

[0477] The terminal sends the user's input data to the server, which then processes the data as an HTTP request.

[0478] Needs and Sentiment Analysis

[0479] The server analyzes the received user input using natural language processing and emotion recognition engines, thereby identifying the user's emotional state as well as their challenges and needs.

[0480] Data lookup and proposal generation

[0481] The server then references relevant behavioral big data based on the analyzed needs and emotions. It retrieves the necessary data sets and uses generative AI to generate optimal recommendations. For example, if it recommends a specific product, it explains why that product is effective and presents options.

[0482] Tailoring suggestions based on emotions

[0483] The server tailors the suggestions based on the user's emotional state as recognized by the emotion recognition engine: for example, if the user expresses anxiety, the suggestion is worded in a reassuring way and includes detailed explanations.

[0484] Sending the proposal results

[0485] The server formats the generated suggestions in JSON format or similar and sends them to the device as an HTTP response, allowing appropriate suggestions to be provided to the user quickly.

[0486] Viewing and taking action on results

[0487] The terminal displays the suggestions received from the server on the user interface of the utility in an easy-to-understand manner for the user, who can then check the suggestions in detail and take specific action as necessary.

[0488] In this way, the entire system works together to make personalized suggestions that take the user's emotions into account, which is expected to improve customer satisfaction.

[0489] Examples of specific examples and prompts

[0490] For example, if a customer types in the app, "I'm looking for new running shoes, but I'm not sure which ones to get," the system will generate appropriate recommendations. Here's an example prompt:

[0491] I'm looking for new running shoes but I'm not sure which ones to get.

[0492] This invention effectively combines an emotion engine with generative AI, making it possible to make personalized suggestions that take emotions into account, something that was difficult to achieve with conventional systems.

[0493] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0494] Step 1:

[0495] The server collects user behavioral big data (search history, location information, purchase history) through API. Specifically, it sends API requests and saves the acquired data in storage. The input is the API request, and the output is the collected behavioral data.

[0496] Step 2:

[0497] The server cleanses the collected data, removing unnecessary information and noise. During this process, the data is shaped and converted into features. The input is the collected behavioral data, and the output is the cleansed, pre-processed data.

[0498] Step 3:

[0499] The server uses the preprocessed data to train the generative AI, specifically optimizing model parameters using a deep learning algorithm. The input is the preprocessed data, and the output is an optimized generative AI model.

[0500] Step 4:

[0501] The server uses a generative AI model to analyze customer needs and issues and generate optimal proposals. The input is user behavior data and the generative AI model, and the output is the generated proposal.

[0502] Step 5:

[0503] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The input is the user's input data, and the output is the recognized emotional state.

[0504] Step 6:

[0505] The server adjusts the generated suggestions based on the user's emotional state. Specifically, it changes the tone and content of the suggestions depending on the emotions detected by the emotion recognition engine. The input is the recognized emotional state and the generated suggestions, and the output is the adjusted suggestions.

[0506] Step 7:

[0507] A user inputs a problem or need through the dialogue interface of a smartphone application, for example, by entering a prompt statement such as "I'm looking for new running shoes, but I'm not sure which ones to get." The input is the prompt statement, and the output is the user input as text.

[0508] Step 8:

[0509] The terminal sends the user's input data to the server. This input data is sent to the server as an HTTP request and processed. The input is the user's text input, and the output is an HTTP request to the server.

[0510] Step 9:

[0511] The server performs needs and emotion analysis, generates optimal proposals, and sends the tailored proposals to the device. The inputs are the user's needs, behavioral data, emotional state, and the generative AI model, and the output is the sending of the proposals to the device.

[0512] Step 10:

[0513] The terminal displays the proposals received from the server on the user interface. The proposals are presented in a way that is easy for the user to understand. The input is the proposal sent from the server, and the output is the proposal displayed on the user interface.

[0514] Step 11:

[0515] The user reviews the generative AI's suggestions and takes specific actions based on them, such as making a decision to purchase the running shoes presented. The input is the suggestions displayed in the user interface, and the output is the user's specific actions.

[0516] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0517] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0518] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0519] [Second embodiment]

[0520] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0521] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0522] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0523] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0524] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0525] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0526] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0527] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0528] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0529] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0530] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0531] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0532] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[0533] System Overview

[0534] Data collection and preprocessing

[0535] 1. Data Collection

[0536] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history.

[0537] 2. Data cleansing and preprocessing

[0538] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[0539] Training generative AI

[0540] 3. Initializing the Model

[0541] The server loads the initial generative AI model and configures it for training.

[0542] 4. Training and optimization

[0543] The server uses the training data to train the generative AI, specifically by using deep learning algorithms to optimize the model's parameters.

[0544] Processing user inquiries

[0545] 5. Receiving User Input

[0546] Users input their issues and needs through a dialogue interface, such as, "I'd like to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic."

[0547] 6. Sending input data

[0548] The terminal transmits the user's input to the server.

[0549] Analysis and suggestions by generative AI

[0550] 7. Needs analysis

[0551] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[0552] 8. Data Reference and Proposal Generation

[0553] Based on the user's needs, the server refers to relevant behavioral big data and uses generative AI to generate optimal suggestions, such as extending business hours, improving the menu, or introducing delivery services.

[0554] 9. Submitting the proposal results

[0555] The server sends the generated proposal to the terminal.

[0556] Viewing and taking action on results

[0557] 10. Display of Offers

[0558] The terminal displays the proposals received from the server on a user interface.

[0559] 11. Taking Action

[0560] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0561] Specific examples

[0562] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[0563] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[0564] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers is increasing at night, but that the opportunity is being missed due to short business hours.

[0565] 3. Proposal Generation: Generate proposals such as extending business hours, improving the menu, or introducing new delivery services.

[0566] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[0567] Through the above process, an efficient data strategy utilizing behavioral big data and generative AI can be provided, making it possible to solve customer problems even without specialized knowledge.

[0568] The processing flow will be explained below.

[0569] Step 1:

[0570] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[0571] Step 2:

[0572] The server cleanses the collected data by filtering out unnecessary information and noise, filling in missing values, and standardizing the data format, thereby ensuring reliable data.

[0573] Step 3:

[0574] The server then preprocesses the cleansed data. Specifically, it converts the data into features and puts them in a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[0575] Step 4:

[0576] The server loads and configures the initial generative AI model, then trains the model using the preprocessed data, specifically by using deep learning algorithms to optimize the model parameters.

[0577] Step 5:

[0578] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[0579] Step 6:

[0580] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[0581] Step 7:

[0582] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[0583] Step 8:

[0584] The server references relevant behavioral big data based on the analyzed needs and retrieves the required data set using a database query.

[0585] Step 9:

[0586] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing delivery services.

[0587] Step 10:

[0588] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[0589] Step 11:

[0590] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[0591] Step 12:

[0592] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0593] Example 1

[0594] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0595] In recent years, the use of behavioral big data has been attracting attention in a wide range of fields. However, there are only a limited number of systems that can efficiently collect and cleanse massive amounts of data and then use generative AI to make useful suggestions. In particular, there are technical challenges in quickly and accurately providing appropriate suggestions that address users' specific needs and challenges.

[0596] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0597] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training the generative AI using the preprocessed data, means for receiving and analyzing customer input, means for referencing the received customer input based on a database and a generative AI model to obtain related data and generate optimal suggestions, and means for displaying the generated suggestions via a user interface, thereby enabling the server to quickly and accurately respond to specific needs and challenges of users and provide optimal suggestions.

[0598] "Behavioral big data" refers to large amounts of data about the behavior of individuals and groups, including, for example, search history, location information, and purchase history.

[0599] "Cleansing" is the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[0600] "Preprocessing" is the process of converting the cleansed data into a format that is easier to analyze and extracting features.

[0601] "Generative AI" refers to algorithms and models that use artificial intelligence technology to learn from data and generate new data.

[0602] "Customer needs and issues" refer to the specific problems and requests that customers have, which the system aims to solve and support.

[0603] "Analysis" is the process of analyzing data or information to understand its structure and trends.

[0604] "Proposals" refer to practical actions or measures derived based on the analysis results.

[0605] "User interface" is a general term for the screens and operating methods that allow users to interact with a system.

[0606] A "database" is a system for efficiently storing and searching data, and is used to manage behavioral big data.

[0607] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[0608] Data collection and preprocessing

[0609] The server collects behavioral big data through APIs. Specifically, it obtains data such as search history, location information, and purchase history from services such as Google Analytics and Foursquare. To do this, it uses tools such as Amazon Web Services (AWS) API Gateway. The collected data is stored in a local database.

[0610] Next, the server cleanses the collected data using Python's pandas library, completing missing values, removing duplicate data, filtering outliers, etc. After that, it uses Scikit-learn to standardize the data and perform feature transformation, including one-hot encoding.

[0611] Training generative AI

[0612] The server loads the initial generative AI model using TensorFlow or PyTorch and configures the training. Specifically, it sets hyperparameters such as the learning rate and batch size. Using the preprocessed data as input, it divides the training data into batches and feeds them into the generative AI, which then updates the model parameters using a deep learning algorithm. Amazon SageMaker can be used to train and optimize the model.

[0613] Processing user inquiries

[0614] Users input their needs and challenges through a dialogue interface. For example, they might enter "I want to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic" into an input field on a smartphone app. The device converts this input data into JSON format and sends it to the server via HTTPS. Specifically, it uses Amazon Lex or Google Dialogflow to accept user input in natural language.

[0615] Analysis and suggestions by generative AI

[0616] The server uses natural language processing to analyze the received user needs. It uses an NLP model, such as OpenAI GPT-3, to tokenize the user's input message and extract the subject and purpose of the sentence. It then queries relevant data from an Oracle database and uses a generative AI model to generate optimal suggestions. Examples include extending business hours, improving the menu, or introducing a new delivery service.

[0617] Sending and viewing proposal results

[0618] The generated suggestions are sent from the server to the device. The server converts the suggestions into JSON format and sends it to the device via HTTPS. The device analyzes the received data and displays it on the screen in a format that is easy for the user to understand. React Native is used to display the suggestions in a mobile application, providing specific instructions that the user can immediately follow.

[0619] Examples of concrete examples and prompts

[0620] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will process the following:

[0621] 1. Data collection: The server collects data such as the number of users at the nearest station and the usage status of restaurant delivery services in the surrounding area.

[0622] 2. Analysis: The server uses AI to analyze the data and understand user trends.

[0623] 3. Proposal Generation: The server generates proposals such as extending business hours, improving the menu, or introducing a new delivery service.

[0624] 4. Presentation: Display the suggestions on the device and provide specific instructions to the user.

[0625] Example prompts to be input to the generative AI model:

[0626] User input: Sales have been declining since the COVID-19 outbreak, so I would like to know specific actions to improve sales.

[0627] Prompt: Analyze the collected data and create effective proposals to increase user sales, such as extending business hours, improving the menu, or introducing a new delivery service.

[0628] In this way, the present invention is a system that utilizes behavioral big data and generative AI to quickly and accurately provide optimal suggestions for specific challenges faced by users.

[0629] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0630] Step 1:

[0631] The server collects the behavioral big data.

[0632] Input: API key and credentials obtained from Google Analytics, Foursquare, etc.

[0633] Data processing: Send API requests to retrieve data such as search history, location information, and purchase history.

[0634] Output: Behavioral big data stored in a local database.

[0635] Specifically, the server executes the API periodically and saves the acquired data in CSV or JSON format.

[0636] Step 2:

[0637] The server cleanses and pre-processes the collected data.

[0638] Input: Behavioral big data obtained in step 1.

[0639] Data processing: We use the Python pandas library to impute missing values, remove duplicates, and filter outliers. We then use Scikit-learn to standardize and one-hot encode the numerical data.

[0640] Output: A clean, pre-processed dataset.

[0641] Specifically, the server executes the processing script, loads the data into memory, cleanses and preprocesses it, and then saves it back to the database.

[0642] Step 3:

[0643] The server trains the generative AI using the preprocessed data.

[0644] Input: The cleaned dataset from step 2 and the initial generative AI model.

[0645] Data computation: Use TensorFlow or PyTorch to set model parameters, batch the training data, and apply deep learning algorithms.

[0646] Output: A trained generative AI model.

[0647] Specifically, the server sets the learning rate and batch size, monitors progress while training the model, and stops early or adjusts the learning rate as needed.

[0648] Step 4:

[0649] Users input their issues and needs through a dialogue interface.

[0650] Input: Text entered by the user into the dialogue interface (e.g., "I would like to know what measures can be taken to increase sales at my ramen shop after the COVID-19 outbreak").

[0651] Data processing: Convert input text into JSON format.

[0652] Output: The converted input data in JSON format.

[0653] Specifically, the user enters text into a smartphone app and taps the send button.

[0654] Step 5:

[0655] The terminal transmits the user's input to the server.

[0656] Input: The input data in JSON format obtained in step 4.

[0657] Data processing: JSON data is sent to the server via HTTPS protocol.

[0658] Output: The user's input data received by the server.

[0659] Specifically, the terminal generates an HTTPS request and sends data to a specific endpoint on the server.

[0660] Step 6:

[0661] The server analyzes the received user needs using natural language processing (NLP).

[0662] Input: The user input data received in step 5.

[0663] Data Computation: Uses NLP models such as OpenAI GPT-3 to tokenize input text and extract the subject and purpose of the sentence.

[0664] Output: Needs analysis results.

[0665] Specifically, the server runs an NLP analysis model to identify key keywords and context from the user's input data.

[0666] Step 7:

[0667] The server references relevant behavioral big data based on the user's needs and uses generative AI to generate optimal suggestions.

[0668] Input: Needs analysis results obtained in step 6 and behavioral big data in an Oracle database.

[0669] Data computation: Query relevant data and input it into generative AI models to generate optimal recommendations.

[0670] Output: The generated optimal proposal.

[0671] Specifically, the server runs a database query, inputs the retrieved data into a generative AI model, and creates recommendations.

[0672] Step 8:

[0673] The server sends the generated proposal to the terminal.

[0674] Input: The best proposal generated in step 7.

[0675] Data processing: The proposal content is converted into JSON format and sent to the terminal via HTTPS protocol.

[0676] Output: The proposal data received by the device.

[0677] Specifically, the server formats the proposal in JSON format, generates an HTTPS request, and sends it to the terminal.

[0678] Step 9:

[0679] The terminal displays the proposals received from the server on a user interface.

[0680] Input: Proposal data received in step 8.

[0681] Data processing: Analyze the received JSON data and display it on the user interface.

[0682] Output: The suggestions displayed in the user interface.

[0683] Specifically, the device analyzes the received data and displays the suggestions on the screen in list or graph format.

[0684] Step 10:

[0685] The user reviews the generative AI's suggestions and takes specific actions based on them.

[0686] Input: The suggestions shown in step 9.

[0687] Data processing: Implementing specific changes or new services based on your suggestions.

[0688] Output: The action taken (e.g., extending business hours, launching delivery service).

[0689] As a specific operation, the user can follow the guide within the app to perform specific actions and provide feedback of the results to the system.

[0690] (Application example 1)

[0691] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0692] Modern brick-and-mortar stores are required to effectively utilize customer behavior data to increase sales and customer satisfaction. However, the systems required to collect, analyze, and provide this data to customers are complex and require a great deal of effort and specialized knowledge. Effectively providing the information obtained through this process is also a challenge. While it is particularly important to provide effective product recommendations for the next time a customer visits the store based on their purchasing history and behavioral patterns, concrete methods for achieving this are still lacking.

[0693] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0694] In this invention, the server includes a means for collecting behavioral big data, a means for cleansing and preprocessing the collected data, a means for training a generating AI using the preprocessed data, a means for analyzing customer needs and issues using the generating AI and generating optimal proposals, and a means for providing the generated proposals to the customer via a smartphone application. This makes it possible to effectively analyze customer behavioral data and present recommended products for the customer's next visit to the store.

[0695] "Behavioral big data" refers to large amounts of data about customer behavior, including search history, location information, purchase history, and time spent in a store.

[0696] "Data cleansing" is the process of removing unnecessary information and noise from collected raw data and preparing it in a form suitable for analysis.

[0697] "Preprocessing" refers to a series of data transformations that convert raw data into a format suitable for analysis and model training.

[0698] "Generative AI" is a type of artificial intelligence that uses deep learning and machine learning techniques to generate optimal suggestions and predictions from data.

[0699] "Natural Language Processing (NLP)" is a technology that enables computers to understand, analyze, and generate human language.

[0700] A "smartphone application" is a software program that runs on a smartphone and provides specific functions and services to users.

[0701] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[0702] System Overview

[0703] The system has the following features:

[0704] 1. Data Collection

[0705] The server collects behavioral big data (e.g., search history, location information, purchase history, and store dwell time) through the API. This data includes customer behavior patterns and purchase history.

[0706] 2. Data cleansing and preprocessing

[0707] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[0708] 3. Training the generative AI

[0709] The server loads the initial generative AI model and configures it for learning, trains the generative AI using training data, and optimizes the model's parameters using deep learning algorithms.

[0710] 4. User Inquiry Processing

[0711] Users input their issues and needs through a dialogue interface on the smartphone application, such as, "What products should I recommend to you the next time I visit the store?"

[0712] 5. Analysis and suggestions by generative AI

[0713] The server uses natural language processing (NLP) to analyze the received user needs. This allows it to understand the specific challenges and requests the user faces. Next, it references the collected behavioral big data and uses generative AI to generate optimal suggestions. For example, it may suggest products based on the customer's purchasing history or the optimal product placement method within the store.

[0714] 6. Submitting the proposal results

[0715] The server sends the generated proposal to the device (smartphone), which displays the proposal to the user on the smartphone application.

[0716] Specific examples

[0717] For example, if a brick-and-mortar store owner types, "I want to know what products to recommend to customers the next time they visit," the system will do the following:

[0718] 1. Data collection: The server collects behavioral big data such as past purchase history, frequency of visits to the store, and length of stay in the store.

[0719] 2. Analysis: Generative AI analyzes the data and determines which products in a particular category are likely to be purchased during the customer's next visit.

[0720] 3. Suggestion generation: A suggestion is generated to recommend products in that category for the customer's next visit.

[0721] 4. Presentation: The smartphone application displays the generated proposal to the user and notifies the customer.

[0722] Prompt Sentence Examples

[0723] For example, the following prompts are used:

[0724] "Based on their recent purchase history, create a list of products to suggest to them the next time they visit."

[0725] "Please analyze in-store dwell time and traffic flow data and suggest optimal product placement."

[0726] This system enables efficient data analysis and proposals using behavioral big data and generative AI, even without specialized knowledge.

[0727] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0728] Step 1:

[0729] The server collects customer behavior big data (e.g., search history, location information, purchase history, and store stay time) through the API. This data includes customer behavior patterns and purchase history.

[0730] Input: Customer behavior data collected via API

[0731] Data processing: collecting and structuring data

[0732] Output: Behavioral big data in Pandas DataFrame format

[0733] Step 2:

[0734] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[0735] Input: Behavioral big data in Pandas DataFrame format

[0736] Data processing: Missing value handling, duplicate data removal, conversion to feature quantities

[0737] Output: Cleansed and preprocessed data

[0738] Step 3:

[0739] The server loads the initial generative AI model, trains it with the preprocessed data, and optimizes the model's parameters using deep learning algorithms.

[0740] Input: Cleansed and preprocessed data

[0741] Data Computation: Optimizing model parameters using deep learning

[0742] Output: A trained generative AI model

[0743] Step 4:

[0744] Users input their issues and needs through a dialogue interface on the smartphone application. For example, "What products should we recommend for you the next time you visit?"

[0745] Input: User needs and challenges (in natural language format)

[0746] Data processing: User input is sent to the server in text format

[0747] Output: User-entered data sent to the server

[0748] Step 5:

[0749] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[0750] Input: User needs and issues (text data)

[0751] Data Computing: Needs Analysis with Natural Language Processing

[0752] Output: Analyzed user needs data

[0753] Step 6:

[0754] The server uses generative AI to generate optimal suggestions based on the analyzed user needs, referencing the collected behavioral big data. For example, product suggestions based on a customer's purchasing history or optimal product placement methods in a store.

[0755] Input: Analyzed user needs data and behavioral big data

[0756] Data calculation: Proposal generation by generative AI

[0757] Output: Generated proposal data

[0758] Step 7:

[0759] The server sends the generated proposal to the terminal (smartphone), and displays the proposal to the user on the smartphone application.

[0760] Input: Generated proposal data

[0761] Data transmission: Sends the proposed data to a smartphone application.

[0762] Output: Suggestions displayed on the smartphone application

[0763] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0764] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user, and is specifically implemented in the following form.

[0765] System Overview

[0766] Data collection and preprocessing

[0767] 1. Data Collection

[0768] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[0769] 2. Data cleansing and preprocessing

[0770] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing. The preprocessed data is then stored in a database.

[0771] Training generative AI

[0772] 3. Initializing the Model

[0773] The server loads and configures the initial generative AI model.

[0774] 4. Training and optimization

[0775] The server uses the preprocessed data to train the generative AI, specifically by using deep learning algorithms to optimize model parameters.

[0776] Emotion engine integration

[0777] 5. Emotion engine integration

[0778] The server integrates an emotion engine and analyzes user input (text and voice) to recognize emotions.

[0779] Processing user inquiries

[0780] 6. Receiving User Input

[0781] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[0782] 7. Sending input data

[0783] The terminal sends the user's input to the server. The input is sent to the server in text format as an HTTP request.

[0784] Sentiment Analysis and Recommendations

[0785] 8. Needs and Sentiment Analysis

[0786] The server analyzes the received user input using natural language processing (NLP) and an emotion engine, thereby identifying the user's emotional state as well as their challenges and needs.

[0787] 9. Data Reference and Proposal Generation

[0788] The server references relevant behavioral big data based on the analyzed needs and emotions, executes database queries to retrieve the necessary data sets, and then uses generative AI to generate optimal suggestions. For example, it generates specific action plans such as extending business hours, improving the menu, or introducing delivery services.

[0789] 10. Adjusting suggestions based on emotions

[0790] The server adjusts the suggestions based on the user's emotional state as recognized by the emotion engine, for example, by changing the suggestions to be more detailed and comforting if the user is feeling anxious.

[0791] 11. Submitting the proposal results

[0792] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[0793] Viewing and taking action on results

[0794] 12. Display of Offers

[0795] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[0796] 13. Taking Action

[0797] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0798] Specific examples

[0799] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[0800] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[0801] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers at night is increasing, but that the business hours are short, meaning the business is missing out. At the same time, the emotion engine recognizes the owner's concerns.

[0802] 3. Proposal Generation: Generates proposals such as extending business hours, improving the menu, and introducing new delivery services. Based on the sentiment engine, the proposals also include wording that will ease customers' concerns.

[0803] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[0804] This concludes the description of the embodiment of the present invention. This system provides an efficient data strategy that utilizes behavioral big data, generative AI, and an emotion engine, making it possible to solve customer problems without requiring specialized knowledge.

[0805] The processing flow will be explained below.

[0806] Step 1:

[0807] The server collects behavioral big data (e.g., search history, location information, purchase history) through APIs, sends API requests, and stores the acquired data in storage.

[0808] Step 2:

[0809] The server cleanses the collected data by imputing missing values, correcting outliers, and filtering out unnecessary information and noise.

[0810] Step 3:

[0811] The server preprocesses the cleansed data, converting it into features and converting it into a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[0812] Step 4:

[0813] The server loads and configures the initial generative AI model, then trains the AI ​​model using the preprocessed data, and uses deep learning algorithms to optimize the model parameters.

[0814] Step 5:

[0815] The server integrates an emotion engine that analyzes user input (text and voice) and recognizes emotions.

[0816] Step 6:

[0817] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[0818] Step 7:

[0819] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[0820] Step 8:

[0821] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[0822] Step 9:

[0823] The server references relevant behavioral big data based on the analyzed needs, executes database queries, and retrieves the required data sets.

[0824] Step 10:

[0825] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing new delivery services.

[0826] Step 11:

[0827] The server tailors the suggestions based on the user's emotional state as recognized by the emotion engine, for example including reassuring words for a user who is feeling anxious.

[0828] Step 12:

[0829] The server sends the generated proposal to the device in a format such as JSON, as an HTTP response.

[0830] Step 13:

[0831] The terminal displays the proposals received from the server on the user interface, presenting the proposals in an easy-to-understand format.

[0832] Step 14:

[0833] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[0834] Example 2

[0835] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0836] In today's data-driven society, companies are required to effectively utilize behavioral big data to gain a deep understanding of their customers' needs and challenges. However, conventional systems require time-consuming cleansing and preprocessing of collected data, and require significant effort to train generative AI models and analyze their needs. Furthermore, it is difficult to analyze customer sentiment and make recommendations based on it, which makes it difficult to improve customer satisfaction and build trusting relationships with customers in real-world business situations.

[0837] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0838] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the data, means for training a generation AI using the preprocessed data, means for analyzing user input and recognizing emotions, means for adjusting proposal content based on the recognized emotions, and means for providing generated proposals to customers. This enables efficient cleansing and preprocessing of collected data, enabling rapid model training and needs analysis. Furthermore, adjusting proposal content based on customer emotions enables more personalized proposals, improving customer satisfaction and building trust in business relationships.

[0839] "Behavioral big data" is a dataset that collects a large amount of information about user behavior, including search history, location information, purchase history, and so on.

[0840] "Data cleansing" refers to the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[0841] "Preprocessing" is the process of formatting data and converting it into features before analysis or model training.

[0842] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to generate new data and information from input data.

[0843] An "emotion engine" is a system that uses natural language processing technology to analyze and recognize emotions from user input data (text and voice).

[0844] "Needs analysis" is the process of identifying what customers want and what issues they face based on their statements and data.

[0845] "Adjusting proposals" is the process of optimizing proposals provided to customers based on the analysis results and recognized emotional information.

[0846] "Providing" refers to the act of presenting the generated proposal to the customer in an appropriate format.

[0847] The present invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user. Specific embodiments for implementing the present invention are described below.

[0848] Data collection and preprocessing

[0849] The server first collects behavioral big data using APIs (e.g., location information API, purchase history API). This includes user search history, location information, purchase history, etc. The collected data is temporarily stored in storage. The server then cleanses the data and removes unnecessary information and noise. Preprocessing is completed by formatting the data and converting it into the required features. The preprocessed data is then stored in a database.

[0850] Training generative AI

[0851] The server initializes and trains the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch). Using the preprocessed data, the deep learning algorithm optimizes the model's parameters. This training process enables the generative AI to efficiently analyze customer needs and challenges.

[0852] Emotion engine integration

[0853] The server also integrates an emotion engine (e.g., IBM Watson Tone Analyzer) that uses natural language processing technology to recognize emotions from user input data (text and voice). The emotion engine identifies the emotion contained in the user input and passes that emotion information to the generative AI.

[0854] Processing user inquiries

[0855] Users use a dialogue interface (e.g., a web form or chatbot) to input their issues and needs. For example, they might input, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic." The device then sends the user's input to the server. The input is sent as an HTTP request to the server, where it is analyzed.

[0856] Sentiment Analysis and Suggestion Generation

[0857] The server analyzes the received user input using natural language processing technology and an emotion engine. This allows it to identify the user's needs, challenges, and even their emotions. It then references the necessary behavioral big data based on the analysis results and uses generative AI to generate optimal proposals. For example, it creates specific action plans, such as extending business hours or introducing new delivery services.

[0858] Tailoring suggestions based on emotions

[0859] The server adjusts the suggestions based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the suggestions will be wrapped in reassuring language and include specific, actionable actions.

[0860] Sending the proposal results

[0861] Finally, the server sends the generated proposal in JSON format to the device, which displays the proposal in a user interface and presents it to the user in an easy-to-understand manner.

[0862] Specific examples

[0863] If a ramen shop owner types in, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will act as follows:

[0864] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[0865] 2. Analysis: Generative AI analyzes the data and determines that nighttime visitors are increasing but that short opening hours are a missed opportunity. The emotion engine also recognizes the owner's concerns.

[0866] 3. Proposal Generation: Propose extended hours, improved menu items, and new delivery services, along with language to ease concerns.

[0867] 4. Presentation: The device displays the generated proposal to the owner and prompts them to take specific action.

[0868] Example prompt sentence:

[0869] "Please suggest specific actions to improve sales for a ramen shop whose sales have declined due to the COVID-19 pandemic."

[0870] This system allows for personalized recommendations based on customer needs and emotions, contributing to business success.

[0871] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0872] Program processing flow

[0873] Step 1:

[0874] Data collection

[0875] The server collects behavioral big data using APIs (e.g., location information API, purchase history API). For example, the server obtains user search history, location information, and purchase history through API requests.

[0876] Input: Raw data obtained from API (search history, location information, purchase history)

[0877] Output: The collected raw data is stored in the server storage.

[0878] Step 2:

[0879] Data Cleansing and Preprocessing

[0880] The server cleanses the collected data by removing duplicates, filling in missing data, and removing noise, and then formats the data as features.

[0881] Input: Raw data collected in step 1

[0882] Output: The cleansed and preprocessed data is stored in a database.

[0883] Step 3:

[0884] Model initialization

[0885] The server initializes the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch), loading the initialization configuration file and setting the necessary parameters.

[0886] Input: Initialization config file for training

[0887] Output: Initialized generative AI model

[0888] Step 4:

[0889] Training and Optimization

[0890] The server uses the preprocessed data to train the generative AI, specifically optimizing the model parameters (weights, biases) using a deep learning algorithm.

[0891] Input: Preprocessed data, initialized generative AI model

[0892] Output: Optimized generative AI model after training

[0893] Step 5:

[0894] Emotion engine collaboration

[0895] The server uses an emotion engine (e.g., natural language processing technology) to analyze text and voice input from the user and recognize emotions. The analysis results are passed to the generative AI.

[0896] Input: User input data (text or voice)

[0897] Output: Emotion analysis results

[0898] Step 6:

[0899] Receiving User Input

[0900] Users input their issues and needs using a dialogue interface (e.g., web form, chatbot), and the input information is sent from the terminal to the server.

[0901] Input: User's question or need (e.g., "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic.")

[0902] Output: HTTP request to the server

[0903] Step 7:

[0904] Sending input data

[0905] The terminal sends the user's input to the server as an HTTP request. The input is sent in text format.

[0906] Input: User input data (text)

[0907] Output: HTTP request sent to the server

[0908] Step 8:

[0909] Needs and Sentiment Analysis

[0910] The server analyzes the received user input using natural language processing technology and an emotion engine, thereby identifying the user's needs, challenges, and emotions.

[0911] Input: HTTP request sent to the server, sentiment analysis results

[0912] Output: Analysis results of needs and emotions

[0913] Step 9:

[0914] Data lookup and proposal generation

[0915] The server then references relevant behavioral big data based on the analysis results and uses generative AI to generate optimal proposals, such as action plans to extend business hours or introduce new delivery services.

[0916] Input: Needs and emotion analysis results, behavioral big data

[0917] Output: Optimal suggestions from generative AI

[0918] Step 10:

[0919] Tailoring suggestions based on emotions

[0920] The server adjusts the content of the suggestions based on the analysis results of the emotion engine. For example, if a user is feeling anxious, the server will make the suggestions more detailed and change the wording to give a sense of security.

[0921] Input: Optimal suggestions by generative AI, emotion analysis results

[0922] Output: Adjusted proposal

[0923] Step 11:

[0924] Sending the proposal results

[0925] The server sends the generated proposal in JSON format to the device, which receives it and displays it in its user interface.

[0926] Input: Adjusted proposal

[0927] Output: HTTP response to the device

[0928] Step 12:

[0929] View Suggestions

[0930] The device displays the proposals received from the server on the user interface in a way that is easy for the user to understand and intuitively understand.

[0931] Input: HTTP response from the server

[0932] Output: The suggestions displayed to the user

[0933] Step 13:

[0934] Execute Action

[0935] The user can review the displayed suggestions made by the generated AI and take specific actions based on them, such as setting new business hours, improving the menu, or launching a new delivery service.

[0936] Input: Displayed suggestion

[0937] Output: The specific actions taken

[0938] (Application example 2)

[0939] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0940] Conventional online shopping sites have systems that recommend products by collecting and analyzing customer behavior data, but they lack a means to recognize the emotional state of the customer and make more appropriate and personalized suggestions. This means that suggestions cannot be made that fully take into account the anxiety and excitement that customers feel, making it difficult to improve customer satisfaction.

[0941] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training a generation AI using the preprocessed data, means for analyzing customer needs and issues using the generation AI and generating optimal proposals, means for detecting user emotions using an emotion recognition engine, means for adjusting the content of the proposal based on the user's emotional state, and means for providing the generated proposals to the customer. This enables personalized proposals that take customer emotions into consideration, thereby increasing customer satisfaction.

[0942] (definition statement)

[0943] "Behavioral big data" refers to large amounts of data based on user behavior, including search history, location information, purchase history, and more.

[0944] "Cleansing" refers to the process of removing unnecessary information and noise from acquired data to improve the quality of the data.

[0945] "Preprocessing" refers to the process of formatting data into a format suitable for analysis, and includes the extraction of features.

[0946] "Generative AI" refers to artificial intelligence models that are trained using deep learning algorithms to generate optimal outputs from given input data.

[0947] An "emotion recognition engine" refers to technology that analyzes a user's input data (text or voice) and identifies their emotional state.

[0948] "Natural language processing (NLP)" is a technology for understanding, analyzing, and generating natural language, and refers to algorithms that interpret the meaning of sentences.

[0949] "Customer needs" refers to the products and services that users desire, or the related issues and demands.

[0950] "Suggestion tailoring" refers to optimizing generated suggestions based on the user's emotional state, which can include changing the tone and content of the suggestions.

[0951] "User's emotional state" refers to the psychological emotions (e.g., anxiety, excitement, sadness, joy) that a user is feeling at a particular moment.

[0952] "Personalization" refers to customization to meet the specific needs and preferences of individual users.

[0953] "Customer satisfaction" refers to an indicator that shows how satisfied customers are with the products and services provided.

[0954] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. The system is implemented in the following form.

[0955] Data collection and preprocessing

[0956] Collecting behavioral big data

[0957] The server collects big data on user behavior through the API. For example, it collects data such as the user's search history on an online shopping site, location information, and purchase history. This allows it to understand the user's behavioral patterns. Upon an API request, the data is sent to the server and stored in storage.

[0958] Data Cleansing and Preprocessing

[0959] The server cleanses the collected data, removing unnecessary information and noise, and formats the data to convert it into features. The preprocessed data is stored in a database for later analysis.

[0960] Training generative AI

[0961] Model initialization

[0962] The server loads and configures the initial generative AI model, which is then trained on the deep learning algorithms that will be used later.

[0963] Training and Optimization

[0964] The server uses the preprocessed data to train the generative AI, which optimizes the model parameters and enables it to generate optimal suggestions based on the user's behavioral data.

[0965] Emotion engine integration

[0966] emotion recognition

[0967] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The emotion recognition engine detects emotions contained in the text entered by the user and sends the results to the server.

[0968] Processing user inquiries

[0969] Receiving User Input

[0970] Users input their problems and needs through the dialogue interface of the smartphone application, for example, by entering a prompt sentence such as, "I'm looking for new running shoes, but I'm not sure which ones to get."

[0971] Sending input data

[0972] The terminal sends the user's input data to the server, which then processes the data as an HTTP request.

[0973] Needs and Sentiment Analysis

[0974] The server analyzes the received user input using natural language processing and emotion recognition engines, thereby identifying the user's emotional state as well as their challenges and needs.

[0975] Data lookup and proposal generation

[0976] The server then references relevant behavioral big data based on the analyzed needs and emotions. It retrieves the necessary data sets and uses generative AI to generate optimal recommendations. For example, if it recommends a specific product, it explains why that product is effective and presents options.

[0977] Tailoring suggestions based on emotions

[0978] The server tailors the suggestions based on the user's emotional state as recognized by the emotion recognition engine: for example, if the user expresses anxiety, the suggestion is worded in a reassuring way and includes detailed explanations.

[0979] Sending the proposal results

[0980] The server formats the generated suggestions in JSON format or similar and sends them to the device as an HTTP response, allowing appropriate suggestions to be provided to the user quickly.

[0981] Viewing and taking action on results

[0982] The terminal displays the suggestions received from the server on the user interface of the utility in an easy-to-understand manner for the user, who can then check the suggestions in detail and take specific action as necessary.

[0983] In this way, the entire system works together to make personalized suggestions that take the user's emotions into account, which is expected to improve customer satisfaction.

[0984] Examples of specific examples and prompts

[0985] For example, if a customer types in the app, "I'm looking for new running shoes, but I'm not sure which ones to get," the system will generate appropriate recommendations. Here's an example prompt:

[0986] I'm looking for new running shoes but I'm not sure which ones to get.

[0987] This invention effectively combines an emotion engine with generative AI, making it possible to make personalized suggestions that take emotions into account, something that was difficult to achieve with conventional systems.

[0988] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0989] Step 1:

[0990] The server collects user behavioral big data (search history, location information, purchase history) through API. Specifically, it sends API requests and saves the acquired data in storage. The input is the API request, and the output is the collected behavioral data.

[0991] Step 2:

[0992] The server cleanses the collected data, removing unnecessary information and noise. During this process, the data is shaped and converted into features. The input is the collected behavioral data, and the output is the cleansed, pre-processed data.

[0993] Step 3:

[0994] The server uses the preprocessed data to train the generative AI, specifically optimizing model parameters using a deep learning algorithm. The input is the preprocessed data, and the output is an optimized generative AI model.

[0995] Step 4:

[0996] The server uses a generative AI model to analyze customer needs and issues and generate optimal proposals. The input is user behavior data and the generative AI model, and the output is the generated proposal.

[0997] Step 5:

[0998] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The input is the user's input data, and the output is the recognized emotional state.

[0999] Step 6:

[1000] The server adjusts the generated suggestions based on the user's emotional state. Specifically, it changes the tone and content of the suggestions depending on the emotions detected by the emotion recognition engine. The input is the recognized emotional state and the generated suggestions, and the output is the adjusted suggestions.

[1001] Step 7:

[1002] A user inputs a problem or need through the dialogue interface of a smartphone application, for example, by entering a prompt statement such as "I'm looking for new running shoes, but I'm not sure which ones to get." The input is the prompt statement, and the output is the user input as text.

[1003] Step 8:

[1004] The terminal sends the user's input data to the server. This input data is sent to the server as an HTTP request and processed. The input is the user's text input, and the output is an HTTP request to the server.

[1005] Step 9:

[1006] The server performs needs and emotion analysis, generates optimal proposals, and sends the tailored proposals to the device. The inputs are the user's needs, behavioral data, emotional state, and the generative AI model, and the output is the sending of the proposals to the device.

[1007] Step 10:

[1008] The terminal displays the proposals received from the server on the user interface. The proposals are presented in a way that is easy for the user to understand. The input is the proposal sent from the server, and the output is the proposal displayed on the user interface.

[1009] Step 11:

[1010] The user reviews the generative AI's suggestions and takes specific actions based on them, such as making a decision to purchase the running shoes presented. The input is the suggestions displayed in the user interface, and the output is the user's specific actions.

[1011] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1012] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1013] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1014] [Third embodiment]

[1015] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1016] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1018] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1019] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1020] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1022] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1023] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1024] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1025] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1026] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1027] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[1028] System Overview

[1029] Data collection and preprocessing

[1030] 1. Data Collection

[1031] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history.

[1032] 2. Data cleansing and preprocessing

[1033] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[1034] Training generative AI

[1035] 3. Initializing the Model

[1036] The server loads the initial generative AI model and configures it for training.

[1037] 4. Training and optimization

[1038] The server uses the training data to train the generative AI, specifically by using deep learning algorithms to optimize the model's parameters.

[1039] Processing user inquiries

[1040] 5. Receiving User Input

[1041] Users input their issues and needs through a dialogue interface, such as, "I'd like to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic."

[1042] 6. Sending input data

[1043] The terminal transmits the user's input to the server.

[1044] Analysis and suggestions by generative AI

[1045] 7. Needs analysis

[1046] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[1047] 8. Data Reference and Proposal Generation

[1048] Based on the user's needs, the server refers to relevant behavioral big data and uses generative AI to generate optimal suggestions, such as extending business hours, improving the menu, or introducing delivery services.

[1049] 9. Submitting the proposal results

[1050] The server sends the generated proposal to the terminal.

[1051] Viewing and taking action on results

[1052] 10. Display of Offers

[1053] The terminal displays the proposals received from the server on a user interface.

[1054] 11. Taking Action

[1055] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1056] Specific examples

[1057] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[1058] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[1059] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers is increasing at night, but that the opportunity is being missed due to short business hours.

[1060] 3. Proposal Generation: Generate proposals such as extending business hours, improving the menu, or introducing new delivery services.

[1061] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[1062] Through the above process, an efficient data strategy utilizing behavioral big data and generative AI can be provided, making it possible to solve customer problems even without specialized knowledge.

[1063] The processing flow will be explained below.

[1064] Step 1:

[1065] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[1066] Step 2:

[1067] The server cleanses the collected data by filtering out unnecessary information and noise, filling in missing values, and standardizing the data format, thereby ensuring reliable data.

[1068] Step 3:

[1069] The server then preprocesses the cleansed data. Specifically, it converts the data into features and puts them in a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[1070] Step 4:

[1071] The server loads and configures the initial generative AI model, then trains the model using the preprocessed data, specifically by using deep learning algorithms to optimize the model parameters.

[1072] Step 5:

[1073] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[1074] Step 6:

[1075] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[1076] Step 7:

[1077] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[1078] Step 8:

[1079] The server references relevant behavioral big data based on the analyzed needs and retrieves the required data set using a database query.

[1080] Step 9:

[1081] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing delivery services.

[1082] Step 10:

[1083] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[1084] Step 11:

[1085] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[1086] Step 12:

[1087] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1088] Example 1

[1089] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1090] In recent years, the use of behavioral big data has been attracting attention in a wide range of fields. However, there are only a limited number of systems that can efficiently collect and cleanse massive amounts of data and then use generative AI to make useful suggestions. In particular, there are technical challenges in quickly and accurately providing appropriate suggestions that address users' specific needs and challenges.

[1091] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1092] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training the generative AI using the preprocessed data, means for receiving and analyzing customer input, means for referencing the received customer input based on a database and a generative AI model to obtain related data and generate optimal suggestions, and means for displaying the generated suggestions via a user interface, thereby enabling the server to quickly and accurately respond to specific needs and challenges of users and provide optimal suggestions.

[1093] "Behavioral big data" refers to large amounts of data about the behavior of individuals and groups, including, for example, search history, location information, and purchase history.

[1094] "Cleansing" is the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[1095] "Preprocessing" is the process of converting the cleansed data into a format that is easier to analyze and extracting features.

[1096] "Generative AI" refers to algorithms and models that use artificial intelligence technology to learn from data and generate new data.

[1097] "Customer needs and issues" refer to the specific problems and requests that customers have, which the system aims to solve and support.

[1098] "Analysis" is the process of analyzing data or information to understand its structure and trends.

[1099] "Proposals" refer to practical actions or measures derived based on the analysis results.

[1100] "User interface" is a general term for the screens and operating methods that allow users to interact with a system.

[1101] A "database" is a system for efficiently storing and searching data, and is used to manage behavioral big data.

[1102] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[1103] Data collection and preprocessing

[1104] The server collects behavioral big data through APIs. Specifically, it obtains data such as search history, location information, and purchase history from services such as Google Analytics and Foursquare. To do this, it uses tools such as Amazon Web Services (AWS) API Gateway. The collected data is stored in a local database.

[1105] Next, the server cleanses the collected data using Python's pandas library, completing missing values, removing duplicate data, filtering outliers, etc. After that, it uses Scikit-learn to standardize the data and perform feature transformation, including one-hot encoding.

[1106] Training generative AI

[1107] The server loads the initial generative AI model using TensorFlow or PyTorch and configures the training. Specifically, it sets hyperparameters such as the learning rate and batch size. Using the preprocessed data as input, it divides the training data into batches and feeds them into the generative AI, which then updates the model parameters using a deep learning algorithm. Amazon SageMaker can be used to train and optimize the model.

[1108] Processing user inquiries

[1109] Users input their needs and challenges through a dialogue interface. For example, they might enter "I want to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic" into an input field on a smartphone app. The device converts this input data into JSON format and sends it to the server via HTTPS. Specifically, it uses Amazon Lex or Google Dialogflow to accept user input in natural language.

[1110] Analysis and suggestions by generative AI

[1111] The server uses natural language processing to analyze the received user needs. It uses an NLP model, such as OpenAI GPT-3, to tokenize the user's input message and extract the subject and purpose of the sentence. It then queries relevant data from an Oracle database and uses a generative AI model to generate optimal suggestions. Examples include extending business hours, improving the menu, or introducing a new delivery service.

[1112] Sending and viewing proposal results

[1113] The generated suggestions are sent from the server to the device. The server converts the suggestions into JSON format and sends it to the device via HTTPS. The device analyzes the received data and displays it on the screen in a format that is easy for the user to understand. React Native is used to display the suggestions in a mobile application, providing specific instructions that the user can immediately follow.

[1114] Examples of concrete examples and prompts

[1115] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will process the following:

[1116] 1. Data collection: The server collects data such as the number of users at the nearest station and the usage status of restaurant delivery services in the surrounding area.

[1117] 2. Analysis: The server uses AI to analyze the data and understand user trends.

[1118] 3. Proposal Generation: The server generates proposals such as extending business hours, improving the menu, or introducing a new delivery service.

[1119] 4. Presentation: Display the suggestions on the device and provide specific instructions to the user.

[1120] Example prompts to be input to the generative AI model:

[1121] User input: Sales have been declining since the COVID-19 outbreak, so I would like to know specific actions to improve sales.

[1122] Prompt: Analyze the collected data and create effective proposals to increase user sales, such as extending business hours, improving the menu, or introducing a new delivery service.

[1123] In this way, the present invention is a system that utilizes behavioral big data and generative AI to quickly and accurately provide optimal suggestions for specific challenges faced by users.

[1124] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1125] Step 1:

[1126] The server collects the behavioral big data.

[1127] Input: API key and credentials obtained from Google Analytics, Foursquare, etc.

[1128] Data processing: Send API requests to retrieve data such as search history, location information, and purchase history.

[1129] Output: Behavioral big data stored in a local database.

[1130] Specifically, the server executes the API periodically and saves the acquired data in CSV or JSON format.

[1131] Step 2:

[1132] The server cleanses and pre-processes the collected data.

[1133] Input: Behavioral big data obtained in step 1.

[1134] Data processing: We use the Python pandas library to impute missing values, remove duplicates, and filter outliers. We then use Scikit-learn to standardize and one-hot encode the numerical data.

[1135] Output: A clean, pre-processed dataset.

[1136] Specifically, the server executes the processing script, loads the data into memory, cleanses and preprocesses it, and then saves it back to the database.

[1137] Step 3:

[1138] The server trains the generative AI using the preprocessed data.

[1139] Input: The cleaned dataset from step 2 and the initial generative AI model.

[1140] Data computation: Use TensorFlow or PyTorch to set model parameters, batch the training data, and apply deep learning algorithms.

[1141] Output: A trained generative AI model.

[1142] Specifically, the server sets the learning rate and batch size, monitors progress while training the model, and stops early or adjusts the learning rate as needed.

[1143] Step 4:

[1144] Users input their issues and needs through a dialogue interface.

[1145] Input: Text entered by the user into the dialogue interface (e.g., "I would like to know what measures can be taken to increase sales at my ramen shop after the COVID-19 outbreak").

[1146] Data processing: Convert input text into JSON format.

[1147] Output: The converted input data in JSON format.

[1148] Specifically, the user enters text into a smartphone app and taps the send button.

[1149] Step 5:

[1150] The terminal transmits the user's input to the server.

[1151] Input: The input data in JSON format obtained in step 4.

[1152] Data processing: JSON data is sent to the server via HTTPS protocol.

[1153] Output: The user's input data received by the server.

[1154] Specifically, the terminal generates an HTTPS request and sends data to a specific endpoint on the server.

[1155] Step 6:

[1156] The server analyzes the received user needs using natural language processing (NLP).

[1157] Input: The user input data received in step 5.

[1158] Data Computation: Uses NLP models such as OpenAI GPT-3 to tokenize input text and extract the subject and purpose of the sentence.

[1159] Output: Needs analysis results.

[1160] Specifically, the server runs an NLP analysis model to identify key keywords and context from the user's input data.

[1161] Step 7:

[1162] The server references relevant behavioral big data based on the user's needs and uses generative AI to generate optimal suggestions.

[1163] Input: Needs analysis results obtained in step 6 and behavioral big data in an Oracle database.

[1164] Data computation: Query relevant data and input it into generative AI models to generate optimal recommendations.

[1165] Output: The generated optimal proposal.

[1166] Specifically, the server runs a database query, inputs the retrieved data into a generative AI model, and creates recommendations.

[1167] Step 8:

[1168] The server sends the generated proposal to the terminal.

[1169] Input: The best proposal generated in step 7.

[1170] Data processing: The proposal content is converted into JSON format and sent to the terminal via HTTPS protocol.

[1171] Output: The proposal data received by the device.

[1172] Specifically, the server formats the proposal in JSON format, generates an HTTPS request, and sends it to the terminal.

[1173] Step 9:

[1174] The terminal displays the proposals received from the server on a user interface.

[1175] Input: Proposal data received in step 8.

[1176] Data processing: Analyze the received JSON data and display it on the user interface.

[1177] Output: The suggestions displayed in the user interface.

[1178] Specifically, the device analyzes the received data and displays the suggestions on the screen in list or graph format.

[1179] Step 10:

[1180] The user reviews the generative AI's suggestions and takes specific actions based on them.

[1181] Input: The suggestions shown in step 9.

[1182] Data processing: Implementing specific changes or new services based on your suggestions.

[1183] Output: The action taken (e.g., extending business hours, launching delivery service).

[1184] As a specific operation, the user can follow the guide within the app to perform specific actions and provide feedback of the results to the system.

[1185] (Application example 1)

[1186] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1187] Modern brick-and-mortar stores are required to effectively utilize customer behavior data to increase sales and customer satisfaction. However, the systems required to collect, analyze, and provide this data to customers are complex and require a great deal of effort and specialized knowledge. Effectively providing the information obtained through this process is also a challenge. While it is particularly important to provide effective product recommendations for the next time a customer visits the store based on their purchasing history and behavioral patterns, concrete methods for achieving this are still lacking.

[1188] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1189] In this invention, the server includes a means for collecting behavioral big data, a means for cleansing and preprocessing the collected data, a means for training a generating AI using the preprocessed data, a means for analyzing customer needs and issues using the generating AI and generating optimal proposals, and a means for providing the generated proposals to the customer via a smartphone application. This makes it possible to effectively analyze customer behavioral data and present recommended products for the customer's next visit to the store.

[1190] "Behavioral big data" refers to large amounts of data about customer behavior, including search history, location information, purchase history, and time spent in a store.

[1191] "Data cleansing" is the process of removing unnecessary information and noise from collected raw data and preparing it in a form suitable for analysis.

[1192] "Preprocessing" refers to a series of data transformations that convert raw data into a format suitable for analysis and model training.

[1193] "Generative AI" is a type of artificial intelligence that uses deep learning and machine learning techniques to generate optimal suggestions and predictions from data.

[1194] "Natural Language Processing (NLP)" is a technology that enables computers to understand, analyze, and generate human language.

[1195] A "smartphone application" is a software program that runs on a smartphone and provides specific functions and services to users.

[1196] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[1197] System Overview

[1198] The system has the following features:

[1199] 1. Data Collection

[1200] The server collects behavioral big data (e.g., search history, location information, purchase history, and store dwell time) through the API. This data includes customer behavior patterns and purchase history.

[1201] 2. Data cleansing and preprocessing

[1202] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[1203] 3. Training the generative AI

[1204] The server loads the initial generative AI model and configures it for learning, trains the generative AI using training data, and optimizes the model's parameters using deep learning algorithms.

[1205] 4. User Inquiry Processing

[1206] Users input their issues and needs through a dialogue interface on the smartphone application, such as, "What products should I recommend to you the next time I visit the store?"

[1207] 5. Analysis and suggestions by generative AI

[1208] The server uses natural language processing (NLP) to analyze the received user needs. This allows it to understand the specific challenges and requests the user faces. Next, it references the collected behavioral big data and uses generative AI to generate optimal suggestions. For example, it may suggest products based on the customer's purchasing history or the optimal product placement method within the store.

[1209] 6. Submitting the proposal results

[1210] The server sends the generated proposal to the device (smartphone), which displays the proposal to the user on the smartphone application.

[1211] Specific examples

[1212] For example, if a brick-and-mortar store owner types, "I want to know what products to recommend to customers the next time they visit," the system will do the following:

[1213] 1. Data collection: The server collects behavioral big data such as past purchase history, frequency of visits to the store, and length of stay in the store.

[1214] 2. Analysis: Generative AI analyzes the data and determines which products in a particular category are likely to be purchased during the customer's next visit.

[1215] 3. Suggestion generation: A suggestion is generated to recommend products in that category for the customer's next visit.

[1216] 4. Presentation: The smartphone application displays the generated proposal to the user and notifies the customer.

[1217] Prompt Sentence Examples

[1218] For example, the following prompts are used:

[1219] "Based on their recent purchase history, create a list of products to suggest to them the next time they visit."

[1220] "Please analyze in-store dwell time and traffic flow data and suggest optimal product placement."

[1221] This system enables efficient data analysis and proposals using behavioral big data and generative AI, even without specialized knowledge.

[1222] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1223] Step 1:

[1224] The server collects customer behavior big data (e.g., search history, location information, purchase history, and store stay time) through the API. This data includes customer behavior patterns and purchase history.

[1225] Input: Customer behavior data collected via API

[1226] Data processing: collecting and structuring data

[1227] Output: Behavioral big data in Pandas DataFrame format

[1228] Step 2:

[1229] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[1230] Input: Behavioral big data in Pandas DataFrame format

[1231] Data processing: Missing value handling, duplicate data removal, conversion to feature quantities

[1232] Output: Cleansed and preprocessed data

[1233] Step 3:

[1234] The server loads the initial generative AI model, trains it with the preprocessed data, and optimizes the model's parameters using deep learning algorithms.

[1235] Input: Cleansed and preprocessed data

[1236] Data Computation: Optimizing model parameters using deep learning

[1237] Output: A trained generative AI model

[1238] Step 4:

[1239] Users input their issues and needs through a dialogue interface on the smartphone application. For example, "What products should we recommend for you the next time you visit?"

[1240] Input: User needs and challenges (in natural language format)

[1241] Data processing: User input is sent to the server in text format

[1242] Output: User-entered data sent to the server

[1243] Step 5:

[1244] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[1245] Input: User needs and issues (text data)

[1246] Data Computing: Needs Analysis with Natural Language Processing

[1247] Output: Analyzed user needs data

[1248] Step 6:

[1249] The server uses generative AI to generate optimal suggestions based on the analyzed user needs, referencing the collected behavioral big data. For example, product suggestions based on a customer's purchasing history or optimal product placement methods in a store.

[1250] Input: Analyzed user needs data and behavioral big data

[1251] Data calculation: Proposal generation by generative AI

[1252] Output: Generated proposal data

[1253] Step 7:

[1254] The server sends the generated proposal to the terminal (smartphone), and displays the proposal to the user on the smartphone application.

[1255] Input: Generated proposal data

[1256] Data transmission: Sends the proposed data to a smartphone application.

[1257] Output: Suggestions displayed on the smartphone application

[1258] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1259] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user, and is specifically implemented in the following form.

[1260] System Overview

[1261] Data collection and preprocessing

[1262] 1. Data Collection

[1263] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[1264] 2. Data cleansing and preprocessing

[1265] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing. The preprocessed data is then stored in a database.

[1266] Training generative AI

[1267] 3. Initializing the Model

[1268] The server loads and configures the initial generative AI model.

[1269] 4. Training and optimization

[1270] The server uses the preprocessed data to train the generative AI, specifically by using deep learning algorithms to optimize model parameters.

[1271] Emotion engine integration

[1272] 5. Emotion engine integration

[1273] The server integrates an emotion engine and analyzes user input (text and voice) to recognize emotions.

[1274] Processing user inquiries

[1275] 6. Receiving User Input

[1276] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[1277] 7. Sending input data

[1278] The terminal sends the user's input to the server. The input is sent to the server in text format as an HTTP request.

[1279] Sentiment Analysis and Recommendations

[1280] 8. Needs and Sentiment Analysis

[1281] The server analyzes the received user input using natural language processing (NLP) and an emotion engine, thereby identifying the user's emotional state as well as their challenges and needs.

[1282] 9. Data Reference and Proposal Generation

[1283] The server references relevant behavioral big data based on the analyzed needs and emotions, executes database queries to retrieve the necessary data sets, and then uses generative AI to generate optimal suggestions. For example, it generates specific action plans such as extending business hours, improving the menu, or introducing delivery services.

[1284] 10. Adjusting suggestions based on emotions

[1285] The server adjusts the suggestions based on the user's emotional state as recognized by the emotion engine, for example, by changing the suggestions to be more detailed and comforting if the user is feeling anxious.

[1286] 11. Submitting the proposal results

[1287] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[1288] Viewing and taking action on results

[1289] 12. Display of Offers

[1290] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[1291] 13. Taking Action

[1292] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1293] Specific examples

[1294] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[1295] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[1296] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers at night is increasing, but that the business hours are short, meaning the business is missing out. At the same time, the emotion engine recognizes the owner's concerns.

[1297] 3. Proposal Generation: Generates proposals such as extending business hours, improving the menu, and introducing new delivery services. Based on the sentiment engine, the proposals also include wording that will ease customers' concerns.

[1298] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[1299] This concludes the description of the embodiment of the present invention. This system provides an efficient data strategy that utilizes behavioral big data, generative AI, and an emotion engine, making it possible to solve customer problems without requiring specialized knowledge.

[1300] The processing flow will be explained below.

[1301] Step 1:

[1302] The server collects behavioral big data (e.g., search history, location information, purchase history) through APIs, sends API requests, and stores the acquired data in storage.

[1303] Step 2:

[1304] The server cleanses the collected data by imputing missing values, correcting outliers, and filtering out unnecessary information and noise.

[1305] Step 3:

[1306] The server preprocesses the cleansed data, converting it into features and converting it into a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[1307] Step 4:

[1308] The server loads and configures the initial generative AI model, then trains the AI ​​model using the preprocessed data, and uses deep learning algorithms to optimize the model parameters.

[1309] Step 5:

[1310] The server integrates an emotion engine that analyzes user input (text and voice) and recognizes emotions.

[1311] Step 6:

[1312] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[1313] Step 7:

[1314] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[1315] Step 8:

[1316] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[1317] Step 9:

[1318] The server references relevant behavioral big data based on the analyzed needs, executes database queries, and retrieves the required data sets.

[1319] Step 10:

[1320] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing new delivery services.

[1321] Step 11:

[1322] The server tailors the suggestions based on the user's emotional state as recognized by the emotion engine, for example including reassuring words for a user who is feeling anxious.

[1323] Step 12:

[1324] The server sends the generated proposal to the device in a format such as JSON, as an HTTP response.

[1325] Step 13:

[1326] The terminal displays the proposals received from the server on the user interface, presenting the proposals in an easy-to-understand format.

[1327] Step 14:

[1328] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1329] Example 2

[1330] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1331] In today's data-driven society, companies are required to effectively utilize behavioral big data to gain a deep understanding of their customers' needs and challenges. However, conventional systems require time-consuming cleansing and preprocessing of collected data, and require significant effort to train generative AI models and analyze their needs. Furthermore, it is difficult to analyze customer sentiment and make recommendations based on it, which makes it difficult to improve customer satisfaction and build trusting relationships with customers in real-world business situations.

[1332] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1333] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the data, means for training a generation AI using the preprocessed data, means for analyzing user input and recognizing emotions, means for adjusting proposal content based on the recognized emotions, and means for providing generated proposals to customers. This enables efficient cleansing and preprocessing of collected data, enabling rapid model training and needs analysis. Furthermore, adjusting proposal content based on customer emotions enables more personalized proposals, improving customer satisfaction and building trust in business relationships.

[1334] "Behavioral big data" is a dataset that collects a large amount of information about user behavior, including search history, location information, purchase history, and so on.

[1335] "Data cleansing" refers to the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[1336] "Preprocessing" is the process of formatting data and converting it into features before analysis or model training.

[1337] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to generate new data and information from input data.

[1338] An "emotion engine" is a system that uses natural language processing technology to analyze and recognize emotions from user input data (text and voice).

[1339] "Needs analysis" is the process of identifying what customers want and what issues they face based on their statements and data.

[1340] "Adjusting proposals" is the process of optimizing proposals provided to customers based on the analysis results and recognized emotional information.

[1341] "Providing" refers to the act of presenting the generated proposal to the customer in an appropriate format.

[1342] The present invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user. Specific embodiments for implementing the present invention are described below.

[1343] Data collection and preprocessing

[1344] The server first collects behavioral big data using APIs (e.g., location information API, purchase history API). This includes user search history, location information, purchase history, etc. The collected data is temporarily stored in storage. The server then cleanses the data and removes unnecessary information and noise. Preprocessing is completed by formatting the data and converting it into the required features. The preprocessed data is then stored in a database.

[1345] Training generative AI

[1346] The server initializes and trains the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch). Using the preprocessed data, the deep learning algorithm optimizes the model's parameters. This training process enables the generative AI to efficiently analyze customer needs and challenges.

[1347] Emotion engine integration

[1348] The server also integrates an emotion engine (e.g., IBM Watson Tone Analyzer) that uses natural language processing technology to recognize emotions from user input data (text and voice). The emotion engine identifies the emotion contained in the user input and passes that emotion information to the generative AI.

[1349] Processing user inquiries

[1350] Users use a dialogue interface (e.g., a web form or chatbot) to input their issues and needs. For example, they might input, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic." The device then sends the user's input to the server. The input is sent as an HTTP request to the server, where it is analyzed.

[1351] Sentiment Analysis and Suggestion Generation

[1352] The server analyzes the received user input using natural language processing technology and an emotion engine. This allows it to identify the user's needs, challenges, and even their emotions. It then references the necessary behavioral big data based on the analysis results and uses generative AI to generate optimal proposals. For example, it creates specific action plans, such as extending business hours or introducing new delivery services.

[1353] Tailoring suggestions based on emotions

[1354] The server adjusts the suggestions based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the suggestions will be wrapped in reassuring language and include specific, actionable actions.

[1355] Sending the proposal results

[1356] Finally, the server sends the generated proposal in JSON format to the device, which displays the proposal in a user interface and presents it to the user in an easy-to-understand manner.

[1357] Specific examples

[1358] If a ramen shop owner types in, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will act as follows:

[1359] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[1360] 2. Analysis: Generative AI analyzes the data and determines that nighttime visitors are increasing but that short opening hours are a missed opportunity. The emotion engine also recognizes the owner's concerns.

[1361] 3. Proposal Generation: Propose extended hours, improved menu items, and new delivery services, along with language to ease concerns.

[1362] 4. Presentation: The device displays the generated proposal to the owner and prompts them to take specific action.

[1363] Example prompt sentence:

[1364] "Please suggest specific actions to improve sales for a ramen shop whose sales have declined due to the COVID-19 pandemic."

[1365] This system allows for personalized recommendations based on customer needs and emotions, contributing to business success.

[1366] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1367] Program processing flow

[1368] Step 1:

[1369] Data collection

[1370] The server collects behavioral big data using APIs (e.g., location information API, purchase history API). For example, the server obtains user search history, location information, and purchase history through API requests.

[1371] Input: Raw data obtained from API (search history, location information, purchase history)

[1372] Output: The collected raw data is stored in the server storage.

[1373] Step 2:

[1374] Data Cleansing and Preprocessing

[1375] The server cleanses the collected data by removing duplicates, filling in missing data, and removing noise, and then formats the data as features.

[1376] Input: Raw data collected in step 1

[1377] Output: The cleansed and preprocessed data is stored in a database.

[1378] Step 3:

[1379] Model initialization

[1380] The server initializes the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch), loading the initialization configuration file and setting the necessary parameters.

[1381] Input: Initialization config file for training

[1382] Output: Initialized generative AI model

[1383] Step 4:

[1384] Training and Optimization

[1385] The server uses the preprocessed data to train the generative AI, specifically optimizing the model parameters (weights, biases) using a deep learning algorithm.

[1386] Input: Preprocessed data, initialized generative AI model

[1387] Output: Optimized generative AI model after training

[1388] Step 5:

[1389] Emotion engine collaboration

[1390] The server uses an emotion engine (e.g., natural language processing technology) to analyze text and voice input from the user and recognize emotions. The analysis results are passed to the generative AI.

[1391] Input: User input data (text or voice)

[1392] Output: Emotion analysis results

[1393] Step 6:

[1394] Receiving User Input

[1395] Users input their issues and needs using a dialogue interface (e.g., web form, chatbot), and the input information is sent from the terminal to the server.

[1396] Input: User's question or need (e.g., "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic.")

[1397] Output: HTTP request to the server

[1398] Step 7:

[1399] Sending input data

[1400] The terminal sends the user's input to the server as an HTTP request. The input is sent in text format.

[1401] Input: User input data (text)

[1402] Output: HTTP request sent to the server

[1403] Step 8:

[1404] Needs and Sentiment Analysis

[1405] The server analyzes the received user input using natural language processing technology and an emotion engine, thereby identifying the user's needs, challenges, and emotions.

[1406] Input: HTTP request sent to the server, sentiment analysis results

[1407] Output: Analysis results of needs and emotions

[1408] Step 9:

[1409] Data lookup and proposal generation

[1410] The server then references relevant behavioral big data based on the analysis results and uses generative AI to generate optimal proposals, such as action plans to extend business hours or introduce new delivery services.

[1411] Input: Needs and emotion analysis results, behavioral big data

[1412] Output: Optimal suggestions from generative AI

[1413] Step 10:

[1414] Tailoring suggestions based on emotions

[1415] The server adjusts the content of the suggestions based on the analysis results of the emotion engine. For example, if a user is feeling anxious, the server will make the suggestions more detailed and change the wording to give a sense of security.

[1416] Input: Optimal suggestions by generative AI, emotion analysis results

[1417] Output: Adjusted proposal

[1418] Step 11:

[1419] Sending the proposal results

[1420] The server sends the generated proposal in JSON format to the device, which receives it and displays it in its user interface.

[1421] Input: Adjusted proposal

[1422] Output: HTTP response to the device

[1423] Step 12:

[1424] View Suggestions

[1425] The device displays the proposals received from the server on the user interface in a way that is easy for the user to understand and intuitively understand.

[1426] Input: HTTP response from the server

[1427] Output: The suggestions displayed to the user

[1428] Step 13:

[1429] Execute Action

[1430] The user can review the displayed suggestions made by the generated AI and take specific actions based on them, such as setting new business hours, improving the menu, or launching a new delivery service.

[1431] Input: Displayed suggestion

[1432] Output: The specific actions taken

[1433] (Application example 2)

[1434] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1435] Conventional online shopping sites have systems that recommend products by collecting and analyzing customer behavior data, but they lack a means to recognize the emotional state of the customer and make more appropriate and personalized suggestions. This means that suggestions cannot be made that fully take into account the anxiety and excitement that customers feel, making it difficult to improve customer satisfaction.

[1436] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training a generation AI using the preprocessed data, means for analyzing customer needs and issues using the generation AI and generating optimal proposals, means for detecting user emotions using an emotion recognition engine, means for adjusting the content of the proposal based on the user's emotional state, and means for providing the generated proposals to the customer. This enables personalized proposals that take customer emotions into consideration, thereby increasing customer satisfaction.

[1437] (definition statement)

[1438] "Behavioral big data" refers to large amounts of data based on user behavior, including search history, location information, purchase history, and more.

[1439] "Cleansing" refers to the process of removing unnecessary information and noise from acquired data to improve the quality of the data.

[1440] "Preprocessing" refers to the process of formatting data into a format suitable for analysis, and includes the extraction of features.

[1441] "Generative AI" refers to artificial intelligence models that are trained using deep learning algorithms to generate optimal outputs from given input data.

[1442] An "emotion recognition engine" refers to technology that analyzes a user's input data (text or voice) and identifies their emotional state.

[1443] "Natural language processing (NLP)" is a technology for understanding, analyzing, and generating natural language, and refers to algorithms that interpret the meaning of sentences.

[1444] "Customer needs" refers to the products and services that users desire, or the related issues and demands.

[1445] "Suggestion tailoring" refers to optimizing generated suggestions based on the user's emotional state, which can include changing the tone and content of the suggestions.

[1446] "User's emotional state" refers to the psychological emotions (e.g., anxiety, excitement, sadness, joy) that a user is feeling at a particular moment.

[1447] "Personalization" refers to customization to meet the specific needs and preferences of individual users.

[1448] "Customer satisfaction" refers to an indicator that shows how satisfied customers are with the products and services provided.

[1449] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. The system is implemented in the following form.

[1450] Data collection and preprocessing

[1451] Collecting behavioral big data

[1452] The server collects big data on user behavior through the API. For example, it collects data such as the user's search history on an online shopping site, location information, and purchase history. This allows it to understand the user's behavioral patterns. Upon an API request, the data is sent to the server and stored in storage.

[1453] Data Cleansing and Preprocessing

[1454] The server cleanses the collected data, removing unnecessary information and noise, and formats the data to convert it into features. The preprocessed data is stored in a database for later analysis.

[1455] Training generative AI

[1456] Model initialization

[1457] The server loads and configures the initial generative AI model, which is then trained on the deep learning algorithms that will be used later.

[1458] Training and Optimization

[1459] The server uses the preprocessed data to train the generative AI, which optimizes the model parameters and enables it to generate optimal suggestions based on the user's behavioral data.

[1460] Emotion engine integration

[1461] emotion recognition

[1462] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The emotion recognition engine detects emotions contained in the text entered by the user and sends the results to the server.

[1463] Processing user inquiries

[1464] Receiving User Input

[1465] Users input their problems and needs through the dialogue interface of the smartphone application, for example, by entering a prompt sentence such as, "I'm looking for new running shoes, but I'm not sure which ones to get."

[1466] Sending input data

[1467] The terminal sends the user's input data to the server, which then processes the data as an HTTP request.

[1468] Needs and Sentiment Analysis

[1469] The server analyzes the received user input using natural language processing and emotion recognition engines, thereby identifying the user's emotional state as well as their challenges and needs.

[1470] Data lookup and proposal generation

[1471] The server then references relevant behavioral big data based on the analyzed needs and emotions. It retrieves the necessary data sets and uses generative AI to generate optimal recommendations. For example, if it recommends a specific product, it explains why that product is effective and presents options.

[1472] Tailoring suggestions based on emotions

[1473] The server tailors the suggestions based on the user's emotional state as recognized by the emotion recognition engine: for example, if the user expresses anxiety, the suggestion is worded in a reassuring way and includes detailed explanations.

[1474] Sending the proposal results

[1475] The server formats the generated suggestions in JSON format or similar and sends them to the device as an HTTP response, allowing appropriate suggestions to be provided to the user quickly.

[1476] Viewing and taking action on results

[1477] The terminal displays the suggestions received from the server on the user interface of the utility in an easy-to-understand manner for the user, who can then check the suggestions in detail and take specific action as necessary.

[1478] In this way, the entire system works together to make personalized suggestions that take the user's emotions into account, which is expected to improve customer satisfaction.

[1479] Examples of specific examples and prompts

[1480] For example, if a customer types in the app, "I'm looking for new running shoes, but I'm not sure which ones to get," the system will generate appropriate recommendations. Here's an example prompt:

[1481] I'm looking for new running shoes but I'm not sure which ones to get.

[1482] This invention effectively combines an emotion engine with generative AI, making it possible to make personalized suggestions that take emotions into account, something that was difficult to achieve with conventional systems.

[1483] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1484] Step 1:

[1485] The server collects user behavioral big data (search history, location information, purchase history) through API. Specifically, it sends API requests and saves the acquired data in storage. The input is the API request, and the output is the collected behavioral data.

[1486] Step 2:

[1487] The server cleanses the collected data, removing unnecessary information and noise. During this process, the data is shaped and converted into features. The input is the collected behavioral data, and the output is the cleansed, pre-processed data.

[1488] Step 3:

[1489] The server uses the preprocessed data to train the generative AI, specifically optimizing model parameters using a deep learning algorithm. The input is the preprocessed data, and the output is an optimized generative AI model.

[1490] Step 4:

[1491] The server uses a generative AI model to analyze customer needs and issues and generate optimal proposals. The input is user behavior data and the generative AI model, and the output is the generated proposal.

[1492] Step 5:

[1493] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The input is the user's input data, and the output is the recognized emotional state.

[1494] Step 6:

[1495] The server adjusts the generated suggestions based on the user's emotional state. Specifically, it changes the tone and content of the suggestions depending on the emotions detected by the emotion recognition engine. The input is the recognized emotional state and the generated suggestions, and the output is the adjusted suggestions.

[1496] Step 7:

[1497] A user inputs a problem or need through the dialogue interface of a smartphone application, for example, by entering a prompt statement such as "I'm looking for new running shoes, but I'm not sure which ones to get." The input is the prompt statement, and the output is the user input as text.

[1498] Step 8:

[1499] The terminal sends the user's input data to the server. This input data is sent to the server as an HTTP request and processed. The input is the user's text input, and the output is an HTTP request to the server.

[1500] Step 9:

[1501] The server performs needs and emotion analysis, generates optimal proposals, and sends the tailored proposals to the device. The inputs are the user's needs, behavioral data, emotional state, and the generative AI model, and the output is the sending of the proposals to the device.

[1502] Step 10:

[1503] The terminal displays the proposals received from the server on the user interface. The proposals are presented in a way that is easy for the user to understand. The input is the proposal sent from the server, and the output is the proposal displayed on the user interface.

[1504] Step 11:

[1505] The user reviews the generative AI's suggestions and takes specific actions based on them, such as making a decision to purchase the running shoes presented. The input is the suggestions displayed in the user interface, and the output is the user's specific actions.

[1506] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1507] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1508] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1509] [Fourth embodiment]

[1510] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1511] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1512] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1513] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1514] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1515] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1516] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1517] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1518] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1519] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1520] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1521] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1522] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1523] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[1524] System Overview

[1525] Data collection and preprocessing

[1526] 1. Data Collection

[1527] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history.

[1528] 2. Data cleansing and preprocessing

[1529] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[1530] Training generative AI

[1531] 3. Initializing the Model

[1532] The server loads the initial generative AI model and configures it for training.

[1533] 4. Training and optimization

[1534] The server uses the training data to train the generative AI, specifically by using deep learning algorithms to optimize the model's parameters.

[1535] Processing user inquiries

[1536] 5. Receiving User Input

[1537] Users input their issues and needs through a dialogue interface, such as, "I'd like to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic."

[1538] 6. Sending input data

[1539] The terminal transmits the user's input to the server.

[1540] Analysis and suggestions by generative AI

[1541] 7. Needs analysis

[1542] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[1543] 8. Data Reference and Proposal Generation

[1544] Based on the user's needs, the server refers to relevant behavioral big data and uses generative AI to generate optimal suggestions, such as extending business hours, improving the menu, or introducing delivery services.

[1545] 9. Submitting the proposal results

[1546] The server sends the generated proposal to the terminal.

[1547] Viewing and taking action on results

[1548] 10. Display of Offers

[1549] The terminal displays the proposals received from the server on a user interface.

[1550] 11. Taking Action

[1551] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1552] Specific examples

[1553] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[1554] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[1555] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers is increasing at night, but that the opportunity is being missed due to short business hours.

[1556] 3. Proposal Generation: Generate proposals such as extending business hours, improving the menu, or introducing new delivery services.

[1557] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[1558] Through the above process, an efficient data strategy utilizing behavioral big data and generative AI can be provided, making it possible to solve customer problems even without specialized knowledge.

[1559] The processing flow will be explained below.

[1560] Step 1:

[1561] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[1562] Step 2:

[1563] The server cleanses the collected data by filtering out unnecessary information and noise, filling in missing values, and standardizing the data format, thereby ensuring reliable data.

[1564] Step 3:

[1565] The server then preprocesses the cleansed data. Specifically, it converts the data into features and puts them in a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[1566] Step 4:

[1567] The server loads and configures the initial generative AI model, then trains the model using the preprocessed data, specifically by using deep learning algorithms to optimize the model parameters.

[1568] Step 5:

[1569] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[1570] Step 6:

[1571] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[1572] Step 7:

[1573] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[1574] Step 8:

[1575] The server references relevant behavioral big data based on the analyzed needs and retrieves the required data set using a database query.

[1576] Step 9:

[1577] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing delivery services.

[1578] Step 10:

[1579] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[1580] Step 11:

[1581] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[1582] Step 12:

[1583] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1584] Example 1

[1585] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1586] In recent years, the use of behavioral big data has been attracting attention in a wide range of fields. However, there are only a limited number of systems that can efficiently collect and cleanse massive amounts of data and then use generative AI to make useful suggestions. In particular, there are technical challenges in quickly and accurately providing appropriate suggestions that address users' specific needs and challenges.

[1587] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1588] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training the generative AI using the preprocessed data, means for receiving and analyzing customer input, means for referencing the received customer input based on a database and a generative AI model to obtain related data and generate optimal suggestions, and means for displaying the generated suggestions via a user interface, thereby enabling the server to quickly and accurately respond to specific needs and challenges of users and provide optimal suggestions.

[1589] "Behavioral big data" refers to large amounts of data about the behavior of individuals and groups, including, for example, search history, location information, and purchase history.

[1590] "Cleansing" is the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[1591] "Preprocessing" is the process of converting the cleansed data into a format that is easier to analyze and extracting features.

[1592] "Generative AI" refers to algorithms and models that use artificial intelligence technology to learn from data and generate new data.

[1593] "Customer needs and issues" refer to the specific problems and requests that customers have, which the system aims to solve and support.

[1594] "Analysis" is the process of analyzing data or information to understand its structure and trends.

[1595] "Proposals" refer to practical actions or measures derived based on the analysis results.

[1596] "User interface" is a general term for the screens and operating methods that allow users to interact with a system.

[1597] A "database" is a system for efficiently storing and searching data, and is used to manage behavioral big data.

[1598] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[1599] Data collection and preprocessing

[1600] The server collects behavioral big data through APIs. Specifically, it obtains data such as search history, location information, and purchase history from services such as Google Analytics and Foursquare. To do this, it uses tools such as Amazon Web Services (AWS) API Gateway. The collected data is stored in a local database.

[1601] Next, the server cleanses the collected data using Python's pandas library, completing missing values, removing duplicate data, filtering outliers, etc. After that, it uses Scikit-learn to standardize the data and perform feature transformation, including one-hot encoding.

[1602] Training generative AI

[1603] The server loads the initial generative AI model using TensorFlow or PyTorch and configures the training. Specifically, it sets hyperparameters such as the learning rate and batch size. Using the preprocessed data as input, it divides the training data into batches and feeds them into the generative AI, which then updates the model parameters using a deep learning algorithm. Amazon SageMaker can be used to train and optimize the model.

[1604] Processing user inquiries

[1605] Users input their needs and challenges through a dialogue interface. For example, they might enter "I want to know what measures I can take to increase sales at my ramen shop after the COVID-19 pandemic" into an input field on a smartphone app. The device converts this input data into JSON format and sends it to the server via HTTPS. Specifically, it uses Amazon Lex or Google Dialogflow to accept user input in natural language.

[1606] Analysis and suggestions by generative AI

[1607] The server uses natural language processing to analyze the received user needs. It uses an NLP model, such as OpenAI GPT-3, to tokenize the user's input message and extract the subject and purpose of the sentence. It then queries relevant data from an Oracle database and uses a generative AI model to generate optimal suggestions. Examples include extending business hours, improving the menu, or introducing a new delivery service.

[1608] Sending and viewing proposal results

[1609] The generated suggestions are sent from the server to the device. The server converts the suggestions into JSON format and sends it to the device via HTTPS. The device analyzes the received data and displays it on the screen in a format that is easy for the user to understand. React Native is used to display the suggestions in a mobile application, providing specific instructions that the user can immediately follow.

[1610] Examples of concrete examples and prompts

[1611] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will process the following:

[1612] 1. Data collection: The server collects data such as the number of users at the nearest station and the usage status of restaurant delivery services in the surrounding area.

[1613] 2. Analysis: The server uses AI to analyze the data and understand user trends.

[1614] 3. Proposal Generation: The server generates proposals such as extending business hours, improving the menu, or introducing a new delivery service.

[1615] 4. Presentation: Display the suggestions on the device and provide specific instructions to the user.

[1616] Example prompts to be input to the generative AI model:

[1617] User input: Sales have been declining since the COVID-19 outbreak, so I would like to know specific actions to improve sales.

[1618] Prompt: Analyze the collected data and create effective proposals to increase user sales, such as extending business hours, improving the menu, or introducing a new delivery service.

[1619] In this way, the present invention is a system that utilizes behavioral big data and generative AI to quickly and accurately provide optimal suggestions for specific challenges faced by users.

[1620] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1621] Step 1:

[1622] The server collects the behavioral big data.

[1623] Input: API key and credentials obtained from Google Analytics, Foursquare, etc.

[1624] Data processing: Send API requests to retrieve data such as search history, location information, and purchase history.

[1625] Output: Behavioral big data stored in a local database.

[1626] Specifically, the server executes the API periodically and saves the acquired data in CSV or JSON format.

[1627] Step 2:

[1628] The server cleanses and pre-processes the collected data.

[1629] Input: Behavioral big data obtained in step 1.

[1630] Data processing: We use the Python pandas library to impute missing values, remove duplicates, and filter outliers. We then use Scikit-learn to standardize and one-hot encode the numerical data.

[1631] Output: A clean, pre-processed dataset.

[1632] Specifically, the server executes the processing script, loads the data into memory, cleanses and preprocesses it, and then saves it back to the database.

[1633] Step 3:

[1634] The server trains the generative AI using the preprocessed data.

[1635] Input: The cleaned dataset from step 2 and the initial generative AI model.

[1636] Data computation: Use TensorFlow or PyTorch to set model parameters, batch the training data, and apply deep learning algorithms.

[1637] Output: A trained generative AI model.

[1638] Specifically, the server sets the learning rate and batch size, monitors progress while training the model, and stops early or adjusts the learning rate as needed.

[1639] Step 4:

[1640] Users input their issues and needs through a dialogue interface.

[1641] Input: Text entered by the user into the dialogue interface (e.g., "I would like to know what measures can be taken to increase sales at my ramen shop after the COVID-19 outbreak").

[1642] Data processing: Convert input text into JSON format.

[1643] Output: The converted input data in JSON format.

[1644] Specifically, the user enters text into a smartphone app and taps the send button.

[1645] Step 5:

[1646] The terminal transmits the user's input to the server.

[1647] Input: The input data in JSON format obtained in step 4.

[1648] Data processing: JSON data is sent to the server via HTTPS protocol.

[1649] Output: The user's input data received by the server.

[1650] Specifically, the terminal generates an HTTPS request and sends data to a specific endpoint on the server.

[1651] Step 6:

[1652] The server analyzes the received user needs using natural language processing (NLP).

[1653] Input: The user input data received in step 5.

[1654] Data Computation: Uses NLP models such as OpenAI GPT-3 to tokenize input text and extract the subject and purpose of the sentence.

[1655] Output: Needs analysis results.

[1656] Specifically, the server runs an NLP analysis model to identify key keywords and context from the user's input data.

[1657] Step 7:

[1658] The server references relevant behavioral big data based on the user's needs and uses generative AI to generate optimal suggestions.

[1659] Input: Needs analysis results obtained in step 6 and behavioral big data in an Oracle database.

[1660] Data computation: Query relevant data and input it into generative AI models to generate optimal recommendations.

[1661] Output: The generated optimal proposal.

[1662] Specifically, the server runs a database query, inputs the retrieved data into a generative AI model, and creates recommendations.

[1663] Step 8:

[1664] The server sends the generated proposal to the terminal.

[1665] Input: The best proposal generated in step 7.

[1666] Data processing: The proposal content is converted into JSON format and sent to the terminal via HTTPS protocol.

[1667] Output: The proposal data received by the device.

[1668] Specifically, the server formats the proposal in JSON format, generates an HTTPS request, and sends it to the terminal.

[1669] Step 9:

[1670] The terminal displays the proposals received from the server on a user interface.

[1671] Input: Proposal data received in step 8.

[1672] Data processing: Analyze the received JSON data and display it on the user interface.

[1673] Output: The suggestions displayed in the user interface.

[1674] Specifically, the device analyzes the received data and displays the suggestions on the screen in list or graph format.

[1675] Step 10:

[1676] The user reviews the generative AI's suggestions and takes specific actions based on them.

[1677] Input: The suggestions shown in step 9.

[1678] Data processing: Implementing specific changes or new services based on your suggestions.

[1679] Output: The action taken (e.g., extending business hours, launching delivery service).

[1680] As a specific operation, the user can follow the guide within the app to perform specific actions and provide feedback of the results to the system.

[1681] (Application example 1)

[1682] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1683] Modern brick-and-mortar stores are required to effectively utilize customer behavior data to increase sales and customer satisfaction. However, the systems required to collect, analyze, and provide this data to customers are complex and require a great deal of effort and specialized knowledge. Effectively providing the information obtained through this process is also a challenge. While it is particularly important to provide effective product recommendations for the next time a customer visits the store based on their purchasing history and behavioral patterns, concrete methods for achieving this are still lacking.

[1684] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1685] In this invention, the server includes a means for collecting behavioral big data, a means for cleansing and preprocessing the collected data, a means for training a generating AI using the preprocessed data, a means for analyzing customer needs and issues using the generating AI and generating optimal proposals, and a means for providing the generated proposals to the customer via a smartphone application. This makes it possible to effectively analyze customer behavioral data and present recommended products for the customer's next visit to the store.

[1686] "Behavioral big data" refers to large amounts of data about customer behavior, including search history, location information, purchase history, and time spent in a store.

[1687] "Data cleansing" is the process of removing unnecessary information and noise from collected raw data and preparing it in a form suitable for analysis.

[1688] "Preprocessing" refers to a series of data transformations that convert raw data into a format suitable for analysis and model training.

[1689] "Generative AI" is a type of artificial intelligence that uses deep learning and machine learning techniques to generate optimal suggestions and predictions from data.

[1690] "Natural Language Processing (NLP)" is a technology that enables computers to understand, analyze, and generate human language.

[1691] A "smartphone application" is a software program that runs on a smartphone and provides specific functions and services to users.

[1692] This invention relates to a system that collects behavioral big data and uses generative AI to analyze and make suggestions. This system consists of three elements: a server, a terminal, and a user.

[1693] System Overview

[1694] The system has the following features:

[1695] 1. Data Collection

[1696] The server collects behavioral big data (e.g., search history, location information, purchase history, and store dwell time) through the API. This data includes customer behavior patterns and purchase history.

[1697] 2. Data cleansing and preprocessing

[1698] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[1699] 3. Training the generative AI

[1700] The server loads the initial generative AI model and configures it for learning, trains the generative AI using training data, and optimizes the model's parameters using deep learning algorithms.

[1701] 4. User Inquiry Processing

[1702] Users input their issues and needs through a dialogue interface on the smartphone application, such as, "What products should I recommend to you the next time I visit the store?"

[1703] 5. Analysis and suggestions by generative AI

[1704] The server uses natural language processing (NLP) to analyze the received user needs. This allows it to understand the specific challenges and requests the user faces. Next, it references the collected behavioral big data and uses generative AI to generate optimal suggestions. For example, it may suggest products based on the customer's purchasing history or the optimal product placement method within the store.

[1705] 6. Submitting the proposal results

[1706] The server sends the generated proposal to the device (smartphone), which displays the proposal to the user on the smartphone application.

[1707] Specific examples

[1708] For example, if a brick-and-mortar store owner types, "I want to know what products to recommend to customers the next time they visit," the system will do the following:

[1709] 1. Data collection: The server collects behavioral big data such as past purchase history, frequency of visits to the store, and length of stay in the store.

[1710] 2. Analysis: Generative AI analyzes the data and determines which products in a particular category are likely to be purchased during the customer's next visit.

[1711] 3. Suggestion generation: A suggestion is generated to recommend products in that category for the customer's next visit.

[1712] 4. Presentation: The smartphone application displays the generated proposal to the user and notifies the customer.

[1713] Prompt Sentence Examples

[1714] For example, the following prompts are used:

[1715] "Based on their recent purchase history, create a list of products to suggest to them the next time they visit."

[1716] "Please analyze in-store dwell time and traffic flow data and suggest optimal product placement."

[1717] This system enables efficient data analysis and proposals using behavioral big data and generative AI, even without specialized knowledge.

[1718] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1719] Step 1:

[1720] The server collects customer behavior big data (e.g., search history, location information, purchase history, and store stay time) through the API. This data includes customer behavior patterns and purchase history.

[1721] Input: Customer behavior data collected via API

[1722] Data processing: collecting and structuring data

[1723] Output: Behavioral big data in Pandas DataFrame format

[1724] Step 2:

[1725] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing.

[1726] Input: Behavioral big data in Pandas DataFrame format

[1727] Data processing: Missing value handling, duplicate data removal, conversion to feature quantities

[1728] Output: Cleansed and preprocessed data

[1729] Step 3:

[1730] The server loads the initial generative AI model, trains it with the preprocessed data, and optimizes the model's parameters using deep learning algorithms.

[1731] Input: Cleansed and preprocessed data

[1732] Data Computation: Optimizing model parameters using deep learning

[1733] Output: A trained generative AI model

[1734] Step 4:

[1735] Users input their issues and needs through a dialogue interface on the smartphone application. For example, "What products should we recommend for you the next time you visit?"

[1736] Input: User needs and challenges (in natural language format)

[1737] Data processing: User input is sent to the server in text format

[1738] Output: User-entered data sent to the server

[1739] Step 5:

[1740] The server analyzes the received user needs using natural language processing (NLP), thereby understanding the specific challenges and requests the user faces.

[1741] Input: User needs and issues (text data)

[1742] Data Computing: Needs Analysis with Natural Language Processing

[1743] Output: Analyzed user needs data

[1744] Step 6:

[1745] The server uses generative AI to generate optimal suggestions based on the analyzed user needs, referencing the collected behavioral big data. For example, product suggestions based on a customer's purchasing history or optimal product placement methods in a store.

[1746] Input: Analyzed user needs data and behavioral big data

[1747] Data calculation: Proposal generation by generative AI

[1748] Output: Generated proposal data

[1749] Step 7:

[1750] The server sends the generated proposal to the terminal (smartphone), and displays the proposal to the user on the smartphone application.

[1751] Input: Generated proposal data

[1752] Data transmission: Sends the proposed data to a smartphone application.

[1753] Output: Suggestions displayed on the smartphone application

[1754] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1755] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user, and is specifically implemented in the following form.

[1756] System Overview

[1757] Data collection and preprocessing

[1758] 1. Data Collection

[1759] The server collects behavioral big data (e.g., search history, location information, purchase history) through the API. This data includes user behavior patterns and purchase history. The server sends an API request and saves the acquired data in storage.

[1760] 2. Data cleansing and preprocessing

[1761] The server cleanses the collected data, removing unnecessary information and noise, then formats the data and converts it into features as preprocessing. The preprocessed data is then stored in a database.

[1762] Training generative AI

[1763] 3. Initializing the Model

[1764] The server loads and configures the initial generative AI model.

[1765] 4. Training and optimization

[1766] The server uses the preprocessed data to train the generative AI, specifically by using deep learning algorithms to optimize model parameters.

[1767] Emotion engine integration

[1768] 5. Emotion engine integration

[1769] The server integrates an emotion engine and analyzes user input (text and voice) to recognize emotions.

[1770] Processing user inquiries

[1771] 6. Receiving User Input

[1772] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[1773] 7. Sending input data

[1774] The terminal sends the user's input to the server. The input is sent to the server in text format as an HTTP request.

[1775] Sentiment Analysis and Recommendations

[1776] 8. Needs and Sentiment Analysis

[1777] The server analyzes the received user input using natural language processing (NLP) and an emotion engine, thereby identifying the user's emotional state as well as their challenges and needs.

[1778] 9. Data Reference and Proposal Generation

[1779] The server references relevant behavioral big data based on the analyzed needs and emotions, executes database queries to retrieve the necessary data sets, and then uses generative AI to generate optimal suggestions. For example, it generates specific action plans such as extending business hours, improving the menu, or introducing delivery services.

[1780] 10. Adjusting suggestions based on emotions

[1781] The server adjusts the suggestions based on the user's emotional state as recognized by the emotion engine, for example, by changing the suggestions to be more detailed and comforting if the user is feeling anxious.

[1782] 11. Submitting the proposal results

[1783] The server sends the generated proposal to the device. The proposal is formatted in JSON format or similar and sent to the device as an HTTP response.

[1784] Viewing and taking action on results

[1785] 12. Display of Offers

[1786] The terminal displays the proposal received from the server on the user interface, and the proposal content is presented in a way that is easy for the user to understand.

[1787] 13. Taking Action

[1788] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1789] Specific examples

[1790] For example, if a ramen shop owner writes, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will execute the following process:

[1791] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[1792] 2. Analysis: The AI ​​analyzes the data and determines that the number of customers at night is increasing, but that the business hours are short, meaning the business is missing out. At the same time, the emotion engine recognizes the owner's concerns.

[1793] 3. Proposal Generation: Generates proposals such as extending business hours, improving the menu, and introducing new delivery services. Based on the sentiment engine, the proposals also include wording that will ease customers' concerns.

[1794] 4. Presentation: The terminal displays the generated suggestions to the ramen shop owner and prompts them to make specific changes.

[1795] This concludes the description of the embodiment of the present invention. This system provides an efficient data strategy that utilizes behavioral big data, generative AI, and an emotion engine, making it possible to solve customer problems without requiring specialized knowledge.

[1796] The processing flow will be explained below.

[1797] Step 1:

[1798] The server collects behavioral big data (e.g., search history, location information, purchase history) through APIs, sends API requests, and stores the acquired data in storage.

[1799] Step 2:

[1800] The server cleanses the collected data by imputing missing values, correcting outliers, and filtering out unnecessary information and noise.

[1801] Step 3:

[1802] The server preprocesses the cleansed data, converting it into features and converting it into a format that the AI ​​model can learn from. The preprocessed data is then stored in a database.

[1803] Step 4:

[1804] The server loads and configures the initial generative AI model, then trains the AI ​​model using the preprocessed data, and uses deep learning algorithms to optimize the model parameters.

[1805] Step 5:

[1806] The server integrates an emotion engine that analyzes user input (text and voice) and recognizes emotions.

[1807] Step 6:

[1808] Users input their issues and needs through a dialogue interface. For example, they might enter, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic."

[1809] Step 7:

[1810] The terminal sends the information entered by the user to the server. The input content is sent to the server in text format as an HTTP request.

[1811] Step 8:

[1812] The server receives user input sent from the device, then uses natural language processing (NLP) to analyze the user's needs and challenges. Text analysis identifies specific challenges and requests.

[1813] Step 9:

[1814] The server references relevant behavioral big data based on the analyzed needs, executes database queries, and retrieves the required data sets.

[1815] Step 10:

[1816] The server uses generative AI to generate optimal proposals based on the acquired data and analysis results, such as specific action plans for extending business hours, improving menus, and introducing new delivery services.

[1817] Step 11:

[1818] The server tailors the suggestions based on the user's emotional state as recognized by the emotion engine, for example including reassuring words for a user who is feeling anxious.

[1819] Step 12:

[1820] The server sends the generated proposal to the device in a format such as JSON, as an HTTP response.

[1821] Step 13:

[1822] The terminal displays the proposals received from the server on the user interface, presenting the proposals in an easy-to-understand format.

[1823] Step 14:

[1824] Users can review the generative AI's suggestions and take specific actions based on them, such as setting new business hours or launching a delivery service.

[1825] Example 2

[1826] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1827] In today's data-driven society, companies are required to effectively utilize behavioral big data to gain a deep understanding of their customers' needs and challenges. However, conventional systems require time-consuming cleansing and preprocessing of collected data, and require significant effort to train generative AI models and analyze their needs. Furthermore, it is difficult to analyze customer sentiment and make recommendations based on it, which makes it difficult to improve customer satisfaction and build trusting relationships with customers in real-world business situations.

[1828] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1829] In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the data, means for training a generation AI using the preprocessed data, means for analyzing user input and recognizing emotions, means for adjusting proposal content based on the recognized emotions, and means for providing generated proposals to customers. This enables efficient cleansing and preprocessing of collected data, enabling rapid model training and needs analysis. Furthermore, adjusting proposal content based on customer emotions enables more personalized proposals, improving customer satisfaction and building trust in business relationships.

[1830] "Behavioral big data" is a dataset that collects a large amount of information about user behavior, including search history, location information, purchase history, and so on.

[1831] "Data cleansing" refers to the process of removing unnecessary information and noise from collected data to improve the quality of the data.

[1832] "Preprocessing" is the process of formatting data and converting it into features before analysis or model training.

[1833] "Generative AI" is an artificial intelligence model that uses machine learning algorithms to generate new data and information from input data.

[1834] An "emotion engine" is a system that uses natural language processing technology to analyze and recognize emotions from user input data (text and voice).

[1835] "Needs analysis" is the process of identifying what customers want and what issues they face based on their statements and data.

[1836] "Adjusting proposals" is the process of optimizing proposals provided to customers based on the analysis results and recognized emotional information.

[1837] "Providing" refers to the act of presenting the generated proposal to the customer in an appropriate format.

[1838] The present invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. This system consists of three elements: a server, a terminal, and a user. Specific embodiments for implementing the present invention are described below.

[1839] Data collection and preprocessing

[1840] The server first collects behavioral big data using APIs (e.g., location information API, purchase history API). This includes user search history, location information, purchase history, etc. The collected data is temporarily stored in storage. The server then cleanses the data and removes unnecessary information and noise. Preprocessing is completed by formatting the data and converting it into the required features. The preprocessed data is then stored in a database.

[1841] Training generative AI

[1842] The server initializes and trains the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch). Using the preprocessed data, the deep learning algorithm optimizes the model's parameters. This training process enables the generative AI to efficiently analyze customer needs and challenges.

[1843] Emotion engine integration

[1844] The server also integrates an emotion engine (e.g., IBM Watson Tone Analyzer) that uses natural language processing technology to recognize emotions from user input data (text and voice). The emotion engine identifies the emotion contained in the user input and passes that emotion information to the generative AI.

[1845] Processing user inquiries

[1846] Users use a dialogue interface (e.g., a web form or chatbot) to input their issues and needs. For example, they might input, "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic." The device then sends the user's input to the server. The input is sent as an HTTP request to the server, where it is analyzed.

[1847] Sentiment Analysis and Suggestion Generation

[1848] The server analyzes the received user input using natural language processing technology and an emotion engine. This allows it to identify the user's needs, challenges, and even their emotions. It then references the necessary behavioral big data based on the analysis results and uses generative AI to generate optimal proposals. For example, it creates specific action plans, such as extending business hours or introducing new delivery services.

[1849] Tailoring suggestions based on emotions

[1850] The server adjusts the suggestions based on the user's emotional information recognized by the emotion engine. For example, if the user is feeling anxious, the suggestions will be wrapped in reassuring language and include specific, actionable actions.

[1851] Sending the proposal results

[1852] Finally, the server sends the generated proposal in JSON format to the device, which displays the proposal in a user interface and presents it to the user in an easy-to-understand manner.

[1853] Specific examples

[1854] If a ramen shop owner types in, "Sales have been declining since the COVID-19 outbreak. I'd like to know specific actions to improve sales," the system will act as follows:

[1855] 1. Data collection: The server collects behavioral big data such as the number of users at the nearest station and the usage of restaurant delivery services in the surrounding area.

[1856] 2. Analysis: Generative AI analyzes the data and determines that nighttime visitors are increasing but that short opening hours are a missed opportunity. The emotion engine also recognizes the owner's concerns.

[1857] 3. Proposal Generation: Propose extended hours, improved menu items, and new delivery services, along with language to ease concerns.

[1858] 4. Presentation: The device displays the generated proposal to the owner and prompts them to take specific action.

[1859] Example prompt sentence:

[1860] "Please suggest specific actions to improve sales for a ramen shop whose sales have declined due to the COVID-19 pandemic."

[1861] This system allows for personalized recommendations based on customer needs and emotions, contributing to business success.

[1862] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1863] Program processing flow

[1864] Step 1:

[1865] Data collection

[1866] The server collects behavioral big data using APIs (e.g., location information API, purchase history API). For example, the server obtains user search history, location information, and purchase history through API requests.

[1867] Input: Raw data obtained from API (search history, location information, purchase history)

[1868] Output: The collected raw data is stored in the server storage.

[1869] Step 2:

[1870] Data Cleansing and Preprocessing

[1871] The server cleanses the collected data by removing duplicates, filling in missing data, and removing noise, and then formats the data as features.

[1872] Input: Raw data collected in step 1

[1873] Output: The cleansed and preprocessed data is stored in a database.

[1874] Step 3:

[1875] Model initialization

[1876] The server initializes the generative AI model using a deep learning framework (e.g., TensorFlow, PyTorch), loading the initialization configuration file and setting the necessary parameters.

[1877] Input: Initialization config file for training

[1878] Output: Initialized generative AI model

[1879] Step 4:

[1880] Training and Optimization

[1881] The server uses the preprocessed data to train the generative AI, specifically optimizing the model parameters (weights, biases) using a deep learning algorithm.

[1882] Input: Preprocessed data, initialized generative AI model

[1883] Output: Optimized generative AI model after training

[1884] Step 5:

[1885] Emotion engine collaboration

[1886] The server uses an emotion engine (e.g., natural language processing technology) to analyze text and voice input from the user and recognize emotions. The analysis results are passed to the generative AI.

[1887] Input: User input data (text or voice)

[1888] Output: Emotion analysis results

[1889] Step 6:

[1890] Receiving User Input

[1891] Users input their issues and needs using a dialogue interface (e.g., web form, chatbot), and the input information is sent from the terminal to the server.

[1892] Input: User's question or need (e.g., "I want to know specific actions to increase sales at my ramen shop after the COVID-19 pandemic.")

[1893] Output: HTTP request to the server

[1894] Step 7:

[1895] Sending input data

[1896] The terminal sends the user's input to the server as an HTTP request. The input is sent in text format.

[1897] Input: User input data (text)

[1898] Output: HTTP request sent to the server

[1899] Step 8:

[1900] Needs and Sentiment Analysis

[1901] The server analyzes the received user input using natural language processing technology and an emotion engine, thereby identifying the user's needs, challenges, and emotions.

[1902] Input: HTTP request sent to the server, sentiment analysis results

[1903] Output: Analysis results of needs and emotions

[1904] Step 9:

[1905] Data lookup and proposal generation

[1906] The server then references relevant behavioral big data based on the analysis results and uses generative AI to generate optimal proposals, such as action plans to extend business hours or introduce new delivery services.

[1907] Input: Needs and emotion analysis results, behavioral big data

[1908] Output: Optimal suggestions from generative AI

[1909] Step 10:

[1910] Tailoring suggestions based on emotions

[1911] The server adjusts the content of the suggestions based on the analysis results of the emotion engine. For example, if a user is feeling anxious, the server will make the suggestions more detailed and change the wording to give a sense of security.

[1912] Input: Optimal suggestions by generative AI, emotion analysis results

[1913] Output: Adjusted proposal

[1914] Step 11:

[1915] Sending the proposal results

[1916] The server sends the generated proposal in JSON format to the device, which receives it and displays it in its user interface.

[1917] Input: Adjusted proposal

[1918] Output: HTTP response to the device

[1919] Step 12:

[1920] View Suggestions

[1921] The device displays the proposals received from the server on the user interface in a way that is easy for the user to understand and intuitively understand.

[1922] Input: HTTP response from the server

[1923] Output: The suggestions displayed to the user

[1924] Step 13:

[1925] Execute Action

[1926] The user can review the displayed suggestions made by the generated AI and take specific actions based on them, such as setting new business hours, improving the menu, or launching a new delivery service.

[1927] Input: Displayed suggestion

[1928] Output: The specific actions taken

[1929] (Application example 2)

[1930] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1931] Conventional online shopping sites have systems that recommend products by collecting and analyzing customer behavior data, but they lack a means to recognize the emotional state of the customer and make more appropriate and personalized suggestions. This means that suggestions cannot be made that fully take into account the anxiety and excitement that customers feel, making it difficult to improve customer satisfaction.

[1932] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting behavioral big data, means for cleansing and preprocessing the collected data, means for training a generation AI using the preprocessed data, means for analyzing customer needs and issues using the generation AI and generating optimal proposals, means for detecting user emotions using an emotion recognition engine, means for adjusting the content of the proposal based on the user's emotional state, and means for providing the generated proposals to the customer. This enables personalized proposals that take customer emotions into consideration, thereby increasing customer satisfaction.

[1933] (definition statement)

[1934] "Behavioral big data" refers to large amounts of data based on user behavior, including search history, location information, purchase history, and more.

[1935] "Cleansing" refers to the process of removing unnecessary information and noise from acquired data to improve the quality of the data.

[1936] "Preprocessing" refers to the process of formatting data into a format suitable for analysis, and includes the extraction of features.

[1937] "Generative AI" refers to artificial intelligence models that are trained using deep learning algorithms to generate optimal outputs from given input data.

[1938] An "emotion recognition engine" refers to technology that analyzes a user's input data (text or voice) and identifies their emotional state.

[1939] "Natural language processing (NLP)" is a technology for understanding, analyzing, and generating natural language, and refers to algorithms that interpret the meaning of sentences.

[1940] "Customer needs" refers to the products and services that users desire, or the related issues and demands.

[1941] "Suggestion tailoring" refers to optimizing generated suggestions based on the user's emotional state, which can include changing the tone and content of the suggestions.

[1942] "User's emotional state" refers to the psychological emotions (e.g., anxiety, excitement, sadness, joy) that a user is feeling at a particular moment.

[1943] "Personalization" refers to customization to meet the specific needs and preferences of individual users.

[1944] "Customer satisfaction" refers to an indicator that shows how satisfied customers are with the products and services provided.

[1945] This invention combines a system that collects behavioral big data, analyzes customer needs and issues using generative AI, and makes optimal proposals with an emotion engine that recognizes user emotions. The system is implemented in the following form.

[1946] Data collection and preprocessing

[1947] Collecting behavioral big data

[1948] The server collects big data on user behavior through the API. For example, it collects data such as the user's search history on an online shopping site, location information, and purchase history. This allows it to understand the user's behavioral patterns. Upon an API request, the data is sent to the server and stored in storage.

[1949] Data Cleansing and Preprocessing

[1950] The server cleanses the collected data, removing unnecessary information and noise, and formats the data to convert it into features. The preprocessed data is stored in a database for later analysis.

[1951] Training generative AI

[1952] Model initialization

[1953] The server loads and configures the initial generative AI model, which is then trained on the deep learning algorithms that will be used later.

[1954] Training and Optimization

[1955] The server uses the preprocessed data to train the generative AI, which optimizes the model parameters and enables it to generate optimal suggestions based on the user's behavioral data.

[1956] Emotion engine integration

[1957] emotion recognition

[1958] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The emotion recognition engine detects emotions contained in the text entered by the user and sends the results to the server.

[1959] Processing user inquiries

[1960] Receiving User Input

[1961] Users input their problems and needs through the dialogue interface of the smartphone application, for example, by entering a prompt sentence such as, "I'm looking for new running shoes, but I'm not sure which ones to get."

[1962] Sending input data

[1963] The terminal sends the user's input data to the server, which then processes the data as an HTTP request.

[1964] Needs and Sentiment Analysis

[1965] The server analyzes the received user input using natural language processing and emotion recognition engines, thereby identifying the user's emotional state as well as their challenges and needs.

[1966] Data lookup and proposal generation

[1967] The server then references relevant behavioral big data based on the analyzed needs and emotions. It retrieves the necessary data sets and uses generative AI to generate optimal recommendations. For example, if it recommends a specific product, it explains why that product is effective and presents options.

[1968] Tailoring suggestions based on emotions

[1969] The server tailors the suggestions based on the user's emotional state as recognized by the emotion recognition engine: for example, if the user expresses anxiety, the suggestion is worded in a reassuring way and includes detailed explanations.

[1970] Sending the proposal results

[1971] The server formats the generated suggestions in JSON format or similar and sends them to the device as an HTTP response, allowing appropriate suggestions to be provided to the user quickly.

[1972] Viewing and taking action on results

[1973] The terminal displays the suggestions received from the server on the user interface of the utility in an easy-to-understand manner for the user, who can then check the suggestions in detail and take specific action as necessary.

[1974] In this way, the entire system works together to make personalized suggestions that take the user's emotions into account, which is expected to improve customer satisfaction.

[1975] Examples of specific examples and prompts

[1976] For example, if a customer types in the app, "I'm looking for new running shoes, but I'm not sure which ones to get," the system will generate appropriate recommendations. Here's an example prompt:

[1977] I'm looking for new running shoes but I'm not sure which ones to get.

[1978] This invention effectively combines an emotion engine with generative AI, making it possible to make personalized suggestions that take emotions into account, something that was difficult to achieve with conventional systems.

[1979] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1980] Step 1:

[1981] The server collects user behavioral big data (search history, location information, purchase history) through API. Specifically, it sends API requests and saves the acquired data in storage. The input is the API request, and the output is the collected behavioral data.

[1982] Step 2:

[1983] The server cleanses the collected data, removing unnecessary information and noise. During this process, the data is shaped and converted into features. The input is the collected behavioral data, and the output is the cleansed, pre-processed data.

[1984] Step 3:

[1985] The server uses the preprocessed data to train the generative AI, specifically optimizing model parameters using a deep learning algorithm. The input is the preprocessed data, and the output is an optimized generative AI model.

[1986] Step 4:

[1987] The server uses a generative AI model to analyze customer needs and issues and generate optimal proposals. The input is user behavior data and the generative AI model, and the output is the generated proposal.

[1988] Step 5:

[1989] The server integrates an emotion recognition engine and analyzes input data (text and voice) from the user to recognize emotions. The input is the user's input data, and the output is the recognized emotional state.

[1990] Step 6:

[1991] The server adjusts the generated suggestions based on the user's emotional state. Specifically, it changes the tone and content of the suggestions depending on the emotions detected by the emotion recognition engine. The input is the recognized emotional state and the generated suggestions, and the output is the adjusted suggestions.

[1992] Step 7:

[1993] A user inputs a problem or need through the dialogue interface of a smartphone application, for example, by entering a prompt statement such as "I'm looking for new running shoes, but I'm not sure which ones to get." The input is the prompt statement, and the output is the user input as text.

[1994] Step 8:

[1995] The terminal sends the user's input data to the server. This input data is sent to the server as an HTTP request and processed. The input is the user's text input, and the output is an HTTP request to the server.

[1996] Step 9:

[1997] The server performs needs and emotion analysis, generates optimal proposals, and sends the tailored proposals to the device. The inputs are the user's needs, behavioral data, emotional state, and the generative AI model, and the output is the sending of the proposals to the device.

[1998] Step 10:

[1999] The terminal displays the proposals received from the server on the user interface. The proposals are presented in a way that is easy for the user to understand. The input is the proposal sent from the server, and the output is the proposal displayed on the user interface.

[2000] Step 11:

[2001] The user reviews the generative AI's suggestions and takes specific actions based on them, such as making a decision to purchase the running shoes presented. The input is the suggestions displayed in the user interface, and the output is the user's specific actions.

[2002] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2003] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2004] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2005] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2006] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2007] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2008] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2009] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2010] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2011] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2012] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2013] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2014] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2015] 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.

[2016] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2017] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2018] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2019] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2020] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2021] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2022] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2023] The following is further disclosed regarding the above embodiment.

[2024] (Claim 1)

[2025] A means of collecting behavioral big data;

[2026] a means of cleansing and pre-processing the collected data;

[2027] a means for training a generative AI using the preprocessed data; and

[2028] A means to analyze customer needs and issues using generative AI and generate optimal proposals,

[2029] a means for providing the generated proposals to the customer;

[2030] A system including:

[2031] (Claim 2)

[2032] 2. The system of claim 1, wherein the behavioral big data includes search history, location information, and purchase history.

[2033] (Claim 3)

[2034] The system of claim 1, wherein the generative AI analyzes customer needs using natural language processing.

[2035] "Example 1"

[2036] (Claim 1)

[2037] A means of collecting behavioral big data;

[2038] a means of cleansing and pre-processing the collected data;

[2039] a means for training a generative AI using the preprocessed data; and

[2040] A means to analyze customer needs and issues using generative AI and generate optimal proposals,

[2041] means for receiving customer input and analyzing said input;

[2042] a means for referencing the received customer input based on a database and a generative AI model to obtain relevant data and generate optimal recommendations;

[2043] means for displaying the generated suggestions via a user interface;

[2044] A system including:

[2045] (Claim 2)

[2046] 2. The system of claim 1, wherein the behavioral big data includes search history, location information, and purchase history.

[2047] (Claim 3)

[2048] The system of claim 1, wherein the generative AI analyzes customer needs using natural language processing.

[2049] "Application Example 1"

[2050] (Claim 1)

[2051] A means of collecting behavioral big data;

[2052] a means of cleansing and pre-processing the collected data;

[2053] a means for training a generative AI using the preprocessed data; and

[2054] A means to analyze customer needs and issues using generative AI and generate optimal proposals,

[2055] a means for providing the generated proposals to the customer via a smartphone application;

[2056] A system including:

[2057] (Claim 2)

[2058] The system of claim 1, wherein the behavioral big data includes search history, location information, purchase history, and time spent in a store.

[2059] (Claim 3)

[2060] The system of claim 1, wherein the generating AI uses natural language processing to analyze the customer's needs and present recommended products for the next time the customer visits the store.

[2061] "Example 2: Combining Emotion Engines"

[2062] (Claim 1)

[2063] A means of collecting behavioral big data;

[2064] a means of cleansing and pre-processing the collected data;

[2065] a means for training a generative AI using the preprocessed data; and

[2066] A means to analyze customer needs and issues using generative AI and generate optimal proposals,

[2067] means for analyzing user input to recognize emotions;

[2068] a means for tailoring recommendations based on the perceived sentiment;

[2069] a means for providing the generated proposals to the customer;

[2070] A system including:

[2071] (Claim 2)

[2072] 2. The system of claim 1, wherein the behavioral big data includes search history, location information, and purchase history.

[2073] (Claim 3)

[2074] The system of claim 1, wherein the generative AI uses natural language processing to analyze customer needs and integrates an emotion recognition engine to tailor suggestions based on the analysis results.

[2075] "Application example 2 when combining emotion engines"

[2076] (Claim 1)

[2077] A means of collecting behavioral big data;

[2078] a means of cleansing and pre-processing the collected data;

[2079] a means for training a generative AI using the preprocessed data; and

[2080] A means to analyze customer needs and issues using generative AI and generate optimal proposals,

[2081] means for detecting a user's emotion using an emotion recognition engine;

[2082] means for adjusting the suggestions based on the emotional state of the user;

[2083] a means for providing the generated proposals to the customer;

[2084] A system including:

[2085] (Claim 2)

[2086] 2. The system of claim 1, wherein the behavioral big data includes search history, location information, and purchase history.

[2087] (Claim 3)

[2088] The system of claim 1, wherein the generative AI analyzes customer needs using natural language processing. [Explanation of symbols]

[2089] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting behavioral big data; a means of cleansing and pre-processing the collected data; a means for training a generative AI using the preprocessed data; and A means to analyze customer needs and issues using generative AI and generate optimal proposals, a means for providing the generated proposals to the customer; A system including:

2. The system of claim 1 , wherein the behavioral big data includes search history, location information, and purchase history.

3. The system of claim 1, wherein the generation AI analyzes customer needs using natural language processing.

Images