system

Abstract

システムを提供する。【解決手段】顧客情報を収集する手段と、収集した顧客情報を前処理する手段と、顧客情報に基づき顧客を分類する手段と、顧客の契約情報を自動的に抽出する手段と、顧客の嗜好を推定する手段と、推定された嗜好に基づきプロモーション施策を生成する手段と、生成されたプロモーション施策を通知する手段と、ユーザーの購買行動履歴を分析する手段と、ユーザーの購買行動履歴に基づき商品の推薦を行う手段と、推奨された商品情報を端末に通知する手段を含むシステム。

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Modern enterprises need to handle a vast amount of customer information, but there are problems that it takes a great deal of time and effort to use this information for contract procedures and customer preference analysis. Also, in order to quickly implement effective promotion measures, a detailed analysis of customers is indispensable, and the existing methods cannot sufficiently solve these problems.

Means for Solving the Problems

[0005] This invention provides a system for efficiently processing collected customer information. This system includes means for collecting customer information, pre-processing the collected information, classifying customers, automatically extracting contract information, estimating customer preferences, and generating promotional strategies. Furthermore, by utilizing natural language processing technology and machine learning algorithms, it enables efficient and highly accurate analysis, significantly improving the operational efficiency of businesses.

[0006] "Customer information" refers to data held by a company that relates to individual customers, including purchase history, contract status, and behavioral logs.

[0007] "Preprocessing" refers to the process of converting raw data into an analyzable format, and includes operations such as data cleaning, normalization, and encoding.

[0008] "Classification" is a method of grouping data based on common characteristics, and involves aggregating customers with similar features using techniques such as clustering.

[0009] "Extracting contract information" refers to automatically extracting necessary contract-related data from documents such as contracts using natural language processing technology.

[0010] "Preference estimation" is the process of identifying the products and services that individual customers prefer, based on their past behavior and purchase history.

[0011] "Promotional measures" refer to proposed campaigns and discount plans implemented as part of sales promotion activities for customers, and are adjusted according to customer preferences.

[0012] "Natural language processing technology" is a technology that enables computers to understand and process human language, and is used when extracting information from documents or analyzing text.

[0013] A "machine learning algorithm" is a computational method used to learn patterns from data and make future predictions and classifications; it is a part of artificial intelligence technology. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14]It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Embodiments for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, a labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention relates to a system for efficiently processing vast amounts of customer information held by a company, enabling rapid contract procedures and effective promotional measures. This system consists of servers, terminals, and users, and functions by integrating various technologies.

[0036] First, the server automatically collects customer information from the company's database. The collected data is preprocessed to remove noise and convert it to the required format. Next, the server uses machine learning algorithms to classify customers into certain groups based on their purchasing patterns and behavioral history. Based on this classification, it estimates customer preferences and interests with high accuracy.

[0037] Next, the server uses natural language processing technology to automatically extract and organize contract information from contract documents and other materials. This makes it possible to quickly grasp contract renewal information and conditions. Furthermore, it generates optimal promotional strategies tailored to estimated customer preferences and notifies the terminal.

[0038] Users can review this information via their devices and approve or modify promotional strategies. This allows users to easily implement the most effective strategies for individual customers. For example, a fashion shop could use past purchase history to infer a customer's preferred brands and send timely notifications about new collections.

[0039] This format allows companies to provide personalized service to customers and significantly improve the overall efficiency of their operations.

[0040] The following describes the processing flow.

[0041] Step 1:

[0042] The server connects to the company's database and selects the customer information to retrieve. This information includes purchase history, access logs, contract history, etc. The server retrieves this data periodically or in real time.

[0043] Step 2:

[0044] The server performs preprocessing on the acquired customer information. Specifically, it checks for missing data, removes outliers, and standardizes data formats to create a clean dataset suitable for analysis.

[0045] Step 3:

[0046] The server uses machine learning algorithms to analyze customer behavior patterns from pre-processed data. This allows it to classify customers by characteristic and evaluate their behavioral tendencies and purchasing intent for each group.

[0047] Step 4:

[0048] The server uses natural language processing technology to extract contract information from the text of contracts and related documents. It organizes important contract elements such as dates, terms, and renewal status, and stores them in digital format.

[0049] Step 5:

[0050] The server analyzes customer preferences. Based on past purchase data and behavioral logs, it estimates products and services that customers might be interested in, and uses the results as reference data for promotional strategies.

[0051] Step 6:

[0052] The server generates optimal promotional strategies based on estimated customer preferences. This includes suggesting discounts and campaigns, and drafting email marketing messages.

[0053] Step 7:

[0054] The generated promotional campaign plan will be notified to the user on their device. The user can review the plan, make any necessary modifications, and then approve its implementation.

[0055] Step 8:

[0056] The user-approved measures are put into action. Promotional measures are deployed from the terminal via the server, and emails and online advertisements are sent to the target customers.

[0057] (Example 1)

[0058] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0059] Effectively utilizing the vast amount of customer data held by companies to automatically generate and notify customers of rapid contract procedures and personalized promotional strategies remains challenging. Traditional methods struggle to accurately understand customer preferences and make effective proposals at the right time, highlighting the need for improved operational efficiency.

[0060] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0061] In this invention, the server includes means for collecting content from information sources that store customer data, means for filtering the collected data and converting it into a standard format, and means for classifying customer behavior patterns using machine learning. This enables the automatic collection and organization of important information from vast amounts of customer data, and the creation and notification of personalized promotional strategies based on customer preferences.

[0062] "Customer data" refers to a collection of information about customers held by a company, including individual purchase history, contract information, and preference patterns.

[0063] "Information source" refers to a location where data is aggregated and recorded, such as a database or repository where a company stores customer data.

[0064] "Filtering" refers to the process of removing noise and unnecessary information from collected data and converting it into a format suitable for analysis.

[0065] "Converting to a standard format" refers to the operation of unifying data that exists in various formats into a consistent format, making it easier to process.

[0066] "Machine learning" is a technology in which computer systems learn patterns from data and perform predictions and classifications, and it is a means of analyzing customer data using algorithms.

[0067] "Classifying based on behavioral patterns" is the process of analyzing a customer's past behavioral history and dividing customers with similar characteristics into similar groups.

[0068] "Document analysis technology" refers to methods of extracting and structuring necessary information from documents using natural language processing.

[0069] An "advertising strategy" is a set of activities and measures planned to promote the sale of a product, and a method for making effective proposals based on customer preferences.

[0070] "Notification" refers to the act of communicating generated information or measures to users and prompting them to prepare for the implementation of those measures.

[0071] This invention is a system that provides services utilizing the vast amount of customer data held by a company, and its implementation is achieved through the cooperation of servers, terminals, and users.

[0072] First, the server connects to the database and collects customer data. This involves extracting necessary data from predefined sources, such as customer purchase history, contract information, and preference data. This collection process is more efficient when using distributed processing frameworks such as Apache Hadoop or Apache Spark.

[0073] Next, the server preprocesses the collected data. This preprocessing includes cleaning the data and converting it to a consistent format, utilizing data processing libraries in Python. Specifically, Pandas is used to format the data, impute missing values, and remove duplicate data.

[0074] Subsequently, the server uses machine learning algorithms to classify customers. Using libraries such as Scikit-learn and TENSORFLOW®, it analyzes customer purchasing behavior and preferences using K-means clustering and random forest classifiers, and classifies them into groups with similar patterns.

[0075] Based on these classification results, the server uses natural language processing techniques to extract and organize important information from the contract documents. In this step, Python's Natural Language Toolkit (NLTK) and SpaCy are used to automatically retrieve elements such as contract renewal dates and terms.

[0076] Next, the generative AI model is used to generate personalized promotional strategies on the server based on the estimated customer preferences. An example of a prompt used here is, "What is the best promotion for this customer?" By entering this prompt, the AI ​​model generates appropriate strategies.

[0077] The device is responsible for notifying users of the generated promotional campaigns. These notifications are delivered through applications using JavaScript® or React Native, allowing users to review them and modify the campaigns as needed. For example, they can adjust discount rates or campaign durations.

[0078] Such systems enable companies to provide more effective and efficient services to their customers and optimize their business operations.

[0079] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0080] Step 1:

[0081] The server connects to the company's database and collects customer data. The input here is raw customer information obtained via database queries. Based on this input, the server uses Apache Hadoop to efficiently retrieve the data and temporarily stores the retrieved data in local storage.

[0082] Step 2:

[0083] The server preprocesses the collected customer data. The input for this step is the raw data obtained in step 1. Pandas is used to remove noise from the data, impute missing values, and perform standardization. The output is a clean and consistent dataset.

[0084] Step 3:

[0085] The server uses a machine learning algorithm to classify customers using the cleaned dataset. The input is the data preprocessed in step 2. Using Scikit-learn, K-means clustering is applied to classify customers into different groups. The output of this operation is a list of clustered customer groups.

[0086] Step 4:

[0087] The server uses Python's SpaCy to perform natural language processing and extract necessary contract information from customer contract documents. The input is contract documents held by companies. Through analysis, elements such as update dates and contract terms are extracted and organized. The output is a structured dataset with important information tagged.

[0088] Step 5:

[0089] The server generates promotional strategies using a generative AI model. The input for this step is the clustering results from step 3 and the contract information obtained in step 4. The prompt "What is the best promotion for this customer?" is passed to the AI ​​model to generate individual strategy suggestions. The output is a customized list of promotional strategies for each customer.

[0090] Step 6:

[0091] The device notifies the user of the promotion generated on the server. The input is the promotional initiative generated in step 5. These initiatives are sent to the user using push notifications in a mobile app using JavaScript or React Native. The output is the notification message sent to the user.

[0092] Step 7:

[0093] Users check notifications via their devices and modify promotional strategies as needed. User input involves detailed editing and approval of the strategies. Through this editing process, the output is an optimized promotional strategy with approvals and necessary changes completed.

[0094] (Application Example 1)

[0095] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0096] In today's business environment, companies are required to efficiently manage vast amounts of customer data and deliver services quickly. However, effectively utilizing this data and conducting promotions tailored to each customer is not easy. Furthermore, e-commerce sites need to accurately provide personalized product recommendations based on customers' purchasing behavior. This invention aims to solve these problems.

[0097] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0098] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for notifying the generated promotional measures, means for analyzing the user's purchase history, means for recommending products based on the user's purchase history, and means for notifying the terminal of the recommended product information. This enables companies to process customer information quickly and efficiently implement personalized promotions and product recommendations.

[0099] "Means of collecting customer information" refers to the processes or technologies used by a company to collect data about its customers.

[0100] "Preprocessing methods" refer to techniques that perform noise reduction and format conversion in order to prepare collected data into an analyzable format.

[0101] "Methods for classifying customers" refer to techniques for dividing customers into various groups based on their attributes and behavior.

[0102] "Methods for automatically extracting contract information" refers to technologies that mechanically extract necessary information from documents such as contracts.

[0103] "Methods for estimating preferences" refer to technologies that analyze a customer's past behavior and purchase history to predict their individual preferences and interests.

[0104] "Means of generating promotional strategies" refers to techniques for creating marketing and sales promotion strategies based on estimated customer preferences.

[0105] "Means of notifying about promotional measures" refers to technologies for delivering generated promotional information to end users' devices.

[0106] "Methods for analyzing purchasing behavior history" refer to technologies used to analyze a customer's purchase history and understand their purchasing patterns and trends.

[0107] "Methods for recommending products" refer to techniques for recommending specific products to customers based on collected information.

[0108] "Means of notifying a device of recommended product information" refers to technologies that display information about recommended products on the user's device.

[0109] In the system that implements this application, the server first collects customer information from the company's database and automatically preprocesses it. Preprocessing includes denoising the data and converting it to a standard format. The server uses machine learning algorithms to classify customers into several groups and estimate the preferences of each group. Next, it leverages natural language processing techniques to extract contract information and organize the contract terms and deadlines relevant to each customer.

[0110] Subsequently, the server generates optimal promotional strategies tailored to the estimated customer preferences and notifies the user's device. Through this device, the user can review the provided promotional information and approve or modify it as needed. Furthermore, based on the user's purchase history, the server recommends products and notifies the user's device accordingly.

[0111] The system is developed using Python and TensorFlow, with Pandas used for data analysis. The user interface is built with Flutter®, and promotional and product information is notified in real time to devices such as smartphones. For example, if a customer purchased sports shoes the previous month, new running wear can be recommended based on their preferences.

[0112] Such systems enable quick and accurate responses tailored to user needs, and are expected to boost sales for businesses. Recommendation systems utilizing generative AI models provide personalized suggestions to individual customers using prompts such as, "Based on the items most frequently purchased in the past six months, suggest the most suitable products for this user."

[0113] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0114] Step 1:

[0115] The server automatically collects customer information from the company's database. It executes database queries to retrieve the necessary customer information and saves it to local storage. The input is raw data from the database, and the output is raw data containing noise.

[0116] Step 2:

[0117] The server preprocesses the customer information collected in the previous step. Denoising and format conversion are performed using the Python Pandas library. The converted data is then prepared in a format that can be analyzed. The input is the raw data output from step 1, and the output is the processed, clean dataset.

[0118] Step 3:

[0119] The server runs a machine learning algorithm using the processed data to classify customers into specific groups. TensorFlow is used for classification, and a clustering method is applied. The input is the clean data obtained in step 2, and the output is a list of customer groups.

[0120] Step 4:

[0121] The server extracts contract information using natural language processing technology. It automatically extracts and organizes necessary information from text data such as contracts. The input is contract-related text information from customers, and the output is structured contract information.

[0122] Step 5:

[0123] The server estimates customer preferences and generates promotional strategies based on that information. It uses a generative AI model for estimation, analyzing customer purchasing patterns using techniques such as random forests. Inputs are historical purchase data and lifestyle information, and the output is the optimal promotional strategy.

[0124] Step 6:

[0125] The server notifies the terminal of the generated promotional campaign. Real-time messaging is used for the notification, and the promotional information is displayed on the terminal immediately. The input is the promotional campaign created in step 5, and the output is the message displayed on the user's device.

[0126] Step 7:

[0127] Users review and approve or modify promotional information provided through their devices. The interface is built with Flutter, providing a user-friendly interface. Input is information displayed on the device, and output is user feedback on the promotional campaign.

[0128] Step 8:

[0129] The server analyzes the user's purchase history and recommends products. Collaborative filtering and deep learning models are used for recommendations to generate information that reflects the user's unique tendencies. The input is past purchase history, and the output is a list of recommended products.

[0130] Step 9:

[0131] The server notifies the terminal of recommended product information. Similarly, real-time notification technology is used to immediately prompt the user to purchase the product. The input is the product recommendation list obtained in step 8, and the output is the product information displayed on the user's terminal.

[0132] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0133] This invention is a system for efficiently processing customer information and automatically generating and notifying users of promotional strategies that take into account their emotional state. This system consists of a server, terminals, and users, and works in conjunction to collect and analyze customer information, recognize emotions, and generate promotional strategies.

[0134] First, the server connects to the company's database and collects customer purchase history, access logs, and various contract information, then performs preprocessing. This preprocessing ensures data consistency while making it ready for analysis. Next, machine learning algorithms are used to classify customers and estimate their preferences. This includes analysis based on customer characteristics derived from past purchase data and behavioral patterns.

[0135] Furthermore, an emotion engine is used to estimate the user's emotional state in real time based on their interactions. The emotion engine comprehensively analyzes data such as text, voice, and facial expressions to recognize emotions like joy, surprise, and dissatisfaction. Based on this, the server formulates promotional strategies that are appropriate for the current emotional state.

[0136] The device notifies the user of promotional initiatives sent from the server. The user reviews these initiatives and adjusts the implementation plan as needed. For example, in the travel industry, offering special travel plans or campaigns when a user is showing positive emotions can increase their purchasing intent.

[0137] This format allows companies to implement flexible and personalized marketing that responds to customer emotions, resulting in improved customer satisfaction and more efficient marketing activities.

[0138] The following describes the processing flow.

[0139] Step 1:

[0140] The server accesses the company's database and automatically collects customer purchase history, access logs, contract information, etc. During this process, the timeliness and completeness of the data are ensured.

[0141] Step 2:

[0142] The server preprocesses the collected information. Specifically, this involves cleaning the data, supplementing missing data, and correcting outliers. This process prepares the dataset for analysis.

[0143] Step 3:

[0144] The server applies machine learning algorithms to analyze customer behavior patterns from pre-processed data. This groups customers based on their purchasing tendencies and interests. Based on these analysis results, it estimates customer preferences.

[0145] Step 4:

[0146] The server uses natural language processing technology to extract necessary contract information from contracts and communication history. This includes information such as contract terms and renewal dates, and is stored in an organized format in the database.

[0147] Step 5:

[0148] Through interaction with the user, the emotion engine recognizes the user's emotions in real time. The emotion engine estimates the user's emotional state by analyzing facial expressions via text input, voice commands, and, in some cases, a camera.

[0149] Step 6:

[0150] The server considers the perceived emotions and generates optimal promotional strategies based on estimated customer preferences. This includes suggesting special offers and recommended products tailored to the emotional state.

[0151] Step 7:

[0152] The device notifies the user of promotional campaigns sent from the server. The user can view the details of the campaign on the device screen and, if necessary, customize the content of the campaign.

[0153] Step 8:

[0154] The device executes the promotional measures that the user has ultimately approved. Specifically, this includes sending emails to target customers and launching online campaigns. This ensures that the company's marketing strategy is effectively implemented.

[0155] (Example 2)

[0156] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0157] In today's information-saturated world, companies are required to effectively utilize large amounts of customer data and quickly develop advertising strategies that take into account the emotional state of individual customers. However, traditional methods have made it difficult to accurately grasp customer preferences and emotional states in real time and generate appropriate advertising proposals.

[0158] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0159] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for analyzing customer dialogue data with an emotion recognition device, means for creating advertising proposals using generative AI technology based on predicted preferences and emotional states, and means for notifying customers of the created advertising proposals via a terminal. This enables the automation of flexible and personalized advertising strategies that respond to customer emotions and preferences.

[0160] "Customer information" refers to data including purchase history, access history, and contract details related to a specific individual or legal entity.

[0161] "Means of aggregation" refers to methods or devices for efficiently collecting data and storing it in one place.

[0162] "Preprocessing" refers to the process of converting collected data into an analyzable format and performing cleansing and formatting adjustments as needed.

[0163] "Methods for classifying customers" refer to methods of dividing customers into groups based on specific attributes or behavioral patterns through data analysis.

[0164] "Preference" is a concept that refers to a customer's taste or priority regarding a particular product or service.

[0165] An "emotion recognition device" is a technology or device for detecting and analyzing an individual's emotional state from text, audio, or video.

[0166] "Generative AI technology" is a technique that uses artificial intelligence to gain insights from data and automatically generate new information and suggestions.

[0167] An "advertising proposal" is a strategy or plan created to effectively propose a specific product or service to customers.

[0168] "Means of notification" refers to a method or technique for transmitting specific information to a designated recipient.

[0169] This system is comprised of three main components: a server, terminals, and users. The server connects to the company's database to collect and pre-process customer information. Specifically, it collects customer purchase history, access history, and contract information, and prepares them for analysis through data cleaning and formatting. This ensures data consistency while making it usable as structured data.

[0170] Next, the server uses machine learning algorithms to classify customers and predict their preferences. For example, it might classify customers into clusters such as travel enthusiasts or technology-oriented individuals. Python libraries such as scikit-learn and TensorFlow can be used for this purpose.

[0171] Subsequently, the server utilizes an emotion recognition device to analyze the user's dialogue data and estimate their emotional state in real time. Text, audio, and image data are analyzed using natural language processing and computer vision technologies to determine the user's emotions.

[0172] The server uses generative AI technology to create advertising proposals based on the customer's preferences and emotional state. For example, the generative AI model utilizes a language model such as GPT-3(registered trademark), and by inputting prompts such as "Create a promotion that suits customers who love to travel and are excited," it generates effective advertising proposals.

[0173] The generated ad proposals are notified to the user via their device. The device uses data reception and notification systems to deliver information to the user in real time. The user can review the received ad proposals and take action according to their content.

[0174] This format allows companies to automatically deploy personalized marketing strategies that respond to customer emotions and preferences, thereby increasing customer satisfaction.

[0175] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0176] Step 1:

[0177] The server collects customer information from the company's database. Inputs include customer purchase history, access logs, and contract information, while output is a dataset that integrates this information in a consistent format. Specifically, it executes database queries and prepares the obtained data through an ETL (Extract, Transform, Load) process.

[0178] Step 2:

[0179] The server preprocesses the aggregated data. The input is a unified dataset, and it performs data cleaning, duplicate removal, and missing value imputation to output a clean, analyzable dataset. Specifically, it uses the Python Pandas library for data formatting.

[0180] Step 3:

[0181] The server classifies customers using pre-processed data. The input is a clean dataset, and machine learning algorithms are applied to classify customers into specific clusters. The output is a classification label for each customer. Specifically, K-means clustering is used to analyze customer behavior patterns and generate cluster labels.

[0182] Step 4:

[0183] The server analyzes user dialogue data using an emotion recognition device. Inputs include text, audio, and image data, and emotion analysis technology outputs emotional states such as "joy" and "dissatisfaction." Specific operations include text analysis using natural language processing technology and tone analysis using speech analysis algorithms.

[0184] Step 5:

[0185] The server generates ad proposals using generative AI technology based on predicted preferences and emotional states. The input consists of customer cluster labels and emotional states, and the prompt text is input to the generative AI model to generate ad proposals as output. Specifically, it automatically generates appropriate campaign messages using technologies such as OpenAI's GPT-3.

[0186] Step 6:

[0187] The device notifies the user of the generated ad draft. The input is the generated ad draft, and the output is the notification received by the user. Specifically, it provides information to the user in real time using push notifications and email distribution systems.

[0188] (Application Example 2)

[0189] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0190] Traditional promotion strategy generation systems rely solely on customer preferences and do not consider their emotional state. This has resulted in the inability to provide appropriate promotions that respond to customers' real-time emotions. Consequently, it is difficult to achieve improved customer satisfaction and efficient marketing activities.

[0191] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0192] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for estimating the user's emotional state in real time, means for adjusting and generating promotional measures according to the estimated emotional state, and means for notifying the generated promotional measures. This makes it possible to provide flexible and individualized promotional measures that respond to the customer's emotional state.

[0193] "Customer information" is a general term for all data related to a customer, including their purchase history, access logs, and contract information.

[0194] "Preprocessing" refers to the process of processing and organizing collected data to make it analyzable while maintaining its consistency.

[0195] "Customer classification" is the process of grouping customers based on the characteristics of the data, using the collected customer information.

[0196] "Contract information" refers to detailed information about the contract that a customer has entered into with a company, and includes contract terms, duration, and conditions.

[0197] "Preference estimation" is the process of predicting a customer's preferences and patterns based on their past behavior and data, using machine learning algorithms and other methods.

[0198] "Promotional measures" refer to advertising activities and campaign strategies implemented by companies to increase customer purchasing intent.

[0199] "Emotional state" refers to a customer's current emotions and is classified into various emotional categories such as joy, surprise, and dissatisfaction.

[0200] "Real-time estimation" is a process that immediately analyzes the situation and derives results as soon as a user's actions or interactions occur.

[0201] "Notification" is the act of sending information or messages from a system to a user, thereby communicating information to the user in real time.

[0202] This invention constructs a system that automatically generates optimal promotional strategies by analyzing customer information and estimating customer emotions in real time. The embodiments of the system are described in detail below.

[0203] The main components of the system are a server, terminals, and users. The server connects to the company's database and collects customer information such as purchase history, access logs, and contract information. The collected data is preprocessed and modified to maintain data consistency. This processing uses libraries such as Python's pandas library.

[0204] Next, the server uses machine learning algorithms to classify customers based on the collected data and estimate their preferences. For this purpose, machine learning libraries such as TensorFlow and PyTorch are utilized. The estimated preference information is then used to generate promotional strategies using generative AI models.

[0205] Next, the server uses an emotion engine to estimate the user's emotional state in real time. This emotion engine analyzes various data from the user, such as text, voice, and images, to identify emotions. Natural language processing (NLP) techniques are applied to the analysis to estimate the emotions contained in the text.

[0206] The device receives notifications of promotional campaigns sent from the server. Designed as a smartphone app, users can view tailored promotions in real time through the app. React Native is used for app development.

[0207] A specific use case would be when a user is online shopping and writes a review for a product they are considering purchasing. Positive emotions are detected from the review, and relevant discount coupons are immediately suggested. This can further increase the user's desire to buy.

[0208] An example of an input prompt statement for a generative AI model is written as follows:

[0209] Username: Taro Tanaka, Recent purchases: Books, Current cart: Books, Review: 'This book is very interesting!'

[0210] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0211] Step 1:

[0212] The server retrieves customer information from the company's database. This information includes purchase history, access logs, and contract information. The retrieved data is preprocessed using a data processing library such as Pandasm to prepare it for easy analysis while maintaining data consistency. The input is raw data, and the output is preprocessed data.

[0213] Step 2:

[0214] The server classifies customers using machine learning algorithms based on pre-processed customer information. This process uses TensorFlow and PyTorch to analyze features such as past purchasing behavior and contract information to estimate customer preferences. The input is pre-processed customer information, and the output is customer preference clusters.

[0215] Step 3:

[0216] The server uses a generative AI model to generate promotional strategies from estimated preferences. In this process, it generates promotional strategies corresponding to specific preferences, which are then used as sales and marketing measures. The input is customer preference clusters, and the output is promotional strategies.

[0217] Step 4:

[0218] The server uses an emotion engine to analyze the user's emotional state in real time. This engine analyzes text, voice, and image data obtained from the user to determine their emotions. It utilizes NLP (Neuro-Linguistic Programming) technology to estimate emotions from text. The input is user interaction data, and the output is the detected emotional state.

[0219] Step 5:

[0220] The server adjusts and generates promotional strategies in real time based on the detected emotional state. In this process, promotions and messages that match the emotion are generated by an AI model. The input is the emotional state and promotional strategy template, and the output is the adjusted promotional strategy.

[0221] Step 6:

[0222] The device notifies the user of the tailored promotional measures received from the server. It provides real-time push notifications, allowing the user to access the appropriate promotional measures through the application. The input is the tailored promotional measures, and the output is the notification to the user.

[0223] Step 7:

[0224] Users can review promotional offers received through their devices and make purchasing decisions based on their content. User behavior data is then fed back into the system for use in future analyses. The input is the promotional offer, and the output is user behavior data.

[0225] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0226] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0227] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0228] [Second Embodiment]

[0229] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0230] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0231] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0232] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0233] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0234] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0235] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0236] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0237] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0238] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0239] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0240] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0241] This invention relates to a system for efficiently processing vast amounts of customer information held by a company, enabling rapid contract procedures and effective promotional measures. This system consists of servers, terminals, and users, and functions by integrating various technologies.

[0242] First, the server automatically collects customer information from the company's database. The collected data is preprocessed to remove noise and convert it to the required format. Next, the server uses machine learning algorithms to classify customers into certain groups based on their purchasing patterns and behavioral history. Based on this classification, it estimates customer preferences and interests with high accuracy.

[0243] Next, the server uses natural language processing technology to automatically extract and organize contract information from contract documents and other materials. This makes it possible to quickly grasp contract renewal information and conditions. Furthermore, it generates optimal promotional strategies tailored to estimated customer preferences and notifies the terminal.

[0244] Users can review this information via their devices and approve or modify promotional strategies. This allows users to easily implement the most effective strategies for individual customers. For example, a fashion shop could use past purchase history to infer a customer's preferred brands and send timely notifications about new collections.

[0245] This format allows companies to provide personalized service to customers and significantly improve the overall efficiency of their operations.

[0246] The following describes the processing flow.

[0247] Step 1:

[0248] The server connects to the company's database and selects the customer information to retrieve. This information includes purchase history, access logs, contract history, etc. The server retrieves this data periodically or in real time.

[0249] Step 2:

[0250] The server performs preprocessing on the acquired customer information. Specifically, it checks for missing data, removes outliers, and standardizes data formats to create a clean dataset suitable for analysis.

[0251] Step 3:

[0252] The server uses machine learning algorithms to analyze customer behavior patterns from pre-processed data. This allows it to classify customers by characteristic and evaluate their behavioral tendencies and purchasing intent for each group.

[0253] Step 4:

[0254] The server uses natural language processing technology to extract contract information from the text of contracts and related documents. It organizes important contract elements such as dates, terms, and renewal status, and stores them in digital format.

[0255] Step 5:

[0256] The server analyzes customer preferences. Based on past purchase data and behavioral logs, it estimates products and services that customers might be interested in, and uses the results as reference data for promotional strategies.

[0257] Step 6:

[0258] The server generates optimal promotional strategies based on estimated customer preferences. This includes suggesting discounts and campaigns, and drafting email marketing messages.

[0259] Step 7:

[0260] The generated promotional campaign plan will be notified to the user on their device. The user can review the plan, make any necessary modifications, and then approve its implementation.

[0261] Step 8:

[0262] The user-approved measures are put into action. Promotional measures are deployed from the terminal via the server, and emails and online advertisements are sent to the target customers.

[0263] (Example 1)

[0264] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0265] Effectively utilizing the vast amount of customer data held by companies to automatically generate and notify customers of rapid contract procedures and personalized promotional strategies remains challenging. Traditional methods struggle to accurately understand customer preferences and make effective proposals at the right time, highlighting the need for improved operational efficiency.

[0266] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0267] In this invention, the server includes means for collecting content from information sources that store customer data, means for filtering the collected data and converting it into a standard format, and means for classifying customer behavior patterns using machine learning. This enables the automatic collection and organization of important information from vast amounts of customer data, and the creation and notification of personalized promotional strategies based on customer preferences.

[0268] "Customer data" refers to a collection of information about customers held by a company, including individual purchase history, contract information, and preference patterns.

[0269] "Information source" refers to a location where data is aggregated and recorded, such as a database or repository where a company stores customer data.

[0270] "Filtering" refers to the process of removing noise and unnecessary information from collected data and converting it into a format suitable for analysis.

[0271] "Converting to a standard format" refers to the operation of unifying data that exists in various formats into a consistent format, making it easier to process.

[0272] "Machine learning" is a technology in which computer systems learn patterns from data and perform predictions and classifications, and it is a means of analyzing customer data using algorithms.

[0273] "Classifying based on behavioral patterns" is the process of analyzing a customer's past behavioral history and dividing customers with similar characteristics into similar groups.

[0274] "Document analysis technology" refers to methods of extracting and structuring necessary information from documents using natural language processing.

[0275] An "advertising strategy" is a set of activities and measures planned to promote the sale of a product, and a method for making effective proposals based on customer preferences.

[0276] "Notification" refers to the act of communicating generated information or measures to users and prompting them to prepare for the implementation of those measures.

[0277] This invention is a system that provides services utilizing the vast amount of customer data held by a company, and its implementation is achieved through the cooperation of servers, terminals, and users.

[0278] First, the server connects to the database and collects customer data. This involves extracting necessary data from predefined sources, such as customer purchase history, contract information, and preference data. This collection process is more efficient when using distributed processing frameworks such as Apache Hadoop or Apache Spark.

[0279] Next, the server preprocesses the collected data. This preprocessing includes cleaning the data and converting it to a consistent format, utilizing data processing libraries in Python. Specifically, Pandas is used to format the data, impute missing values, and remove duplicate data.

[0280] After that, the server classifies customers using machine learning algorithms. By using libraries such as Scikit-learn or TensorFlow, it analyzes customers' purchase behaviors and preferences using K-means clustering or random forest classifiers, and classifies them into groups with similar patterns.

[0281] Based on this classification result, the server uses natural language processing technology to extract and organize important information from the contract documents. In this step, libraries such as Python's Natural Language Toolkit (NLTK) or SpaCy are used to automatically obtain elements such as the contract update date and conditions.

[0282] Subsequently, the server generates personalized promotion measures based on the estimated customer preferences using a generative AI model. An example of the prompt text used here is "What is the optimal promotion for this customer?" By inputting this prompt, the AI model generates appropriate measures.

[0283] The terminal is responsible for notifying the user of the generated promotion measures. The notification is carried out through applications using JavaScript or React Native, and the user can check it and modify the measures if necessary. For example, it is possible to adjust the discount rate or campaign period.

[0284] Such a system enables companies to provide more effective and efficient services to customers and achieve optimization of business operations.

[0285] The flow of the specific process in Example 1 will be described using Figure 11.

[0286] Step 1:

[0287] The server connects to the company's database and collects customer data. The input here is raw customer information obtained via database queries. Based on this input, the server uses Apache Hadoop to efficiently retrieve the data and temporarily stores the retrieved data in local storage.

[0288] Step 2:

[0289] The server preprocesses the collected customer data. The input for this step is the raw data obtained in step 1. Pandas is used to remove noise from the data, impute missing values, and perform standardization. The output is a clean and consistent dataset.

[0290] Step 3:

[0291] The server uses a machine learning algorithm to classify customers using the cleaned dataset. The input is the data preprocessed in step 2. Using Scikit-learn, K-means clustering is applied to classify customers into different groups. The output of this operation is a list of clustered customer groups.

[0292] Step 4:

[0293] The server uses Python's SpaCy to perform natural language processing and extract necessary contract information from customer contract documents. The input is contract documents held by companies. Through analysis, elements such as update dates and contract terms are extracted and organized. The output is a structured dataset with important information tagged.

[0294] Step 5:

[0295] The server generates promotional strategies using a generative AI model. The input for this step is the clustering results from step 3 and the contract information obtained in step 4. The prompt "What is the best promotion for this customer?" is passed to the AI ​​model to generate individual strategy suggestions. The output is a customized list of promotional strategies for each customer.

[0296] Step 6:

[0297] The device notifies the user of the promotion generated on the server. The input is the promotional initiative generated in step 5. These initiatives are sent to the user using push notifications in a mobile app using JavaScript or React Native. The output is the notification message sent to the user.

[0298] Step 7:

[0299] Users check notifications via their devices and modify promotional strategies as needed. User input involves detailed editing and approval of the strategies. Through this editing process, the output is an optimized promotional strategy with approvals and necessary changes completed.

[0300] (Application Example 1)

[0301] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0302] In today's business environment, companies are required to efficiently manage vast amounts of customer data and deliver services quickly. However, effectively utilizing this data and conducting promotions tailored to each customer is not easy. Furthermore, e-commerce sites need to accurately provide personalized product recommendations based on customers' purchasing behavior. This invention aims to solve these problems.

[0303] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 1 is realized by the following respective means.

[0304] In this invention, the server includes means for collecting customer information, means for preprocessing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting the contract information of customers, means for estimating customer preferences, means for generating a promotion measure based on the estimated preferences, means for notifying the generated promotion measure, means for analyzing the purchase behavior history of users, means for recommending products based on the purchase behavior history of users, and means for notifying the recommended product information to the terminal. Thereby, the enterprise can quickly process customer information and efficiently implement individualized promotions and product recommendations.

[0305] The "means for collecting customer information" is a process or technology used to collect data related to customers held by an enterprise.

[0306] The "means for preprocessing" is a technology for performing noise removal and format conversion to arrange the collected data in an analyzable format.

[0307] The "means for classifying customers" is a technology for dividing customers into various groups based on customer attributes and behaviors.

[0308] The "means for automatically extracting contract information" is a technology for mechanically extracting necessary information from contracts and the like.

[0309] The "means for estimating preferences" is a technology for analyzing the past behaviors and purchase histories of customers to predict personal preferences and interests.

[0310] The "means for generating a promotion measure" is a technology for creating strategies for marketing and sales promotion based on the estimated preferences of customers.

[0311] "Means of notifying about promotional measures" refers to technologies for delivering generated promotional information to end users' devices.

[0312] "Methods for analyzing purchasing behavior history" refer to technologies used to analyze a customer's purchase history and understand their purchasing patterns and trends.

[0313] "Methods for recommending products" refer to techniques for recommending specific products to customers based on collected information.

[0314] "Means of notifying a device of recommended product information" refers to technologies that display information about recommended products on the user's device.

[0315] In the system that implements this application, the server first collects customer information from the company's database and automatically preprocesses it. Preprocessing includes denoising the data and converting it to a standard format. The server uses machine learning algorithms to classify customers into several groups and estimate the preferences of each group. Next, it leverages natural language processing techniques to extract contract information and organize the contract terms and deadlines relevant to each customer.

[0316] Subsequently, the server generates optimal promotional strategies tailored to the estimated customer preferences and notifies the user's device. Through this device, the user can review the provided promotional information and approve or modify it as needed. Furthermore, based on the user's purchase history, the server recommends products and notifies the user's device accordingly.

[0317] The system is developed using Python and TensorFlow, with Pandas used for data analysis. The user interface is built with Flutter, and promotional and product information is notified in real time to devices such as smartphones. For example, if a customer purchased sports shoes the previous month, new running wear can be recommended based on their preferences.

[0318] Such systems enable quick and accurate responses tailored to user needs, and are expected to boost sales for businesses. Recommendation systems utilizing generative AI models provide personalized suggestions to individual customers using prompts such as, "Based on the items most frequently purchased in the past six months, suggest the most suitable products for this user."

[0319] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0320] Step 1:

[0321] The server automatically collects customer information from the company's database. It executes database queries to retrieve the necessary customer information and saves it to local storage. The input is raw data from the database, and the output is raw data containing noise.

[0322] Step 2:

[0323] The server preprocesses the customer information collected in the previous step. Denoising and format conversion are performed using the Python Pandas library. The converted data is then prepared in a format that can be analyzed. The input is the raw data output from step 1, and the output is the processed, clean dataset.

[0324] Step 3:

[0325] The server runs a machine learning algorithm using the processed data to classify customers into specific groups. TensorFlow is used for classification, and a clustering method is applied. The input is the clean data obtained in step 2, and the output is a list of customer groups.

[0326] Step 4:

[0327] The server extracts contract information using natural language processing technology. It automatically extracts and organizes necessary information from text data such as contracts. The input is contract-related text information from customers, and the output is structured contract information.

[0328] Step 5:

[0329] The server estimates customer preferences and generates promotional strategies based on that information. It uses a generative AI model for estimation, analyzing customer purchasing patterns using techniques such as random forests. Inputs are historical purchase data and lifestyle information, and the output is the optimal promotional strategy.

[0330] Step 6:

[0331] The server notifies the terminal of the generated promotional campaign. Real-time messaging is used for the notification, and the promotional information is displayed on the terminal immediately. The input is the promotional campaign created in step 5, and the output is the message displayed on the user's device.

[0332] Step 7:

[0333] Users review and approve or modify promotional information provided through their devices. The interface is built with Flutter, providing a user-friendly interface. Input is information displayed on the device, and output is user feedback on the promotional campaign.

[0334] Step 8:

[0335] The server analyzes the user's purchase history and recommends products. Collaborative filtering and deep learning models are used for recommendations to generate information that reflects the user's unique tendencies. The input is past purchase history, and the output is a list of recommended products.

[0336] Step 9:

[0337] The server notifies the terminal of recommended product information. Similarly, real-time notification technology is used to immediately prompt the user to purchase the product. The input is the product recommendation list obtained in step 8, and the output is the product information displayed on the user's terminal.

[0338] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0339] This invention is a system for efficiently processing customer information and automatically generating and notifying users of promotional strategies that take into account their emotional state. This system consists of a server, terminals, and users, and works in conjunction to collect and analyze customer information, recognize emotions, and generate promotional strategies.

[0340] First, the server connects to the company's database and collects customer purchase history, access logs, and various contract information, then performs preprocessing. This preprocessing ensures data consistency while making it ready for analysis. Next, machine learning algorithms are used to classify customers and estimate their preferences. This includes analysis based on customer characteristics derived from past purchase data and behavioral patterns.

[0341] Furthermore, an emotion engine is used to estimate the user's emotional state in real time based on their interactions. The emotion engine comprehensively analyzes data such as text, voice, and facial expressions to recognize emotions like joy, surprise, and dissatisfaction. Based on this, the server formulates promotional strategies that are appropriate for the current emotional state.

[0342] The device notifies the user of promotional initiatives sent from the server. The user reviews these initiatives and adjusts the implementation plan as needed. For example, in the travel industry, offering special travel plans or campaigns when a user is showing positive emotions can increase their purchasing intent.

[0343] This format allows companies to implement flexible and personalized marketing that responds to customer emotions, resulting in improved customer satisfaction and more efficient marketing activities.

[0344] The following describes the processing flow.

[0345] Step 1:

[0346] The server accesses the company's database and automatically collects customer purchase history, access logs, contract information, etc. During this process, the timeliness and completeness of the data are ensured.

[0347] Step 2:

[0348] The server preprocesses the collected information. Specifically, this involves cleaning the data, supplementing missing data, and correcting outliers. This process prepares the dataset for analysis.

[0349] Step 3:

[0350] The server applies machine learning algorithms to analyze customer behavior patterns from pre-processed data. This groups customers based on their purchasing tendencies and interests. Based on these analysis results, it estimates customer preferences.

[0351] Step 4:

[0352] The server uses natural language processing technology to extract necessary contract information from contracts and communication history. This includes information such as contract terms and renewal dates, and is stored in an organized format in the database.

[0353] Step 5:

[0354] Through interaction with the user, the emotion engine recognizes the user's emotions in real time. The emotion engine estimates the user's emotional state by analyzing facial expressions via text input, voice commands, and, in some cases, a camera.

[0355] Step 6:

[0356] The server considers the perceived emotions and generates optimal promotional strategies based on estimated customer preferences. This includes suggesting special offers and recommended products tailored to the emotional state.

[0357] Step 7:

[0358] The device notifies the user of promotional campaigns sent from the server. The user can view the details of the campaign on the device screen and, if necessary, customize the content of the campaign.

[0359] Step 8:

[0360] The device executes the promotional measures that the user has ultimately approved. Specifically, this includes sending emails to target customers and launching online campaigns. This ensures that the company's marketing strategy is effectively implemented.

[0361] (Example 2)

[0362] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0363] In today's information-saturated world, companies are required to effectively utilize large amounts of customer data and quickly develop advertising strategies that take into account the emotional state of individual customers. However, traditional methods have made it difficult to accurately grasp customer preferences and emotional states in real time and generate appropriate advertising proposals.

[0364] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0365] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for analyzing customer dialogue data with an emotion recognition device, means for creating advertising proposals using generative AI technology based on predicted preferences and emotional states, and means for notifying customers of the created advertising proposals via a terminal. This enables the automation of flexible and personalized advertising strategies that respond to customer emotions and preferences.

[0366] "Customer information" refers to data including purchase history, access history, and contract details related to a specific individual or legal entity.

[0367] "Means of aggregation" refers to methods or devices for efficiently collecting data and storing it in one place.

[0368] "Preprocessing" refers to the process of converting collected data into an analyzable format and performing cleansing and formatting adjustments as needed.

[0369] "Methods for classifying customers" refer to methods of dividing customers into groups based on specific attributes or behavioral patterns through data analysis.

[0370] "Preference" is a concept that refers to a customer's taste or priority regarding a particular product or service.

[0371] An "emotion recognition device" is a technology or device for detecting and analyzing an individual's emotional state from text, audio, or video.

[0372] "Generative AI technology" is a technique that uses artificial intelligence to gain insights from data and automatically generate new information and suggestions.

[0373] An "advertising proposal" is a strategy or plan created to effectively propose a specific product or service to customers.

[0374] "Means of notification" refers to a method or technique for transmitting specific information to a designated recipient.

[0375] This system is comprised of three main components: a server, terminals, and users. The server connects to the company's database to collect and pre-process customer information. Specifically, it collects customer purchase history, access history, and contract information, and prepares them for analysis through data cleaning and formatting. This ensures data consistency while making it usable as structured data.

[0376] Next, the server uses machine learning algorithms to classify customers and predict their preferences. For example, it might classify customers into clusters such as travel enthusiasts or technology-oriented individuals. Python libraries such as scikit-learn and TensorFlow can be used for this purpose.

[0377] Subsequently, the server utilizes an emotion recognition device to analyze the user's dialogue data and estimate their emotional state in real time. Text, audio, and image data are analyzed using natural language processing and computer vision technologies to determine the user's emotions.

[0378] The server uses generative AI technology to create ad ideas based on the customer's preferences and emotional state. For example, it utilizes a language model like GPT-3 in its generative AI model, and by inputting prompts such as "Create a promotion that suits customers who love to travel and are excited," it generates effective ad ideas.

[0379] The generated ad proposals are notified to the user via their device. The device uses data reception and notification systems to deliver information to the user in real time. The user can review the received ad proposals and take action according to their content.

[0380] This format allows companies to automatically deploy personalized marketing strategies that respond to customer emotions and preferences, thereby increasing customer satisfaction.

[0381] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0382] Step 1:

[0383] The server collects customer information from the company's database. Inputs include customer purchase history, access logs, and contract information, while output is a dataset that integrates this information in a consistent format. Specifically, it executes database queries and prepares the obtained data through an ETL (Extract, Transform, Load) process.

[0384] Step 2:

[0385] The server preprocesses the aggregated data. The input is a unified dataset, and it performs data cleaning, duplicate removal, and missing value imputation to output a clean, analyzable dataset. Specifically, it uses the Python Pandas library for data formatting.

[0386] Step 3:

[0387] The server classifies customers using pre-processed data. The input is a clean dataset, and machine learning algorithms are applied to classify customers into specific clusters. The output is a classification label for each customer. Specifically, K-means clustering is used to analyze customer behavior patterns and generate cluster labels.

[0388] Step 4:

[0389] The server analyzes user dialogue data using an emotion recognition device. Inputs include text, audio, and image data, and emotion analysis technology outputs emotional states such as "joy" and "dissatisfaction." Specific operations include text analysis using natural language processing technology and tone analysis using speech analysis algorithms.

[0390] Step 5:

[0391] The server generates ad proposals using generative AI technology based on predicted preferences and emotional states. The input consists of customer cluster labels and emotional states, and the prompt text is fed into the generative AI model to generate ad proposals as output. Specifically, it automatically generates appropriate campaign messages using technologies such as OpenAI's GPT-3.

[0392] Step 6:

[0393] The device notifies the user of the generated ad draft. The input is the generated ad draft, and the output is the notification received by the user. Specifically, it provides information to the user in real time using push notifications and email distribution systems.

[0394] (Application Example 2)

[0395] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0396] Traditional promotion strategy generation systems rely solely on customer preferences and do not consider their emotional state. This has resulted in the inability to provide appropriate promotions that respond to customers' real-time emotions. Consequently, it is difficult to achieve improved customer satisfaction and efficient marketing activities.

[0397] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0398] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for estimating the user's emotional state in real time, means for adjusting and generating promotional measures according to the estimated emotional state, and means for notifying the generated promotional measures. This makes it possible to provide flexible and individualized promotional measures that respond to the customer's emotional state.

[0399] "Customer information" is a general term for all data related to a customer, including their purchase history, access logs, and contract information.

[0400] "Preprocessing" refers to the process of processing and organizing collected data to make it analyzable while maintaining its consistency.

[0401] "Customer classification" is the process of grouping customers based on the characteristics of the data, using the collected customer information.

[0402] "Contract information" refers to detailed information about the contract that a customer has entered into with a company, and includes contract terms, duration, and conditions.

[0403] "Preference estimation" is the process of predicting a customer's preferences and patterns based on their past behavior and data, using machine learning algorithms and other methods.

[0404] "Promotional measures" refer to advertising activities and campaign strategies implemented by companies to increase customer purchasing intent.

[0405] "Emotional state" refers to a customer's current emotions and is classified into various emotional categories such as joy, surprise, and dissatisfaction.

[0406] "Real-time estimation" is a process that immediately analyzes the situation and derives results as soon as a user's actions or interactions occur.

[0407] "Notification" is the act of sending information or messages from a system to a user, thereby communicating information to the user in real time.

[0408] This invention constructs a system that automatically generates optimal promotional strategies by analyzing customer information and estimating customer emotions in real time. The embodiments of the system are described in detail below.

[0409] The main components of the system are a server, terminals, and users. The server connects to the company's database and collects customer information such as purchase history, access logs, and contract information. The collected data is preprocessed and modified to maintain data consistency. This processing uses libraries such as Python's pandas library.

[0410] Next, the server uses machine learning algorithms to classify customers based on the collected data and estimate their preferences. For this purpose, machine learning libraries such as TensorFlow and PyTorch are utilized. The estimated preference information is then used to generate promotional strategies using generative AI models.

[0411] Next, the server uses an emotion engine to estimate the user's emotional state in real time. This emotion engine analyzes various data from the user, such as text, voice, and images, to identify emotions. Natural language processing (NLP) techniques are applied to the analysis to estimate the emotions contained in the text.

[0412] The device receives notifications of promotional campaigns sent from the server. Designed as a smartphone app, users can view tailored promotions in real time through the app. React Native is used for app development.

[0413] A specific use case would be when a user is online shopping and writes a review for a product they are considering purchasing. Positive emotions are detected from the review, and relevant discount coupons are immediately suggested. This can further increase the user's desire to buy.

[0414] An example of an input prompt statement for a generative AI model is written as follows:

[0415] Username: Taro Tanaka, Recent purchases: Books, Current cart: Books, Review: 'This book is very interesting!'

[0416] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0417] Step 1:

[0418] The server retrieves customer information from the company's database. This information includes purchase history, access logs, and contract information. The retrieved data is preprocessed using a data processing library such as Pandasm to prepare it for easy analysis while maintaining data consistency. The input is raw data, and the output is preprocessed data.

[0419] Step 2:

[0420] The server classifies customers using machine learning algorithms based on pre-processed customer information. This process uses TensorFlow and PyTorch to analyze features such as past purchasing behavior and contract information to estimate customer preferences. The input is pre-processed customer information, and the output is customer preference clusters.

[0421] Step 3:

[0422] The server uses a generative AI model to generate promotional strategies from estimated preferences. In this process, it generates promotional strategies corresponding to specific preferences, which are then used as sales and marketing measures. The input is customer preference clusters, and the output is promotional strategies.

[0423] Step 4:

[0424] The server uses an emotion engine to analyze the user's emotional state in real time. This engine analyzes text, voice, and image data obtained from the user to determine their emotions. It utilizes NLP (Neuro-Linguistic Programming) technology to estimate emotions from text. The input is user interaction data, and the output is the detected emotional state.

[0425] Step 5:

[0426] The server adjusts and generates promotional strategies in real time based on the detected emotional state. In this process, promotions and messages that match the emotion are generated by an AI model. The input is the emotional state and promotional strategy template, and the output is the adjusted promotional strategy.

[0427] Step 6:

[0428] The device notifies the user of the tailored promotional measures received from the server. It provides real-time push notifications, allowing the user to access the appropriate promotional measures through the application. The input is the tailored promotional measures, and the output is the notification to the user.

[0429] Step 7:

[0430] Users can review promotional offers received through their devices and make purchasing decisions based on their content. User behavior data is then fed back into the system for use in future analyses. The input is the promotional offer, and the output is user behavior data.

[0431] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0432] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0433] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0434] [Third Embodiment]

[0435] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0436] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0437] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0438] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0439] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0440] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0441] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0442] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0443] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0444] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0445] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0446] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0447] This invention relates to a system for efficiently processing vast amounts of customer information held by a company, enabling rapid contract procedures and effective promotional measures. This system consists of servers, terminals, and users, and functions by integrating various technologies.

[0448] First, the server automatically collects customer information from the company's database. The collected data is preprocessed to remove noise and convert it to the required format. Next, the server uses machine learning algorithms to classify customers into certain groups based on their purchasing patterns and behavioral history. Based on this classification, it estimates customer preferences and interests with high accuracy.

[0449] Next, the server uses natural language processing technology to automatically extract and organize contract information from contract documents and other materials. This makes it possible to quickly grasp contract renewal information and conditions. Furthermore, it generates optimal promotional strategies tailored to estimated customer preferences and notifies the terminal.

[0450] Users can review this information via their devices and approve or modify promotional strategies. This allows users to easily implement the most effective strategies for individual customers. For example, a fashion shop could use past purchase history to infer a customer's preferred brands and send timely notifications about new collections.

[0451] This format allows companies to provide personalized service to customers and significantly improve the overall efficiency of their operations.

[0452] The following describes the processing flow.

[0453] Step 1:

[0454] The server connects to the company's database and selects the customer information to retrieve. This information includes purchase history, access logs, contract history, etc. The server retrieves this data periodically or in real time.

[0455] Step 2:

[0456] The server performs preprocessing on the acquired customer information. Specifically, it checks for missing data, removes outliers, and standardizes data formats to create a clean dataset suitable for analysis.

[0457] Step 3:

[0458] The server uses machine learning algorithms to analyze customer behavior patterns from pre-processed data. This allows it to classify customers by characteristic and evaluate their behavioral tendencies and purchasing intent for each group.

[0459] Step 4:

[0460] The server uses natural language processing technology to extract contract information from the text of contracts and related documents. It organizes important contract elements such as dates, terms, and renewal status, and stores them in digital format.

[0461] Step 5:

[0462] The server analyzes customer preferences. Based on past purchase data and behavioral logs, it estimates products and services that customers might be interested in, and uses the results as reference data for promotional strategies.

[0463] Step 6:

[0464] The server generates optimal promotional strategies based on estimated customer preferences. This includes suggesting discounts and campaigns, and drafting email marketing messages.

[0465] Step 7:

[0466] The generated promotional campaign plan will be notified to the user on their device. The user can review the plan, make any necessary modifications, and then approve its implementation.

[0467] Step 8:

[0468] The user-approved measures are put into action. Promotional measures are deployed from the terminal via the server, and emails and online advertisements are sent to the target customers.

[0469] (Example 1)

[0470] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0471] Effectively utilizing the vast amount of customer data held by companies to automatically generate and notify customers of rapid contract procedures and personalized promotional strategies remains challenging. Traditional methods struggle to accurately understand customer preferences and make effective proposals at the right time, highlighting the need for improved operational efficiency.

[0472] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0473] In this invention, the server includes means for collecting content from information sources that store customer data, means for filtering the collected data and converting it into a standard format, and means for classifying customer behavior patterns using machine learning. This enables the automatic collection and organization of important information from vast amounts of customer data, and the creation and notification of personalized promotional strategies based on customer preferences.

[0474] "Customer data" refers to a collection of information about customers held by a company, including individual purchase history, contract information, and preference patterns.

[0475] "Information source" refers to a location where data is aggregated and recorded, such as a database or repository where a company stores customer data.

[0476] "Filtering" refers to the process of removing noise and unnecessary information from collected data and converting it into a format suitable for analysis.

[0477] "Converting to a standard format" refers to the operation of unifying data that exists in various formats into a consistent format, making it easier to process.

[0478] "Machine learning" is a technology in which computer systems learn patterns from data and perform predictions and classifications, and it is a means of analyzing customer data using algorithms.

[0479] "Classifying based on behavioral patterns" is the process of analyzing a customer's past behavioral history and dividing customers with similar characteristics into similar groups.

[0480] "Document analysis technology" refers to methods of extracting and structuring necessary information from documents using natural language processing.

[0481] An "advertising strategy" is a set of activities and measures planned to promote the sale of a product, and a method for making effective proposals based on customer preferences.

[0482] "Notification" refers to the act of communicating generated information or measures to users and prompting them to prepare for the implementation of those measures.

[0483] This invention is a system that provides services utilizing the vast amount of customer data held by a company, and its implementation is achieved through the cooperation of servers, terminals, and users.

[0484] First, the server connects to the database and collects customer data. This involves extracting necessary data from predefined sources, such as customer purchase history, contract information, and preference data. This collection process is more efficient when using distributed processing frameworks such as Apache Hadoop or Apache Spark.

[0485] Next, the server preprocesses the collected data. This preprocessing includes cleaning the data and converting it to a consistent format, utilizing data processing libraries in Python. Specifically, Pandas is used to format the data, impute missing values, and remove duplicate data.

[0486] The server then uses machine learning algorithms to classify customers. Using libraries such as Scikit-learn and TensorFlow, it analyzes customer purchasing behavior and preferences using K-means clustering and random forest classifiers, and classifies them into groups with similar patterns.

[0487] Based on these classification results, the server uses natural language processing techniques to extract and organize important information from the contract documents. In this step, Python's Natural Language Toolkit (NLTK) and SpaCy are used to automatically retrieve elements such as contract renewal dates and terms.

[0488] Next, the generative AI model is used to generate personalized promotional strategies on the server based on the estimated customer preferences. An example of a prompt used here is, "What is the best promotion for this customer?" By entering this prompt, the AI ​​model generates appropriate strategies.

[0489] The device is responsible for notifying users of the generated promotional campaigns. These notifications are delivered through applications using JavaScript or React Native, allowing users to review them and modify the campaigns as needed. For example, they can adjust discount rates or campaign durations.

[0490] Such systems enable companies to provide more effective and efficient services to their customers and optimize their business operations.

[0491] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0492] Step 1:

[0493] The server connects to the company's database and collects customer data. The input here is raw customer information obtained via database queries. Based on this input, the server uses Apache Hadoop to efficiently retrieve the data and temporarily stores the retrieved data in local storage.

[0494] Step 2:

[0495] The server preprocesses the collected customer data. The input for this step is the raw data obtained in step 1. Pandas is used to remove noise from the data, impute missing values, and perform standardization. The output is a clean and consistent dataset.

[0496] Step 3:

[0497] The server uses a machine learning algorithm to classify customers using the cleaned dataset. The input is the data preprocessed in step 2. Using Scikit-learn, K-means clustering is applied to classify customers into different groups. The output of this operation is a list of clustered customer groups.

[0498] Step 4:

[0499] The server uses Python's SpaCy to perform natural language processing and extract necessary contract information from customer contract documents. The input is contract documents held by companies. Through analysis, elements such as update dates and contract terms are extracted and organized. The output is a structured dataset with important information tagged.

[0500] Step 5:

[0501] The server generates promotional strategies using a generative AI model. The input for this step is the clustering results from step 3 and the contract information obtained in step 4. The prompt "What is the best promotion for this customer?" is passed to the AI ​​model to generate individual strategy suggestions. The output is a customized list of promotional strategies for each customer.

[0502] Step 6:

[0503] The device notifies the user of the promotion generated on the server. The input is the promotional initiative generated in step 5. These initiatives are sent to the user using push notifications in a mobile app using JavaScript or React Native. The output is the notification message sent to the user.

[0504] Step 7:

[0505] Users check notifications via their devices and modify promotional strategies as needed. User input involves detailed editing and approval of the strategies. Through this editing process, the output is an optimized promotional strategy with approvals and necessary changes completed.

[0506] (Application Example 1)

[0507] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0508] In today's business environment, companies are required to efficiently manage vast amounts of customer data and deliver services quickly. However, effectively utilizing this data and conducting promotions tailored to each customer is not easy. Furthermore, e-commerce sites need to accurately provide personalized product recommendations based on customers' purchasing behavior. This invention aims to solve these problems.

[0509] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0510] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for notifying the generated promotional measures, means for analyzing the user's purchase history, means for recommending products based on the user's purchase history, and means for notifying the terminal of the recommended product information. This enables companies to process customer information quickly and efficiently implement personalized promotions and product recommendations.

[0511] "Means of collecting customer information" refers to the processes or technologies used by a company to collect data about its customers.

[0512] "Preprocessing methods" refer to techniques that perform noise reduction and format conversion in order to prepare collected data into an analyzable format.

[0513] "Methods for classifying customers" refer to techniques for dividing customers into various groups based on their attributes and behavior.

[0514] "Methods for automatically extracting contract information" refers to technologies that mechanically extract necessary information from documents such as contracts.

[0515] "Methods for estimating preferences" refer to technologies that analyze a customer's past behavior and purchase history to predict their individual preferences and interests.

[0516] "Means of generating promotional strategies" refers to techniques for creating marketing and sales promotion strategies based on estimated customer preferences.

[0517] "Means of notifying about promotional measures" refers to technologies for delivering generated promotional information to end users' devices.

[0518] "Methods for analyzing purchasing behavior history" refer to technologies used to analyze a customer's purchase history and understand their purchasing patterns and trends.

[0519] "Methods for recommending products" refer to techniques for recommending specific products to customers based on collected information.

[0520] "Means of notifying a device of recommended product information" refers to technologies that display information about recommended products on the user's device.

[0521] In the system that implements this application, the server first collects customer information from the company's database and automatically preprocesses it. Preprocessing includes denoising the data and converting it to a standard format. The server uses machine learning algorithms to classify customers into several groups and estimate the preferences of each group. Next, it leverages natural language processing techniques to extract contract information and organize the contract terms and deadlines relevant to each customer.

[0522] Subsequently, the server generates optimal promotional strategies tailored to the estimated customer preferences and notifies the user's device. Through this device, the user can review the provided promotional information and approve or modify it as needed. Furthermore, based on the user's purchase history, the server recommends products and notifies the user's device accordingly.

[0523] The system is developed using Python and TensorFlow, with Pandas used for data analysis. The user interface is built with Flutter, and promotional and product information is notified in real time to devices such as smartphones. For example, if a customer purchased sports shoes the previous month, new running wear can be recommended based on their preferences.

[0524] Such systems enable quick and accurate responses tailored to user needs, and are expected to boost sales for businesses. Recommendation systems utilizing generative AI models provide personalized suggestions to individual customers using prompts such as, "Based on the items most frequently purchased in the past six months, suggest the most suitable products for this user."

[0525] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0526] Step 1:

[0527] The server automatically collects customer information from the company's database. It executes database queries to retrieve the necessary customer information and saves it to local storage. The input is raw data from the database, and the output is raw data containing noise.

[0528] Step 2:

[0529] The server preprocesses the customer information collected in the previous step. Denoising and format conversion are performed using the Python Pandas library. The converted data is then prepared in a format that can be analyzed. The input is the raw data output from step 1, and the output is the processed, clean dataset.

[0530] Step 3:

[0531] The server runs a machine learning algorithm using the processed data to classify customers into specific groups. TensorFlow is used for classification, and a clustering method is applied. The input is the clean data obtained in step 2, and the output is a list of customer groups.

[0532] Step 4:

[0533] The server extracts contract information using natural language processing technology. It automatically extracts and organizes necessary information from text data such as contracts. The input is contract-related text information from customers, and the output is structured contract information.

[0534] Step 5:

[0535] The server estimates customer preferences and generates promotional strategies based on that information. It uses a generative AI model for estimation, analyzing customer purchasing patterns using techniques such as random forests. Inputs are historical purchase data and lifestyle information, and the output is the optimal promotional strategy.

[0536] Step 6:

[0537] The server notifies the terminal of the generated promotional campaign. Real-time messaging is used for the notification, and the promotional information is displayed on the terminal immediately. The input is the promotional campaign created in step 5, and the output is the message displayed on the user's device.

[0538] Step 7:

[0539] Users review and approve or modify promotional information provided through their devices. The interface is built with Flutter, providing a user-friendly interface. Input is information displayed on the device, and output is user feedback on the promotional campaign.

[0540] Step 8:

[0541] The server analyzes the user's purchase history and recommends products. Collaborative filtering and deep learning models are used for recommendations to generate information that reflects the user's unique tendencies. The input is past purchase history, and the output is a list of recommended products.

[0542] Step 9:

[0543] The server notifies the terminal of recommended product information. Similarly, real-time notification technology is used to immediately prompt the user to purchase the product. The input is the product recommendation list obtained in step 8, and the output is the product information displayed on the user's terminal.

[0544] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0545] This invention is a system for efficiently processing customer information and automatically generating and notifying users of promotional strategies that take into account their emotional state. This system consists of a server, terminals, and users, and works in conjunction to collect and analyze customer information, recognize emotions, and generate promotional strategies.

[0546] First, the server connects to the company's database and collects customer purchase history, access logs, and various contract information, then performs preprocessing. This preprocessing ensures data consistency while making it ready for analysis. Next, machine learning algorithms are used to classify customers and estimate their preferences. This includes analysis based on customer characteristics derived from past purchase data and behavioral patterns.

[0547] Furthermore, an emotion engine is used to estimate the user's emotional state in real time based on their interactions. The emotion engine comprehensively analyzes data such as text, voice, and facial expressions to recognize emotions like joy, surprise, and dissatisfaction. Based on this, the server formulates promotional strategies that are appropriate for the current emotional state.

[0548] The device notifies the user of promotional initiatives sent from the server. The user reviews these initiatives and adjusts the implementation plan as needed. For example, in the travel industry, offering special travel plans or campaigns when a user is showing positive emotions can increase their purchasing intent.

[0549] This format allows companies to implement flexible and personalized marketing that responds to customer emotions, resulting in improved customer satisfaction and more efficient marketing activities.

[0550] The following describes the processing flow.

[0551] Step 1:

[0552] The server accesses the company's database and automatically collects customer purchase history, access logs, contract information, etc. During this process, the timeliness and completeness of the data are ensured.

[0553] Step 2:

[0554] The server preprocesses the collected information. Specifically, this involves cleaning the data, supplementing missing data, and correcting outliers. This process prepares the dataset for analysis.

[0555] Step 3:

[0556] The server applies machine learning algorithms to analyze customer behavior patterns from pre-processed data. This groups customers based on their purchasing tendencies and interests. Based on these analysis results, it estimates customer preferences.

[0557] Step 4:

[0558] The server uses natural language processing technology to extract necessary contract information from contracts and communication history. This includes information such as contract terms and renewal dates, and is stored in an organized format in the database.

[0559] Step 5:

[0560] Through interaction with the user, the emotion engine recognizes the user's emotions in real time. The emotion engine estimates the user's emotional state by analyzing facial expressions via text input, voice commands, and, in some cases, a camera.

[0561] Step 6:

[0562] The server considers the perceived emotions and generates optimal promotional strategies based on estimated customer preferences. This includes suggesting special offers and recommended products tailored to the emotional state.

[0563] Step 7:

[0564] The device notifies the user of promotional campaigns sent from the server. The user can view the details of the campaign on the device screen and, if necessary, customize the content of the campaign.

[0565] Step 8:

[0566] The device executes the promotional measures that the user has ultimately approved. Specifically, this includes sending emails to target customers and launching online campaigns. This ensures that the company's marketing strategy is effectively implemented.

[0567] (Example 2)

[0568] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0569] In today's information-saturated world, companies are required to effectively utilize large amounts of customer data and quickly develop advertising strategies that take into account the emotional state of individual customers. However, traditional methods have made it difficult to accurately grasp customer preferences and emotional states in real time and generate appropriate advertising proposals.

[0570] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0571] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for analyzing customer dialogue data with an emotion recognition device, means for creating advertising proposals using generative AI technology based on predicted preferences and emotional states, and means for notifying customers of the created advertising proposals via a terminal. This enables the automation of flexible and personalized advertising strategies that respond to customer emotions and preferences.

[0572] "Customer information" refers to data including purchase history, access history, and contract details related to a specific individual or legal entity.

[0573] "Means of aggregation" refers to methods or devices for efficiently collecting data and storing it in one place.

[0574] "Preprocessing" refers to the process of converting collected data into an analyzable format and performing cleansing and formatting adjustments as needed.

[0575] "Methods for classifying customers" refer to methods of dividing customers into groups based on specific attributes or behavioral patterns through data analysis.

[0576] "Preference" is a concept that refers to a customer's taste or priority regarding a particular product or service.

[0577] An "emotion recognition device" is a technology or device for detecting and analyzing an individual's emotional state from text, audio, or video.

[0578] "Generative AI technology" is a technique that uses artificial intelligence to gain insights from data and automatically generate new information and suggestions.

[0579] An "advertising proposal" is a strategy or plan created to effectively propose a specific product or service to customers.

[0580] "Means of notification" refers to a method or technique for transmitting specific information to a designated recipient.

[0581] This system is comprised of three main components: a server, terminals, and users. The server connects to the company's database to collect and pre-process customer information. Specifically, it collects customer purchase history, access history, and contract information, and prepares them for analysis through data cleaning and formatting. This ensures data consistency while making it usable as structured data.

[0582] Next, the server uses machine learning algorithms to classify customers and predict their preferences. For example, it might classify customers into clusters such as travel enthusiasts or technology-oriented individuals. Python libraries such as scikit-learn and TensorFlow can be used for this purpose.

[0583] Subsequently, the server utilizes an emotion recognition device to analyze the user's dialogue data and estimate their emotional state in real time. Text, audio, and image data are analyzed using natural language processing and computer vision technologies to determine the user's emotions.

[0584] The server uses generative AI technology to create ad ideas based on the customer's preferences and emotional state. For example, it utilizes a language model like GPT-3 in its generative AI model, and by inputting prompts such as "Create a promotion that suits customers who love to travel and are excited," it generates effective ad ideas.

[0585] The generated ad proposals are notified to the user via their device. The device uses data reception and notification systems to deliver information to the user in real time. The user can review the received ad proposals and take action according to their content.

[0586] This format allows companies to automatically deploy personalized marketing strategies that respond to customer emotions and preferences, thereby increasing customer satisfaction.

[0587] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0588] Step 1:

[0589] The server collects customer information from the company's database. Inputs include customer purchase history, access logs, and contract information, while output is a dataset that integrates this information in a consistent format. Specifically, it executes database queries and prepares the obtained data through an ETL (Extract, Transform, Load) process.

[0590] Step 2:

[0591] The server preprocesses the aggregated data. The input is a unified dataset, and it performs data cleaning, duplicate removal, and missing value imputation to output a clean, analyzable dataset. Specifically, it uses the Python Pandas library for data formatting.

[0592] Step 3:

[0593] The server classifies customers using pre-processed data. The input is a clean dataset, and machine learning algorithms are applied to classify customers into specific clusters. The output is a classification label for each customer. Specifically, K-means clustering is used to analyze customer behavior patterns and generate cluster labels.

[0594] Step 4:

[0595] The server analyzes user dialogue data using an emotion recognition device. Inputs include text, audio, and image data, and emotion analysis technology outputs emotional states such as "joy" and "dissatisfaction." Specific operations include text analysis using natural language processing technology and tone analysis using speech analysis algorithms.

[0596] Step 5:

[0597] The server generates ad proposals using generative AI technology based on predicted preferences and emotional states. The input consists of customer cluster labels and emotional states, and the prompt text is fed into the generative AI model to generate ad proposals as output. Specifically, it automatically generates appropriate campaign messages using technologies such as OpenAI's GPT-3.

[0598] Step 6:

[0599] The device notifies the user of the generated ad draft. The input is the generated ad draft, and the output is the notification received by the user. Specifically, it provides information to the user in real time using push notifications and email distribution systems.

[0600] (Application Example 2)

[0601] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0602] Traditional promotion strategy generation systems rely solely on customer preferences and do not consider their emotional state. This has resulted in the inability to provide appropriate promotions that respond to customers' real-time emotions. Consequently, it is difficult to achieve improved customer satisfaction and efficient marketing activities.

[0603] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0604] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for estimating the user's emotional state in real time, means for adjusting and generating promotional measures according to the estimated emotional state, and means for notifying the generated promotional measures. This makes it possible to provide flexible and individualized promotional measures that respond to the customer's emotional state.

[0605] "Customer information" is a general term for all data related to a customer, including their purchase history, access logs, and contract information.

[0606] "Preprocessing" refers to the process of processing and organizing collected data to make it analyzable while maintaining its consistency.

[0607] "Customer classification" is the process of grouping customers based on the characteristics of the data, using the collected customer information.

[0608] "Contract information" refers to detailed information about the contract that a customer has entered into with a company, and includes contract terms, duration, and conditions.

[0609] "Preference estimation" is the process of predicting a customer's preferences and patterns based on their past behavior and data, using machine learning algorithms and other methods.

[0610] "Promotional measures" refer to advertising activities and campaign strategies implemented by companies to increase customer purchasing intent.

[0611] "Emotional state" refers to a customer's current emotions and is classified into various emotional categories such as joy, surprise, and dissatisfaction.

[0612] "Real-time estimation" is a process that immediately analyzes the situation and derives results as soon as a user's actions or interactions occur.

[0613] "Notification" is the act of sending information or messages from a system to a user, thereby communicating information to the user in real time.

[0614] This invention constructs a system that automatically generates optimal promotional strategies by analyzing customer information and estimating customer emotions in real time. The embodiments of the system are described in detail below.

[0615] The main components of the system are a server, terminals, and users. The server connects to the company's database and collects customer information such as purchase history, access logs, and contract information. The collected data is preprocessed and modified to maintain data consistency. This processing uses libraries such as Python's pandas library.

[0616] Next, the server uses machine learning algorithms to classify customers based on the collected data and estimate their preferences. For this purpose, machine learning libraries such as TensorFlow and PyTorch are utilized. The estimated preference information is then used to generate promotional strategies using generative AI models.

[0617] Next, the server uses an emotion engine to estimate the user's emotional state in real time. This emotion engine analyzes various data from the user, such as text, voice, and images, to identify emotions. Natural language processing (NLP) techniques are applied to the analysis to estimate the emotions contained in the text.

[0618] The device receives notifications of promotional campaigns sent from the server. Designed as a smartphone app, users can view tailored promotions in real time through the app. React Native is used for app development.

[0619] A specific use case would be when a user is online shopping and writes a review for a product they are considering purchasing. Positive emotions are detected from the review, and relevant discount coupons are immediately suggested. This can further increase the user's desire to buy.

[0620] An example of an input prompt statement for a generative AI model is written as follows:

[0621] Username: Taro Tanaka, Recent purchases: Books, Current cart: Books, Review: 'This book is very interesting!'

[0622] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0623] Step 1:

[0624] The server retrieves customer information from the company's database. This information includes purchase history, access logs, and contract information. The retrieved data is preprocessed using a data processing library such as Pandasm to prepare it for easy analysis while maintaining data consistency. The input is raw data, and the output is preprocessed data.

[0625] Step 2:

[0626] The server classifies customers using machine learning algorithms based on pre-processed customer information. This process uses TensorFlow and PyTorch to analyze features such as past purchasing behavior and contract information to estimate customer preferences. The input is pre-processed customer information, and the output is customer preference clusters.

[0627] Step 3:

[0628] The server uses a generative AI model to generate promotional strategies from estimated preferences. In this process, it generates promotional strategies corresponding to specific preferences, which are then used as sales and marketing measures. The input is customer preference clusters, and the output is promotional strategies.

[0629] Step 4:

[0630] The server uses an emotion engine to analyze the user's emotional state in real time. This engine analyzes text, voice, and image data obtained from the user to determine their emotions. It utilizes NLP (Neuro-Linguistic Programming) technology to estimate emotions from text. The input is user interaction data, and the output is the detected emotional state.

[0631] Step 5:

[0632] The server adjusts and generates promotional strategies in real time based on the detected emotional state. In this process, promotions and messages that match the emotion are generated by an AI model. The input is the emotional state and promotional strategy template, and the output is the adjusted promotional strategy.

[0633] Step 6:

[0634] The device notifies the user of the tailored promotional measures received from the server. It provides real-time push notifications, allowing the user to access the appropriate promotional measures through the application. The input is the tailored promotional measures, and the output is the notification to the user.

[0635] Step 7:

[0636] Users can review promotional offers received through their devices and make purchasing decisions based on their content. User behavior data is then fed back into the system for use in future analyses. The input is the promotional offer, and the output is user behavior data.

[0637] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0638] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0639] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0640] [Fourth Embodiment]

[0641] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0642] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0643] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0644] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0645] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0646] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0647] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0648] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0649] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0650] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0651] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0652] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0653] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0654] This invention relates to a system for efficiently processing vast amounts of customer information held by a company, enabling rapid contract procedures and effective promotional measures. This system consists of servers, terminals, and users, and functions by integrating various technologies.

[0655] First, the server automatically collects customer information from the company's database. The collected data is preprocessed to remove noise and convert it to the required format. Next, the server uses machine learning algorithms to classify customers into certain groups based on their purchasing patterns and behavioral history. Based on this classification, it estimates customer preferences and interests with high accuracy.

[0656] Next, the server uses natural language processing technology to automatically extract and organize contract information from contract documents and other materials. This makes it possible to quickly grasp contract renewal information and conditions. Furthermore, it generates optimal promotional strategies tailored to estimated customer preferences and notifies the terminal.

[0657] Users can review this information via their devices and approve or modify promotional strategies. This allows users to easily implement the most effective strategies for individual customers. For example, a fashion shop could use past purchase history to infer a customer's preferred brands and send timely notifications about new collections.

[0658] This format allows companies to provide personalized service to customers and significantly improve the overall efficiency of their operations.

[0659] The following describes the processing flow.

[0660] Step 1:

[0661] The server connects to the company's database and selects the customer information to retrieve. This information includes purchase history, access logs, contract history, etc. The server retrieves this data periodically or in real time.

[0662] Step 2:

[0663] The server performs preprocessing on the acquired customer information. Specifically, it checks for missing data, removes outliers, and standardizes data formats to create a clean dataset suitable for analysis.

[0664] Step 3:

[0665] The server uses machine learning algorithms to analyze customer behavior patterns from pre-processed data. This allows it to classify customers by characteristic and evaluate their behavioral tendencies and purchasing intent for each group.

[0666] Step 4:

[0667] The server uses natural language processing technology to extract contract information from the text of contracts and related documents. It organizes important contract elements such as dates, terms, and renewal status, and stores them in digital format.

[0668] Step 5:

[0669] The server analyzes customer preferences. Based on past purchase data and behavioral logs, it estimates products and services that customers might be interested in, and uses the results as reference data for promotional strategies.

[0670] Step 6:

[0671] The server generates optimal promotional strategies based on estimated customer preferences. This includes suggesting discounts and campaigns, and drafting email marketing messages.

[0672] Step 7:

[0673] The generated promotional campaign plan will be notified to the user on their device. The user can review the plan, make any necessary modifications, and then approve its implementation.

[0674] Step 8:

[0675] The user-approved measures are put into action. Promotional measures are deployed from the terminal via the server, and emails and online advertisements are sent to the target customers.

[0676] (Example 1)

[0677] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0678] Effectively utilizing the vast amount of customer data held by companies to automatically generate and notify customers of rapid contract procedures and personalized promotional strategies remains challenging. Traditional methods struggle to accurately understand customer preferences and make effective proposals at the right time, highlighting the need for improved operational efficiency.

[0679] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0680] In this invention, the server includes means for collecting content from information sources that store customer data, means for filtering the collected data and converting it into a standard format, and means for classifying customer behavior patterns using machine learning. This enables the automatic collection and organization of important information from vast amounts of customer data, and the creation and notification of personalized promotional strategies based on customer preferences.

[0681] "Customer data" refers to a collection of information about customers held by a company, including individual purchase history, contract information, and preference patterns.

[0682] "Information source" refers to a location where data is aggregated and recorded, such as a database or repository where a company stores customer data.

[0683] "Filtering" refers to the process of removing noise and unnecessary information from collected data and converting it into a format suitable for analysis.

[0684] "Converting to a standard format" refers to the operation of unifying data that exists in various formats into a consistent format, making it easier to process.

[0685] "Machine learning" is a technology in which computer systems learn patterns from data and perform predictions and classifications, and it is a means of analyzing customer data using algorithms.

[0686] "Classifying based on behavioral patterns" is the process of analyzing a customer's past behavioral history and dividing customers with similar characteristics into similar groups.

[0687] "Document analysis technology" refers to methods of extracting and structuring necessary information from documents using natural language processing.

[0688] An "advertising strategy" is a set of activities and measures planned to promote the sale of a product, and a method for making effective proposals based on customer preferences.

[0689] "Notification" refers to the act of communicating generated information or measures to users and prompting them to prepare for the implementation of those measures.

[0690] This invention is a system that provides services utilizing the vast amount of customer data held by a company, and its implementation is achieved through the cooperation of servers, terminals, and users.

[0691] First, the server connects to the database and collects customer data. This involves extracting necessary data from predefined sources, such as customer purchase history, contract information, and preference data. This collection process is more efficient when using distributed processing frameworks such as Apache Hadoop or Apache Spark.

[0692] Next, the server preprocesses the collected data. This preprocessing includes cleaning the data and converting it to a consistent format, utilizing data processing libraries in Python. Specifically, Pandas is used to format the data, impute missing values, and remove duplicate data.

[0693] The server then uses machine learning algorithms to classify customers. Using libraries such as Scikit-learn and TensorFlow, it analyzes customer purchasing behavior and preferences using K-means clustering and random forest classifiers, and classifies them into groups with similar patterns.

[0694] Based on these classification results, the server uses natural language processing techniques to extract and organize important information from the contract documents. In this step, Python's Natural Language Toolkit (NLTK) and SpaCy are used to automatically retrieve elements such as contract renewal dates and terms.

[0695] Next, the generative AI model is used to generate personalized promotional strategies on the server based on the estimated customer preferences. An example of a prompt used here is, "What is the best promotion for this customer?" By entering this prompt, the AI ​​model generates appropriate strategies.

[0696] The device is responsible for notifying users of the generated promotional campaigns. These notifications are delivered through applications using JavaScript or React Native, allowing users to review them and modify the campaigns as needed. For example, they can adjust discount rates or campaign durations.

[0697] Such systems enable companies to provide more effective and efficient services to their customers and optimize their business operations.

[0698] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0699] Step 1:

[0700] The server connects to the company's database and collects customer data. The input here is raw customer information obtained via database queries. Based on this input, the server uses Apache Hadoop to efficiently retrieve the data and temporarily stores the retrieved data in local storage.

[0701] Step 2:

[0702] The server preprocesses the collected customer data. The input for this step is the raw data obtained in step 1. Pandas is used to remove noise from the data, impute missing values, and perform standardization. The output is a clean and consistent dataset.

[0703] Step 3:

[0704] The server uses a machine learning algorithm to classify customers using the cleaned dataset. The input is the data preprocessed in step 2. Using Scikit-learn, K-means clustering is applied to classify customers into different groups. The output of this operation is a list of clustered customer groups.

[0705] Step 4:

[0706] The server uses Python's SpaCy to perform natural language processing and extract necessary contract information from customer contract documents. The input is contract documents held by companies. Through analysis, elements such as update dates and contract terms are extracted and organized. The output is a structured dataset with important information tagged.

[0707] Step 5:

[0708] The server generates promotional strategies using a generative AI model. The input for this step is the clustering results from step 3 and the contract information obtained in step 4. The prompt "What is the best promotion for this customer?" is passed to the AI ​​model to generate individual strategy suggestions. The output is a customized list of promotional strategies for each customer.

[0709] Step 6:

[0710] The device notifies the user of the promotion generated on the server. The input is the promotional initiative generated in step 5. These initiatives are sent to the user using push notifications in a mobile app using JavaScript or React Native. The output is the notification message sent to the user.

[0711] Step 7:

[0712] Users check notifications via their devices and modify promotional strategies as needed. User input involves detailed editing and approval of the strategies. Through this editing process, the output is an optimized promotional strategy with approvals and necessary changes completed.

[0713] (Application Example 1)

[0714] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0715] In today's business environment, companies are required to efficiently manage vast amounts of customer data and deliver services quickly. However, effectively utilizing this data and conducting promotions tailored to each customer is not easy. Furthermore, e-commerce sites need to accurately provide personalized product recommendations based on customers' purchasing behavior. This invention aims to solve these problems.

[0716] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0717] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for notifying the generated promotional measures, means for analyzing the user's purchase history, means for recommending products based on the user's purchase history, and means for notifying the terminal of the recommended product information. This enables companies to process customer information quickly and efficiently implement personalized promotions and product recommendations.

[0718] "Means of collecting customer information" refers to the processes or technologies used by a company to collect data about its customers.

[0719] "Preprocessing methods" refer to techniques that perform noise reduction and format conversion in order to prepare collected data into an analyzable format.

[0720] "Methods for classifying customers" refer to techniques for dividing customers into various groups based on their attributes and behavior.

[0721] "Methods for automatically extracting contract information" refers to technologies that mechanically extract necessary information from documents such as contracts.

[0722] "Methods for estimating preferences" refer to technologies that analyze a customer's past behavior and purchase history to predict their individual preferences and interests.

[0723] "Means of generating promotional strategies" refers to techniques for creating marketing and sales promotion strategies based on estimated customer preferences.

[0724] "Means of notifying about promotional measures" refers to technologies for delivering generated promotional information to end users' devices.

[0725] "Methods for analyzing purchasing behavior history" refer to technologies used to analyze a customer's purchase history and understand their purchasing patterns and trends.

[0726] "Methods for recommending products" refer to techniques for recommending specific products to customers based on collected information.

[0727] "Means of notifying a device of recommended product information" refers to technologies that display information about recommended products on the user's device.

[0728] In the system that implements this application, the server first collects customer information from the company's database and automatically preprocesses it. Preprocessing includes denoising the data and converting it to a standard format. The server uses machine learning algorithms to classify customers into several groups and estimate the preferences of each group. Next, it leverages natural language processing techniques to extract contract information and organize the contract terms and deadlines relevant to each customer.

[0729] Subsequently, the server generates optimal promotional strategies tailored to the estimated customer preferences and notifies the user's device. Through this device, the user can review the provided promotional information and approve or modify it as needed. Furthermore, based on the user's purchase history, the server recommends products and notifies the user's device accordingly.

[0730] The system is developed using Python and TensorFlow, with Pandas used for data analysis. The user interface is built with Flutter, and promotional and product information is notified in real time to devices such as smartphones. For example, if a customer purchased sports shoes the previous month, new running wear can be recommended based on their preferences.

[0731] Such systems enable quick and accurate responses tailored to user needs, and are expected to boost sales for businesses. Recommendation systems utilizing generative AI models provide personalized suggestions to individual customers using prompts such as, "Based on the items most frequently purchased in the past six months, suggest the most suitable products for this user."

[0732] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0733] Step 1:

[0734] The server automatically collects customer information from the company's database. It executes database queries to retrieve the necessary customer information and saves it to local storage. The input is raw data from the database, and the output is raw data containing noise.

[0735] Step 2:

[0736] The server preprocesses the customer information collected in the previous step. Denoising and format conversion are performed using the Python Pandas library. The converted data is then prepared in a format that can be analyzed. The input is the raw data output from step 1, and the output is the processed, clean dataset.

[0737] Step 3:

[0738] The server runs a machine learning algorithm using the processed data to classify customers into specific groups. TensorFlow is used for classification, and a clustering method is applied. The input is the clean data obtained in step 2, and the output is a list of customer groups.

[0739] Step 4:

[0740] The server extracts contract information using natural language processing technology. It automatically extracts and organizes necessary information from text data such as contracts. The input is contract-related text information from customers, and the output is structured contract information.

[0741] Step 5:

[0742] The server estimates customer preferences and generates promotional strategies based on that information. It uses a generative AI model for estimation, analyzing customer purchasing patterns using techniques such as random forests. Inputs are historical purchase data and lifestyle information, and the output is the optimal promotional strategy.

[0743] Step 6:

[0744] The server notifies the terminal of the generated promotional campaign. Real-time messaging is used for the notification, and the promotional information is displayed on the terminal immediately. The input is the promotional campaign created in step 5, and the output is the message displayed on the user's device.

[0745] Step 7:

[0746] Users review and approve or modify promotional information provided through their devices. The interface is built with Flutter, providing a user-friendly interface. Input is information displayed on the device, and output is user feedback on the promotional campaign.

[0747] Step 8:

[0748] The server analyzes the user's purchase history and recommends products. Collaborative filtering and deep learning models are used for recommendations to generate information that reflects the user's unique tendencies. The input is past purchase history, and the output is a list of recommended products.

[0749] Step 9:

[0750] The server notifies the terminal of recommended product information. Similarly, real-time notification technology is used to immediately prompt the user to purchase the product. The input is the product recommendation list obtained in step 8, and the output is the product information displayed on the user's terminal.

[0751] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0752] This invention is a system for efficiently processing customer information and automatically generating and notifying users of promotional strategies that take into account their emotional state. This system consists of a server, terminals, and users, and works in conjunction to collect and analyze customer information, recognize emotions, and generate promotional strategies.

[0753] First, the server connects to the company's database and collects customer purchase history, access logs, and various contract information, then performs preprocessing. This preprocessing ensures data consistency while making it ready for analysis. Next, machine learning algorithms are used to classify customers and estimate their preferences. This includes analysis based on customer characteristics derived from past purchase data and behavioral patterns.

[0754] Furthermore, an emotion engine is used to estimate the user's emotional state in real time based on their interactions. The emotion engine comprehensively analyzes data such as text, voice, and facial expressions to recognize emotions like joy, surprise, and dissatisfaction. Based on this, the server formulates promotional strategies that are appropriate for the current emotional state.

[0755] The device notifies the user of promotional initiatives sent from the server. The user reviews these initiatives and adjusts the implementation plan as needed. For example, in the travel industry, offering special travel plans or campaigns when a user is showing positive emotions can increase their purchasing intent.

[0756] This format allows companies to implement flexible and personalized marketing that responds to customer emotions, resulting in improved customer satisfaction and more efficient marketing activities.

[0757] The following describes the processing flow.

[0758] Step 1:

[0759] The server accesses the company's database and automatically collects customer purchase history, access logs, contract information, etc. During this process, the timeliness and completeness of the data are ensured.

[0760] Step 2:

[0761] The server preprocesses the collected information. Specifically, this involves cleaning the data, supplementing missing data, and correcting outliers. This process prepares the dataset for analysis.

[0762] Step 3:

[0763] The server applies machine learning algorithms to analyze customer behavior patterns from pre-processed data. This groups customers based on their purchasing tendencies and interests. Based on these analysis results, it estimates customer preferences.

[0764] Step 4:

[0765] The server uses natural language processing technology to extract necessary contract information from contracts and communication history. This includes information such as contract terms and renewal dates, and is stored in an organized format in the database.

[0766] Step 5:

[0767] Through interaction with the user, the emotion engine recognizes the user's emotions in real time. The emotion engine estimates the user's emotional state by analyzing facial expressions via text input, voice commands, and, in some cases, a camera.

[0768] Step 6:

[0769] The server considers the perceived emotions and generates optimal promotional strategies based on estimated customer preferences. This includes suggesting special offers and recommended products tailored to the emotional state.

[0770] Step 7:

[0771] The device notifies the user of promotional campaigns sent from the server. The user can view the details of the campaign on the device screen and, if necessary, customize the content of the campaign.

[0772] Step 8:

[0773] The device executes the promotional measures that the user has ultimately approved. Specifically, this includes sending emails to target customers and launching online campaigns. This ensures that the company's marketing strategy is effectively implemented.

[0774] (Example 2)

[0775] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0776] In today's information-saturated world, companies are required to effectively utilize large amounts of customer data and quickly develop advertising strategies that take into account the emotional state of individual customers. However, traditional methods have made it difficult to accurately grasp customer preferences and emotional states in real time and generate appropriate advertising proposals.

[0777] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0778] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for analyzing customer dialogue data with an emotion recognition device, means for creating advertising proposals using generative AI technology based on predicted preferences and emotional states, and means for notifying customers of the created advertising proposals via a terminal. This enables the automation of flexible and personalized advertising strategies that respond to customer emotions and preferences.

[0779] "Customer information" refers to data including purchase history, access history, and contract details related to a specific individual or legal entity.

[0780] "Means of aggregation" refers to methods or devices for efficiently collecting data and storing it in one place.

[0781] "Preprocessing" refers to the process of converting collected data into an analyzable format and performing cleansing and formatting adjustments as needed.

[0782] "Methods for classifying customers" refer to methods of dividing customers into groups based on specific attributes or behavioral patterns through data analysis.

[0783] "Preference" is a concept that refers to a customer's taste or priority regarding a particular product or service.

[0784] An "emotion recognition device" is a technology or device for detecting and analyzing an individual's emotional state from text, audio, or video.

[0785] "Generative AI technology" is a technique that uses artificial intelligence to gain insights from data and automatically generate new information and suggestions.

[0786] An "advertising proposal" is a strategy or plan created to effectively propose a specific product or service to customers.

[0787] "Means of notification" refers to a method or technique for transmitting specific information to a designated recipient.

[0788] This system is comprised of three main components: a server, terminals, and users. The server connects to the company's database to collect and pre-process customer information. Specifically, it collects customer purchase history, access history, and contract information, and prepares them for analysis through data cleaning and formatting. This ensures data consistency while making it usable as structured data.

[0789] Next, the server uses machine learning algorithms to classify customers and predict their preferences. For example, it might classify customers into clusters such as travel enthusiasts or technology-oriented individuals. Python libraries such as scikit-learn and TensorFlow can be used for this purpose.

[0790] Subsequently, the server utilizes an emotion recognition device to analyze the user's dialogue data and estimate their emotional state in real time. Text, audio, and image data are analyzed using natural language processing and computer vision technologies to determine the user's emotions.

[0791] The server uses generative AI technology to create ad ideas based on the customer's preferences and emotional state. For example, it utilizes a language model like GPT-3 in its generative AI model, and by inputting prompts such as "Create a promotion that suits customers who love to travel and are excited," it generates effective ad ideas.

[0792] The generated ad proposals are notified to the user via their device. The device uses data reception and notification systems to deliver information to the user in real time. The user can review the received ad proposals and take action according to their content.

[0793] This format allows companies to automatically deploy personalized marketing strategies that respond to customer emotions and preferences, thereby increasing customer satisfaction.

[0794] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0795] Step 1:

[0796] The server collects customer information from the company's database. Inputs include customer purchase history, access logs, and contract information, while output is a dataset that integrates this information in a consistent format. Specifically, it executes database queries and prepares the obtained data through an ETL (Extract, Transform, Load) process.

[0797] Step 2:

[0798] The server preprocesses the aggregated data. The input is a unified dataset, and it performs data cleaning, duplicate removal, and missing value imputation to output a clean, analyzable dataset. Specifically, it uses the Python Pandas library for data formatting.

[0799] Step 3:

[0800] The server classifies customers using pre-processed data. The input is a clean dataset, and machine learning algorithms are applied to classify customers into specific clusters. The output is a classification label for each customer. Specifically, K-means clustering is used to analyze customer behavior patterns and generate cluster labels.

[0801] Step 4:

[0802] The server analyzes user dialogue data using an emotion recognition device. Inputs include text, audio, and image data, and emotion analysis technology outputs emotional states such as "joy" and "dissatisfaction." Specific operations include text analysis using natural language processing technology and tone analysis using speech analysis algorithms.

[0803] Step 5:

[0804] The server generates ad proposals using generative AI technology based on predicted preferences and emotional states. The input consists of customer cluster labels and emotional states, and the prompt text is fed into the generative AI model to generate ad proposals as output. Specifically, it automatically generates appropriate campaign messages using technologies such as OpenAI's GPT-3.

[0805] Step 6:

[0806] The device notifies the user of the generated ad draft. The input is the generated ad draft, and the output is the notification received by the user. Specifically, it provides information to the user in real time using push notifications and email distribution systems.

[0807] (Application Example 2)

[0808] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0809] Traditional promotion strategy generation systems rely solely on customer preferences and do not consider their emotional state. This has resulted in the inability to provide appropriate promotions that respond to customers' real-time emotions. Consequently, it is difficult to achieve improved customer satisfaction and efficient marketing activities.

[0810] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0811] In this invention, the server includes means for collecting customer information, means for pre-processing the collected customer information, means for classifying customers based on the customer information, means for automatically extracting customer contract information, means for estimating customer preferences, means for generating promotional measures based on the estimated preferences, means for estimating the user's emotional state in real time, means for adjusting and generating promotional measures according to the estimated emotional state, and means for notifying the generated promotional measures. This makes it possible to provide flexible and individualized promotional measures that respond to the customer's emotional state.

[0812] "Customer information" is a general term for all data related to a customer, including their purchase history, access logs, and contract information.

[0813] "Preprocessing" refers to the process of processing and organizing collected data to make it analyzable while maintaining its consistency.

[0814] "Customer classification" is the process of grouping customers based on the characteristics of the data, using the collected customer information.

[0815] "Contract information" refers to detailed information about the contract that a customer has entered into with a company, and includes contract terms, duration, and conditions.

[0816] "Preference estimation" is the process of predicting a customer's preferences and patterns based on their past behavior and data, using machine learning algorithms and other methods.

[0817] "Promotional measures" refer to advertising activities and campaign strategies implemented by companies to increase customer purchasing intent.

[0818] "Emotional state" refers to a customer's current emotions and is classified into various emotional categories such as joy, surprise, and dissatisfaction.

[0819] "Real-time estimation" is a process that immediately analyzes the situation and derives results as soon as a user's actions or interactions occur.

[0820] "Notification" is the act of sending information or messages from a system to a user, thereby communicating information to the user in real time.

[0821] This invention constructs a system that automatically generates optimal promotional strategies by analyzing customer information and estimating customer emotions in real time. The embodiments of the system are described in detail below.

[0822] The main components of the system are a server, terminals, and users. The server connects to the company's database and collects customer information such as purchase history, access logs, and contract information. The collected data is preprocessed and modified to maintain data consistency. This processing uses libraries such as Python's pandas library.

[0823] Next, the server uses machine learning algorithms to classify customers based on the collected data and estimate their preferences. For this purpose, machine learning libraries such as TensorFlow and PyTorch are utilized. The estimated preference information is then used to generate promotional strategies using generative AI models.

[0824] Next, the server uses an emotion engine to estimate the user's emotional state in real time. This emotion engine analyzes various data from the user, such as text, voice, and images, to identify emotions. Natural language processing (NLP) techniques are applied to the analysis to estimate the emotions contained in the text.

[0825] The device receives notifications of promotional campaigns sent from the server. Designed as a smartphone app, users can view tailored promotions in real time through the app. React Native is used for app development.

[0826] A specific use case would be when a user is online shopping and writes a review for a product they are considering purchasing. Positive emotions are detected from the review, and relevant discount coupons are immediately suggested. This can further increase the user's desire to buy.

[0827] An example of an input prompt statement for a generative AI model is written as follows:

[0828] Username: Taro Tanaka, Recent purchases: Books, Current cart: Books, Review: 'This book is very interesting!'

[0829] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0830] Step 1:

[0831] The server retrieves customer information from the company's database. This information includes purchase history, access logs, and contract information. The retrieved data is preprocessed using a data processing library such as Pandasm to prepare it for easy analysis while maintaining data consistency. The input is raw data, and the output is preprocessed data.

[0832] Step 2:

[0833] The server classifies customers using machine learning algorithms based on pre-processed customer information. This process uses TensorFlow and PyTorch to analyze features such as past purchasing behavior and contract information to estimate customer preferences. The input is pre-processed customer information, and the output is customer preference clusters.

[0834] Step 3:

[0835] The server uses a generative AI model to generate promotional strategies from estimated preferences. In this process, it generates promotional strategies corresponding to specific preferences, which are then used as sales and marketing measures. The input is customer preference clusters, and the output is promotional strategies.

[0836] Step 4:

[0837] The server uses an emotion engine to analyze the user's emotional state in real time. This engine analyzes text, voice, and image data obtained from the user to determine their emotions. It utilizes NLP (Neuro-Linguistic Programming) technology to estimate emotions from text. The input is user interaction data, and the output is the detected emotional state.

[0838] Step 5:

[0839] The server adjusts and generates promotional strategies in real time based on the detected emotional state. In this process, promotions and messages that match the emotion are generated by an AI model. The input is the emotional state and promotional strategy template, and the output is the adjusted promotional strategy.

[0840] Step 6:

[0841] The device notifies the user of the tailored promotional measures received from the server. It provides real-time push notifications, allowing the user to access the appropriate promotional measures through the application. The input is the tailored promotional measures, and the output is the notification to the user.

[0842] Step 7:

[0843] Users can review promotional offers received through their devices and make purchasing decisions based on their content. User behavior data is then fed back into the system for use in future analyses. The input is the promotional offer, and the output is user behavior data.

[0844] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0845] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0846] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0847] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0848] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0849] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0850] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0851] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0852] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0853] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0854] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0855] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0856] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0857] 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.

[0858] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0859] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0860] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0861] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0862] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0863] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0864] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0865] The following is further disclosed regarding the embodiments described above.

[0866] (Claim 1)

[0867] Means of collecting customer information,

[0868] A means for pre-processing the collected customer information,

[0869] A means of classifying customers based on customer information,

[0870] A means of automatically extracting customer contract information,

[0871] A means of estimating customer preferences,

[0872] A means of generating promotional strategies based on estimated preferences,

[0873] A system that includes means for notifying users of generated promotional measures.

[0874] (Claim 2)

[0875] The system according to claim 1, which extracts customer contract information using natural language processing technology.

[0876] (Claim 3)

[0877] The system according to claim 1, which uses a machine learning algorithm to analyze customer preferences.

[0878] "Example 1"

[0879] (Claim 1)

[0880] Means for collecting content from information sources that store customer data,

[0881] A means of filtering the collected data and converting it to a standard format,

[0882] A method of classifying customers based on their behavior patterns using machine learning,

[0883] A method using natural language processing technology to analyze important contractual elements from documents,

[0884] A means of predicting customer preferences by utilizing the classified results,

[0885] A means of building advertising strategies based on predicted preferences,

[0886] A system that includes means for notifying users of the advertising strategy that has been developed.

[0887] (Claim 2)

[0888] The system according to claim 1, which acquires contract information using document analysis technology.

[0889] (Claim 3)

[0890] The system according to claim 1, which uses a learning algorithm to identify customer interests.

[0891] "Application Example 1"

[0892] (Claim 1)

[0893] Means of collecting customer information,

[0894] A means for pre-processing the collected customer information,

[0895] A means of classifying customers based on customer information,

[0896] A means of automatically extracting customer contract information,

[0897] A means of estimating customer preferences,

[0898] A means of generating promotional strategies based on estimated preferences,

[0899] A means of notifying about the generated promotional measures,

[0900] Methods for analyzing users' purchasing behavior history,

[0901] A means of recommending products based on the user's purchase history,

[0902] A method for notifying the device of recommended product information.

[0903] A system that includes this.

[0904] (Claim 2)

[0905] The system according to claim 1, which extracts customer contract information using natural language processing technology.

[0906] (Claim 3)

[0907] The system according to claim 1, which uses a machine learning algorithm to analyze customer preferences.

[0908] "Example 2 of combining an emotion engine"

[0909] (Claim 1)

[0910] Means for collecting customer information,

[0911] A means of pre-processing the collected customer information,

[0912] A means of classifying customers based on customer information,

[0913] Means for predicting customer preferences,

[0914] A means of analyzing customer dialogue data with an emotion recognition device,

[0915] A means of creating advertising proposals using generative AI technology based on predicted preferences and emotional states,

[0916] A system that includes a means of notifying users of created advertisement proposals via a terminal.

[0917] (Claim 2)

[0918] The system according to claim 1, which analyzes customer dialogue data to estimate their emotional state.

[0919] (Claim 3)

[0920] The system according to claim 1, which analyzes customer preferences using machine learning methods.

[0921] "Application example 2 when combining with an emotional engine"

[0922] (Claim 1)

[0923] Means of collecting customer information,

[0924] A means for pre-processing the collected customer information,

[0925] A means of classifying customers based on customer information,

[0926] A means of automatically extracting customer contract information,

[0927] A means of estimating customer preferences,

[0928] A means of generating promotional strategies based on estimated preferences,

[0929] A means of estimating the user's emotional state in real time,

[0930] A means of adjusting and generating promotional measures according to the estimated emotional state,

[0931] A system that includes means for notifying users of generated promotional measures.

[0932] (Claim 2)

[0933] The system according to claim 1, which extracts customer contract information using natural language processing technology.

[0934] (Claim 3)

[0935] The system according to claim 1, which uses a machine learning algorithm to analyze customer preferences and perform emotion recognition. [Explanation of Symbols]

[0936] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means of collecting customer information, A means for pre-processing the collected customer information, A means of classifying customers based on customer information, A means of automatically extracting customer contract information, A means of estimating customer preferences, A means of generating promotional strategies based on estimated preferences, A means of notifying about the generated promotional measures, Methods for analyzing users' purchasing behavior history, A means of recommending products based on the user's purchase history, A method for notifying the device of recommended product information. A system that includes this.

2. The system according to claim 1, which extracts customer contract information using natural language processing technology.

3. The system according to claim 1, which uses a machine learning algorithm to analyze customer preferences.

Images