system

Abstract

We provide the system. [Solution] A means for users to input their preferred customer service style, A means of collecting users' past purchase history and preference data, A means of generating a virtual customer service representative suited to the customer service style based on collected user data, A means of allowing the generated virtual customer service representative to interact with the user, A means of answering user questions in real time, A means of providing effective talk information in real time, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot performed by at least one processor, including the 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 Document

Patent Document

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the modern consumer market, a more personalized customer experience that better responds to consumer needs is required. In conventional systems, it is difficult to fully reflect the individual needs of consumers, and there are problems such as ineffective real-time conversations and proposals. Furthermore, in the customer experience in physical stores or the metaverse environment, there is also a lack of effective means to attract consumers. A customer service system using specific famous people or characters has also been complicated and difficult to implement in the past. It is necessary to solve these problems, increase consumers' purchasing desire, and provide a more satisfying purchasing experience.

Means for Solving the Problems

[0005] The present invention is a system that includes means for the user to input their desired customer service style, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for the generated virtual customer service representative to interact with the user, means for answering the user's questions in real time, and means for providing effective talk information in real time. Furthermore, the system has a configuration that includes means for the user to select a specific person model from among the customer service styles they desire, means for collecting data on the selected person model, means for generating a virtual customer service representative based on the specific person model based on the collected data, and means for providing the user with limited offers and cross-sell suggestions in real time. As a result, the user can obtain an ideal customer service experience, and in particular, receiving effective talk and offers in real time increases their desire to purchase, resulting in a highly satisfying purchasing experience.

[0006] "Customer service style" refers to the type of service, attitude, and interaction that users prefer.

[0007] "User's past purchase history" refers to a record of products and services that a user has purchased in the past.

[0008] "Preference data" refers to information that indicates a user's preferences, interests, and concerns.

[0009] A "virtual customer service representative" refers to a virtual customer service representative created using computer generation.

[0010] "Means of interaction with the user" refers to methods and devices for users to communicate with virtual customer service representatives.

[0011] "Means of providing real-time responses" refers to technologies and methods that enable immediate responses to user questions.

[0012] "Effective sales pitch information" refers to persuasive information that attracts the user's attention and increases their desire to purchase.

[0013] A "human model" refers to data that is modeled after a specific celebrity or character.

[0014] A "limited-time offer" refers to a special offer or discount provided for a specific period or under specific conditions.

[0015] "Cross-selling" refers to suggesting other products related to a product that a user is interested in.

[0016] "Means of input" refers to the methods or devices that users use to provide information to a system.

[0017] "Means of collection" refers to the technologies and methods used to gather data and information.

[0018] "Means of generation" refers to the technologies and methods used to construct new information or objects based on data.

[0019] "Means of provision" refers to the technologies and methods used to provide information and services to users. [Brief explanation of the drawing]

[0020] [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]It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It 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] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It 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] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0021] 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.

[0022] First, the language used in the following description will be explained.

[0023] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0024] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0025] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0026] 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).

[0027] 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."

[0028] [First Embodiment]

[0029] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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".

[0041] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[0042] System Processing Flow Overview

[0043] User's preference input

[0044] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a customer service style selection form, such as "casual and friendly customer service style." This information is sent from the device to the server.

[0045] Consumer data collection

[0046] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0047] Virtual Customer Service Representative Generation

[0048] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0049] Virtual customer service representative display and interaction

[0050] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[0051] User interaction

[0052] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[0053] Generating answers to questions

[0054] The server generates responses based on the collected product information. The generated response data is sent to the terminal, which then displays it to the user. For example, specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes offers excellent cushioning and is ideal for long-distance running," might be displayed.

[0055] Providing an effective talk

[0056] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal displays this information to the user, offering special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0057] Customer service scenarios with celebrities (optional feature)

[0058] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[0059] Specific example

[0060] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[0061] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[0062] In this way, the system of the present invention provides a practical solution for enhancing purchasing intent by offering a personalized customer service experience that meets the user's needs.

[0063] The following describes the processing flow.

[0064] Step 1:

[0065] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[0066] Step 2:

[0067] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[0068] Step 3:

[0069] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[0070] Step 4:

[0071] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[0072] Step 5:

[0073] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0074] Step 6:

[0075] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[0076] Step 7:

[0077] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[0078] Step 8:

[0079] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[0080] Step 9:

[0081] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0082] Step 10:

[0083] The server sends the generated response data to the terminal. The terminal displays the response to the user.

[0084] Step 11:

[0085] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual customer service representative.

[0086] Step 12:

[0087] The terminal displays exclusive offer information to the user via a virtual customer service representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also recommend socks that go perfectly with these shoes."

[0088] Through these steps, users will be able to have an ideal customer service experience, which will likely increase their desire to purchase.

[0089] (Example 1)

[0090] 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."

[0091] Traditional online shopping systems required users to spend a lot of time finding the products they wanted to buy, and lacked personalized customer service experiences tailored to individual needs. Furthermore, it was difficult for users with specific interests or styles to receive appropriate advice and suggestions. This could lead to decreased purchasing intent and missed sales opportunities. Additionally, it was challenging to adequately accommodate users who desired customer service representatives based on specific celebrities or characters.

[0092] 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.

[0093] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for transmitting the generated virtual customer service representative's text data, talk script, and character information to the terminal, means for displaying the virtual customer service representative, means for transmitting the user's questions from the terminal to the server, means for searching for appropriate product information from a database based on the user's questions, and means for generating answers from a generated AI model based on the collected product information. This makes it possible to provide a personalized customer service experience that matches the user's wishes and interests, thereby increasing their willingness to purchase. Furthermore, it is possible to generate virtual customer service representatives based on specific celebrities or characters and provide personalized responses to users.

[0094] A "user" refers to a person who uses the system to search for and purchase goods and services.

[0095] "Device" refers to electronic devices used by users, such as smartphones, tablets, and personal computers.

[0096] A "server" refers to a central system that processes data, stores information, and handles requests from terminals.

[0097] "Customer service style" refers to the way a virtual customer service representative responds and their attitude, as chosen by the user.

[0098] "Purchase history" refers to a record of products that a user has purchased through the system so far.

[0099] "Preference data" refers to data that includes information about users' preferences and interests.

[0100] A "virtual customer service representative" refers to a virtual character or assistant created using generative AI to provide customer service to users.

[0101] "Interaction" refers to the dialogue and exchange of information that takes place between the user and the virtual customer service representative.

[0102] A "generative AI model" refers to an artificial intelligence model that generates text or responses based on user input or data.

[0103] A "talk script" refers to a script or pattern for a virtual customer service representative to use when speaking to a user.

[0104] "Character information" refers to data about the appearance, personality, and speaking style of the virtual customer service representative.

[0105] A "prompt statement" refers to a command statement that is input to a generating AI model.

[0106] A "database" refers to an entire system used for collecting, storing, and retrieving data.

[0107] "Real-time" refers to processing or responding to user input almost simultaneously.

[0108] A "specific person model" refers to a virtual customer service representative generated based on data of a celebrity or character selected by the user.

[0109] "Limited-time offers" refer to information about special products or discounts that are available only under specific conditions.

[0110] "Cross-selling" refers to a marketing technique where related products are suggested simultaneously when a customer purchases a particular product.

[0111] Modes for carrying out the invention

[0112] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[0113] Hardware and software to be used

[0114] Devices: Smartphones, tablets, PCs, etc.

[0115] Server: A central system that processes data, stores information, and handles requests from each terminal.

[0116] Database: Stores users' past purchase history and preference data (e.g., MySQL®).

[0117] Generative AI models: AI models for generating natural language (e.g., OpenAI®, GPT-4®)

[0118] UI framework: A framework for displaying content on a device (e.g., React Native)

[0119] Operating procedures and specific examples

[0120] Users input their preferred customer service style using a device such as a smartphone or computer. For example, they might enter "casual and friendly customer service style" and press the send button. The device then sends this information to the server.

[0121] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server's process begins with executing a SELECT query to extract all data related to the user ID. Next, edge analytics is used to filter the collected data and create a data set that matches the user's requests.

[0122] The server generates a virtual customer service representative using a generative AI (for example, OpenAI's GPT-4 model) based on the collected data and user requests. A prompt is used during the generation process. An example of a prompt might be, "Generate a virtual customer service representative with a casual customer service style based on the user's preferences." The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal in JSON format.

[0123] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. Using a UI framework such as React Native, the virtual customer service representative's avatar is displayed on the screen and greets the user with "Hello! What products are you looking for today?"

[0124] When a user enters the question "What running shoes do you recommend?", the device sends this question to the server. The device receives the user's input and sends it to the server as a POST request in JSON format.

[0125] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to collect the relevant product information. After that, it uses a generative AI model to create an answer. The generated answer data is sent to the terminal in JSON format.

[0126] The device analyzes the received data and displays a response to the user. For example, it can make specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0127] Furthermore, the server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal receives this information, and the sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that would go perfectly with these shoes."

[0128] Furthermore, if a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the device, which then displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?"

[0129] The above describes embodiments of the present invention, which are detailed procedures for providing a personalized customer service experience tailored to user needs and increasing purchasing intent.

[0130] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0131] Step 1: User input

[0132] The user uses their device to input their preferred service style. Specifically, they launch an application on their device and enter a service style selection form, such as "casual and friendly service style." This action causes the device to send the user's input data to the server. An HTTP POST request is used for this transmission. The input data, "casual and friendly service style," is sent to the server.

[0133] Step 2: Consumer Data Collection

[0134] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server first executes a SELECT query to extract all data related to the user ID. This data includes items the user has purchased, browsing history, and favorite categories. The extracted data is then filtered using edge analytics to create a data set that matches the user's requirements. The input is the user ID, and the output is the filtered data set.

[0135] Step 3: Generate a virtual customer service representative

[0136] The server generates a virtual customer service representative using a generative AI (e.g., OpenAI's GPT-4 model) based on the collected data and user requests. In this process, a prompt is input to the generative AI model. For example, "Generate a virtual customer service representative with a casual service style based on the user's preferences" might be the prompt. The AI ​​model analyzes this prompt, generates text data, talk scripts, and character information for a virtual customer service representative that meets the user's preferences, and sends it to the terminal in JSON format. The input is a filtered data set and a prompt, and the output is virtual customer service representative data in JSON format.

[0137] Step 4: Virtual customer service representative display and interaction

[0138] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. A UI framework (e.g., React Native) is used for this. The terminal screen displays an avatar of the virtual customer service representative, who says to the user, "Hello! What products are you looking for today?" The input is virtual customer service representative data in JSON format, and the output is the virtual customer service representative displayed on the screen.

[0139] Step 5: Interaction with the user

[0140] When a user enters the question "What running shoes do you recommend?", the terminal sends this question to the server. The terminal receives the user's input and sends it to the server as a POST request in JSON format. The input is the user's question, and the output is the request to the server.

[0141] Step 6: Generate answers to the questions

[0142] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to extract the relevant product information. After that, it uses a generative AI model to create an answer and sends the generated answer data to the terminal in JSON format. The input is the user's question, and the output is the answer data in JSON format.

[0143] Step 7: Displaying the response to the user

[0144] The device parses the received JSON data and displays a response to the user. For example, it might display specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The input is response data in JSON format, and the output is the response displayed on the screen.

[0145] Step 8: Delivering an effective talk

[0146] The server generates effective chat information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The generated chat information is sent from the server to the terminal. The terminal receives this information, and the virtual sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes." The input is the generated chat information, and the output is the chat information displayed on the screen.

[0147] Step 9: Customer service scenario with a celebrity (optional feature)

[0148] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the terminal, which displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?" The input is the data of the selected person model, and the output is the virtual customer service representative displayed on the screen.

[0149] (Application Example 1)

[0150] 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."

[0151] Current customer service systems offer limited personalized service experiences. Furthermore, there's a challenge in providing effective recommendations to users in physical stores by leveraging their past purchase history and preference data. Additionally, there are no systems that can generate virtual sales representatives tailored to user preferences and provide interactive product suggestions within the store. This results in users being unable to fully develop their purchasing intent and making optimal product selection difficult.

[0152] 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.

[0153] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model. This makes it possible to provide users with a personalized customer service experience even in a physical store and increase their willingness to purchase.

[0154] "Customer service style" refers to the method and attitude of interaction that users prefer.

[0155] "Purchase history" refers to information that records a list of products a user has purchased in the past.

[0156] "Preference data" refers to information that reflects a user's preferences and interests.

[0157] A "virtual customer service representative" refers to a customer service representative created virtually using generative AI, who interacts with the user on screen.

[0158] "Interaction" refers to a two-way exchange between the user and the system.

[0159] "Real-time" refers to responding immediately to user input or actions.

[0160] "Talk information" refers to information that includes the content of conversations with users and the suggestions they made.

[0161] A "physical store" refers to a retail store that provides goods and services in a physical location.

[0162] A "smartphone application" refers to a software program that runs on a smartphone.

[0163] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate text and responses.

[0164] A "prompt message" refers to an instruction message that is input into a generative AI model.

[0165] The system of the present invention includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model.

[0166] Hardware and software

[0167] server

[0168] The server includes a database (e.g., MySQL or PostgreSQL) to store user purchase history and preference data. It also provides computing resources to generate responses from virtual customer service representatives using generative AI models (e.g., GPT-3® or the latest ChatGPT® technology). Furthermore, it generates prompt messages and sends the generated virtual customer service representative data to the terminal.

[0169] terminal

[0170] The terminal is a mobile device such as a smartphone or tablet, and interaction between the user and a virtual customer service representative is realized through a dedicated application. The application receives user input and sends it to a server. Based on the data received from the server, the virtual customer service representative is displayed, and real-time interaction takes place.

[0171] System Operation Overview

[0172] User's preference input

[0173] Users enter their preferred customer service style using a smartphone application. For example, they might enter "casual and friendly customer service style."

[0174] Data collection

[0175] The server collects past purchase history and preference data from the database based on the user's ID. This data includes items the user has purchased, browsing history, and favorite categories.

[0176] Generation of virtual customer service representatives

[0177] Based on the collected data and user preferences, the server generates prompt messages, which are then input into a generation AI model to create a virtual customer service representative. Examples of prompt messages are as follows:

[0178] User preference: Casual and friendly customer service style

[0179] User data: Past purchase history: running shoes, sportswear. Favorite category: sporting goods.

[0180] Interaction

[0181] The generated virtual customer service representative data is sent to the terminal and displayed within the smartphone application. The virtual customer service representative begins by saying something like, "Hello! What products are you looking for today?" and interacts with the user.

[0182] Providing real-time answers and talk information.

[0183] When a user asks a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server searches its database for appropriate product information and sends the generated answer back to the terminal. Effective sales talk information, such as limited-time offers and cross-selling suggestions, is also provided in real time.

[0184] This makes it possible to provide users with a personalized customer service experience through interaction with virtual sales representatives, thereby increasing their willingness to purchase.

[0185] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0186] Step 1:

[0187] Users input their desired customer service style into a smartphone application. Specifically, they might enter "casual and friendly customer service style" and submit that information. This input data is then sent from the device to the server.

[0188] Input: User's preferred customer service style (e.g., "Casual and friendly customer service style")

[0189] Output: User's requested input data sent to the server

[0190] Step 2:

[0191] The server uses the user's ID to collect past purchase history and preference data from the database. Specifically, it retrieves information such as products the user has purchased, browsing history, and favorite categories from the database. This collected data is used in subsequent processes.

[0192] Input: User ID

[0193] Output: User's past purchase history and preference data

[0194] Step 3:

[0195] The server generates prompt sentences suitable for the generating AI model based on the collected user data. These prompt sentences include the user's desired customer service style and user data. This is then input into the generating AI model to create a virtual customer service representative.

[0196] Input: User's desired input data, past purchase history, and preference data.

[0197] Output: Prompt text to input into the generating AI model (Example: "User preference: Casual and friendly customer service style. User data: Past purchase history: Running shoes, sportswear. Favorite category: Sports goods.")

[0198] Step 4:

[0199] The server uses a generation AI model to generate a virtual customer service representative based on the prompt text. It then sends the generated virtual customer service representative's text data, talk script, and character information to the terminal.

[0200] Input: Prompt message

[0201] Output: Generated virtual customer service representative data (text data, talk script, character information)

[0202] Step 5:

[0203] The terminal displays the virtual customer service representative's data received from the server. Interaction with the user begins, and the virtual customer service representative starts speaking to the user, saying things like, "Hello! What kind of products are you looking for today?"

[0204] Input: Data of the generated virtual customer service representative

[0205] Output: Virtual customer service representative displayed on the screen

[0206] Step 6:

[0207] The user enters a question into the virtual customer service representative. For example, they might type, "What running shoes would you recommend?" This question is then sent from the terminal to the server.

[0208] Input: User's question (e.g., "What running shoes do you recommend?")

[0209] Output: User questions sent to the server

[0210] Step 7:

[0211] The server analyzes the user's question and searches the database for appropriate product information. Specifically, it searches for recommended running shoe models and features, and generates answer data.

[0212] Input: User's question

[0213] Output: Response data generated based on appropriate product information

[0214] Step 8:

[0215] The server sends the generated response data to the terminal. This data includes specific suggestions, such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0216] Input: Generated response data

[0217] Output: Sending response data to the terminal

[0218] Step 9:

[0219] The terminal displays the received response data to the user. Suggestions from a virtual customer service representative are displayed on the screen, increasing the user's purchasing intent.

[0220] Input: Response data sent from the server

[0221] Output: Response data displayed on the user screen

[0222] Step 10:

[0223] The server generates effective sales pitches, exclusive offers, and cross-selling suggestions in real time and incorporates them into the virtual sales representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also show you socks that would go perfectly with them."

[0224] Input: User's current interaction status, reward information

[0225] Output: Generation of talk data incorporating bonus information

[0226] Step 11:

[0227] The terminal displays updated virtual customer service representatives' conversation information in real time, providing users with effective conversation and special offer information. This can further increase users' purchasing intent.

[0228] Input: Talk data sent from the server

[0229] Output: Talk information from the virtual customer service representative displayed on the screen in real time.

[0230] 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.

[0231] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[0232] System Processing Flow Overview

[0233] User's preference input

[0234] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a selection form, such as "casual and friendly customer service style," and presses the submit button. This information is sent from the device to the server.

[0235] Consumer data collection

[0236] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0237] Virtual Customer Service Representative Generation

[0238] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0239] Virtual customer service representative display and interaction

[0240] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[0241] User emotion recognition

[0242] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. For example, when a user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[0243] Adjusting interactions

[0244] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[0245] User interaction

[0246] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[0247] Generating answers to questions

[0248] The server generates responses based on the collected product information. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated response data is sent to the terminal, which then displays it to the user.

[0249] Providing an effective talk

[0250] The server generates effective sales talk information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0251] Customer service scenarios with celebrities (optional feature)

[0252] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[0253] Specific example

[0254] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[0255] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[0256] During this time, the emotion engine analyzes the user's facial expressions and tone of voice, and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time and look around."

[0257] In this way, the system of the present invention provides a personalized customer service experience that meets the user's needs and further enables responses that take into account the user's emotions in real time, thereby increasing purchasing intent and providing a highly satisfying purchasing experience.

[0258] The following describes the processing flow.

[0259] Step 1:

[0260] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[0261] Step 2:

[0262] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[0263] Step 3:

[0264] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[0265] Step 4:

[0266] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[0267] Step 5:

[0268] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0269] Step 6:

[0270] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[0271] Step 7:

[0272] The device uses its built-in camera and microphone to activate an emotion engine that analyzes the user's facial expressions and voice tone in real time. The emotion engine detects the user's emotional state from their facial expressions and voice tone.

[0273] Step 8:

[0274] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[0275] Step 9:

[0276] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[0277] Step 10:

[0278] The server generates a response based on the product information it has collected. For example, it generates information such as "This pair of shoes is lightweight and has excellent breathability. The next pair of shoes has high cushioning and is optimal for long-distance running."

[0279] Step 11:

[0280] The server sends the response data it has generated to the terminal. The terminal displays the response to the user. For example, product features and price information are displayed.

[0281] Step 12:

[0282] The emotion engine analyzes the user's facial expressions and voice tones to identify whether the user is calm, tired, excited, etc. The emotion engine sends the identified emotional state to the server.

[0283] Step 13:

[0284] Based on the identified emotional state, the server adjusts the response and tone of the virtual receptionist. For example, if the user is judged to be tired, the virtual receptionist greets the user in a gentle tone such as "You seem tired today. Please take your time to look around."

[0285] Step 14:

[0286] The server generates real-time effective conversation information (e.g., limited offers or cross-sell proposals) and sends the data to be incorporated into the virtual receptionist to the terminal. As a result, the virtual receptionist provides special offer information such as "Right now, this pair of shoes is 20% off! Also, I'd like to introduce socks that are perfect for these shoes."

[0287] Step 15:

[0288] The device displays exclusive offers and cross-sell suggestions to the user via a virtual sales representative. If the user is interested in the information, more detailed information is provided to increase their purchase intent.

[0289] In this way, the system of the present invention provides a personalized customer service experience based on the user's wishes and real-time emotional state, thereby increasing the user's desire to purchase. Furthermore, by providing effective conversational information and adjusting responses based on emotion recognition, it realizes a highly satisfying purchasing experience.

[0290] (Example 2)

[0291] 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".

[0292] In modern online shopping, users demand personalized customer service experiences, but current systems fail to meet these demands. In particular, there is a lack of customized customer service styles tailored to user preferences and in-depth responses that consider user emotional states. Furthermore, the creation of virtual customer service representatives using specific person models, as well as real-time limited-time offers and cross-selling suggestions, are not yet realized. A system is needed to address these challenges and provide users with a highly satisfying purchasing experience.

[0293] 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.

[0294] In this invention, the server includes means for inputting the user's desired customer service style, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for displaying the generated virtual customer service representative on the user terminal and allowing interaction, means for recognizing the user's emotional state using an emotion engine, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotional state, means for searching for appropriate product information from a database based on the user's questions and generating answers, and means for providing effective talk information in real time. This enables a personalized customer service experience that meets the user's needs and responses that respond to their emotional state, and also enables the provision of limited offers and cross-selling suggestions in real time.

[0295] A "user" refers to an individual who uses the system to input their preferred customer service style or interact with a virtual customer service representative.

[0296] A "terminal" is a device used by a user to access a system and input or receive information, and includes smartphones, tablets, and personal computers.

[0297] A "server" is a central control unit that collects, processes, and generates data based on user input.

[0298] "Customer service style" refers to the communication characteristics and attitudes of the virtual customer service representative that the user desires.

[0299] "User data" refers to a collection of data related to a user, including their past purchase history and preference data.

[0300] A "virtual customer service representative" refers to a virtual character that is generated based on the user's preferences and data, and that interacts with the user online.

[0301] "Interaction" refers to the two-way communication that takes place between the user and the virtual receptionist.

[0302] "Emotion engine" refers to the technology that analyzes the user's facial expressions, voice tones, and input text to identify the user's emotional state.

[0303] "Emotional state" refers to the user's psychological and emotional situation, including, for example, being tired or excited.

[0304] "Response" refers to the reply or reaction made by the virtual receptionist to the user's questions or requests.

[0305] "Tone" refers to the way of speaking and the attitude of the virtual receptionist.

[0306] "Database" refers to an information management system for storing user data, product information, etc.

[0307] "Product information" refers to the detailed data about the products stored in the system, including features, prices, inventory status, etc.

[0308] "Conversation information" refers to the conversation content and its progress used by the virtual receptionist in the interaction with the user.

[0309] "Limited offer" refers to the discounts or benefits provided specifically for the purchase of certain products.

[0310] "Cross-sell proposal" refers to the proposal of additional products related to the product that the user is considering purchasing.

[0311] "Person model" refers to a set of data for generating a virtual receptionist based on the characteristics of a specific famous person or character.

[0312] The customer service system of this invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[0313] Users input their preferred customer service style using devices such as smartphones, tablets, or personal computers. For example, a user might enter "casual and friendly customer service style" into a selection form and press the submit button. This information is then sent from the device to the server.

[0314] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0315] The server generates virtual customer service representatives using a generation AI (e.g., the latest chatGPT technology). During this process, the server inputs prompt text into the generation AI, which then generates text data, talk scripts, and character information for the virtual customer service representative. This data is then sent to the terminal.

[0316] The terminal displays a virtual sales representative on the screen and begins interacting with the user. For example, the virtual sales representative might start by saying, "Hello! What kind of product are you looking for today?"

[0317] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. When the user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[0318] The server receives emotional data sent from the terminal and adjusts the virtual customer service representative's response and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[0319] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (for example, recommended running shoe models and their features). Based on the collected product information, the server generates an answer. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated answer data is sent to the terminal, which then displays it to the user.

[0320] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0321] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the terminal. The terminal displays this virtual customer service representative and begins interacting with the user.

[0322] Specific example

[0323] The user selects a "casual and friendly customer service style" in the smartphone app and presses the send button. The device sends this information to the server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a casual, sports-savvy virtual customer service representative and sends that data to the device. The device displays the virtual customer service representative and begins interacting with the user. When the user asks, "What running shoes do you recommend?", the device sends the question to the server, which generates an answer based on appropriate product information and sends it to the device. The device displays the generated answer to the user and also displays more effective talk and limited offers to increase the user's desire to purchase. During this time, an emotion engine analyzes the user's facial expressions and voice tone and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying, "You seem tired today. Please take your time browsing."

[0324] Example of a prompt

[0325] "Specify the user's preferred customer service style as 'casual and friendly,' and generate text recommending running shoes based on their past purchase history and preference data."

[0326] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0327] Step 1:

[0328] The user enters their preferred customer service style.

[0329] The user opens the app on their device (e.g., smartphone, tablet, or computer) and enters their preferred customer service style.

[0330] The user enters "Casual and friendly customer service style" into the customer service style selection form and presses the submit button.

[0331] Input: User's preferred customer service style

[0332] Output: Preferred customer service style information is sent from the terminal to the server.

[0333] Step 2:

[0334] User data collection

[0335] The server accesses the database based on the user's ID.

[0336] The server collects the user's past purchase history and preference data. This data includes items the user has previously purchased, browsing history, and favorite categories.

[0337] The server filters the collected data and creates a dataset that matches the user's requirements.

[0338] Input: User ID, preferred customer service style

[0339] Output: Filtered historical purchase history and preference data

[0340] Step 3:

[0341] Generation of virtual customer service representatives

[0342] The server generates prompt messages based on the collected user data and user requests.

[0343] The server inputs prompt text into a generating AI (e.g., the latest chatGPT technology) to create a virtual customer service representative.

[0344] The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal.

[0345] Input: Filtered user data, preferred customer service style information

[0346] Output: Text data, talk scripts, and character information of the virtual customer service representative.

[0347] Step 4:

[0348] Display of the virtual customer service representative and initiation of interaction.

[0349] The device displays a virtual customer service representative on the screen.

[0350] The virtual customer service representative begins by saying, "Hello! What kind of product are you looking for today?"

[0351] Input: Text data, talk script, and character information of the virtual customer service representative.

[0352] Output: A virtual customer service representative is displayed on the user's terminal, and the interaction begins.

[0353] Step 5:

[0354] User emotion recognition

[0355] The device has a built-in emotion engine that analyzes the user's facial expressions, voice tone, and input text.

[0356] When a user is visible through the camera, the device uses facial recognition technology to identify their emotional state.

[0357] When a user speaks, the device performs voice tone analysis to understand emotional nuances.

[0358] Input: User facial expression data, voice tone data, input text

[0359] Output: The emotional state of the identified user.

[0360] Step 6:

[0361] Response and tone adjustment

[0362] The server receives emotion data sent from the terminal.

[0363] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak in a gentler tone.

[0364] Input: Identified user's emotional state, virtual customer service representative's text data

[0365] Output: Adjusted virtual customer service representative responses and tone

[0366] Step 7:

[0367] Handling user inquiries

[0368] The user enters the question, "What running shoes do you recommend?"

[0369] The terminal sends the question to the server.

[0370] Input: User's question

[0371] Output: Question data sent to the server

[0372] Step 8:

[0373] Generating answers to questions

[0374] The server analyzes the user's question.

[0375] The server searches the database for appropriate product information (e.g., recommended running shoe models and their features).

[0376] The server generates a response based on the collected product information.

[0377] Input: User's question, product information in the database

[0378] Output: Generated response data

[0379] Step 9:

[0380] Submit and display of responses

[0381] The server sends the generated response data to the terminal.

[0382] The device displays the generated response to the user.

[0383] Input: Generated response data

[0384] Output: Answer displayed on the user's terminal

[0385] Step 10:

[0386] Providing effective communication information

[0387] The server generates effective, real-time sales information (e.g., limited-time offers and cross-selling suggestions).

[0388] The server sends data to the terminal to be incorporated into the virtual customer service representative.

[0389] A virtual customer service representative provides users with special offers such as, "These shoes are 20% off right now! We can also recommend socks that would go perfectly with these shoes."

[0390] Input: Real-time conversation information, text data from virtual customer service representatives

[0391] Output: Virtual customer service representative speech with effective conversational information incorporated.

[0392] Step 11:

[0393] Customer service scenarios with celebrities (optional feature)

[0394] When a user selects a customer service style, they can choose a specific person model (for example, a celebrity).

[0395] The server collects data from the selected person model (e.g., voice data, feature data).

[0396] The server generates a virtual customer service representative based on a specific person model and sends it to the terminal.

[0397] The terminal displays this virtual customer service representative and begins interacting with the user.

[0398] Input: User-selected person model, data for the selected person model

[0399] Output: Data and display of a virtual customer service representative generated based on a specific person model.

[0400] Through these steps, the system provides a personalized customer service experience tailored to the user's needs, and further enhances purchasing intent and delivers a highly satisfying shopping experience by enabling real-time responses that take the user's emotions into account.

[0401] (Application Example 2)

[0402] 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".

[0403] Traditional in-store customer service systems faced challenges in providing personalized service tailored to the individual needs and emotions of each user. Furthermore, the inability to make effective real-time suggestions and cross-selling proposals limited the ability to improve customer satisfaction and boost sales. Additionally, there was a lack of customer service systems that could leverage new technologies such as smart glasses.

[0404] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for analyzing the user's facial expressions and voice tone to recognize emotions, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotions, and means for interacting with the virtual customer service representative by wearing smart glasses in a physical store. This enables a personalized customer service experience that responds to the user's needs and emotions, thereby improving customer satisfaction and promoting sales.

[0405] A "virtual customer service representative" is a virtual customer service representative generated based on the user's preferences and collected data.

[0406] "Interaction" refers to two-way communication between the user and the virtual customer service representative.

[0407] "User's past purchase history" refers to a list of products and historical data that the user has purchased so far.

[0408] "Preference data" refers to data related to a user's preferences and interests, and is primarily based on past behavior and preferences.

[0409] "Facial recognition" is a technology that analyzes a user's facial expressions to identify their emotional state at that time.

[0410] "Voice tone" is a technique that identifies a user's emotional state by analyzing the nuances of their voice, including its pitch and volume.

[0411] "Smart glasses" are glasses-type devices that have the function of displaying information from the real world as augmented reality (AR).

[0412] "Real-time response" refers to a response that provides appropriate information immediately in response to a user's question.

[0413] "Effective sales information" refers to information generated to promote sales and make suggestions, tailored to the user's interests and needs.

[0414] "Cross-selling" is a sales strategy that suggests related products that users might be interested in.

[0415] The system program for implementing this invention operates in cooperation between a server and a terminal and is designed to provide users with a personalized customer service experience. This system includes the following main components:

[0416] Component and Processing Overview

[0417] server

[0418] The server is primarily responsible for data collection, execution of generative AI models, and response generation.

[0419] 1. Collection of user data:

[0420] The system collects the user's past purchase history and preference data from a database. This includes a list of products the user has purchased, their browsing history, and their preferences.

[0421] 2. Virtual customer service representative generation:

[0422] Based on the collected data and the user's preferred customer service style, a virtual customer service representative is generated using a generative AI model (e.g., the T5 model). This generation process involves inputting appropriate prompts to the generative AI model and obtaining character information for the virtual customer service representative.

[0423] 3. Real-time response to questions:

[0424] The system analyzes user inquiries and generates appropriate product information and answers. The generated information is then sent to the device.

[0425] terminal

[0426] The device is responsible for user interaction and presents data in real time.

[0427] 1. Emotion recognition:

[0428] The device is equipped with a camera and a voice input device, and recognizes emotions by analyzing the user's facial expressions and voice tone. This analysis utilizes facial recognition technology such as the DeepFace library and voice tone analysis technology.

[0429] 2. Performing an interaction:

[0430] The terminal displays a generated virtual customer service representative and interacts with the user. For example, it provides an environment where the user can converse with a virtual customer service representative through smart glasses.

[0431] 3. Adjusting the response:

[0432] Based on recognized emotion data, the system receives instructions from the server and adjusts the virtual customer service representative's responses and tone. For example, if the user is tired, it will respond in a gentle tone.

[0433] Specific example

[0434] Let's assume the user is wearing smart glasses in a physical store. These smart glasses are equipped with a camera and microphone, and can analyze the user's facial expressions and voice tone in real time.

[0435] When a user requests a "casual and friendly customer service style," the device sends this information to the server. The server collects the user's past purchase history and preference data, and uses a generative AI model to generate an appropriate virtual customer service representative. The generated virtual representative will engage in friendly conversation, for example, saying, "Hello! What products are you looking for today?"

[0436] An example of a prompt is, "Create a virtual assistant with sports and casual preferences and a casual and friendly interaction style." This prompt prompt causes the AI ​​model to generate a virtual customer service representative that aligns with the user's preferences.

[0437] In this way, personalized customer service experiences tailored to user needs and emotions can be provided. This system is highly effective in improving customer satisfaction and boosting sales.

[0438] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0439] Step 1:

[0440] The user inputs their desired customer service style through a device such as smart glasses or a smartphone. The entered information, along with the user's ID, is sent to the server. The input data is in text format and represents the user's preferences.

[0441] Step 2:

[0442] Based on the received user preference data, the server collects the user's past purchase history and preference data from the database. Database queries are used to retrieve a list of products the user has previously purchased, their browsing history, and their preferences. The retrieved data is temporarily stored as a user dataset.

[0443] Step 3:

[0444] The server inputs a prompt message into a generative AI model based on the collected user dataset, generating a virtual customer service representative. The prompt message is "Create a virtual assistant with sports and casual preferences and a casual and friendly interaction style." and the text data output from the generative model (e.g., T5 model) is retrieved. The output text data contains the character information of the virtual customer service representative.

[0445] Step 4:

[0446] The device initiates interaction with the user based on the character information of the virtual customer service representative received from the server. The virtual customer service representative is displayed on the smart glasses' screen and speaks to the user, saying things like, "Hello! What kind of product are you looking for today?" During this process, the displayed text and audio data are generated as interaction data.

[0447] Step 5:

[0448] The device uses its built-in camera and microphone to analyze the user's facial expressions and voice tone, and uses an emotion recognition engine to identify the user's emotions. It analyzes facial expression data using the DeepFace library and voice tone analysis technology to analyze voice data. Detected emotions are stored as emotion data.

[0449] Step 6:

[0450] The server receives emotional data and runs an algorithm that adjusts the virtual customer service representative's responses and tone based on it. For example, if the emotional data identifies the user as "tired," the virtual customer service representative's response will be changed to a gentler tone. This provides the user with a pleasant customer service experience.

[0451] Step 7:

[0452] When a user enters a specific question, the device sends that question to the server. The server analyzes the question and searches its database for appropriate product information. The search results are generated as answer data and sent back to the device.

[0453] Step 8:

[0454] The terminal displays response data received from the server to the user and provides answers through a virtual customer service representative. For example, it might present information such as, "These shoes are lightweight and highly breathable." In this process, the displayed text and audio data are generated as response data.

[0455] Step 9:

[0456] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and sends it to the terminal. The terminal then incorporates this information into the virtual sales representative and presents it to the user. This allows the server to provide users with special offers such as, "These shoes are 20% off right now."

[0457] 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.

[0458] 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 (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.

[0459] 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.

[0460] [Second Embodiment]

[0461] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0462] 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.

[0463] 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).

[0464] 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.

[0465] 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.

[0466] 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).

[0467] 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.

[0468] 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.

[0469] 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.

[0470] 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.

[0471] 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.

[0472] 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".

[0473] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[0474] System Processing Flow Overview

[0475] User's preference input

[0476] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a customer service style selection form, such as "casual and friendly customer service style." This information is sent from the device to the server.

[0477] Consumer data collection

[0478] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0479] Virtual Customer Service Representative Generation

[0480] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0481] Virtual customer service representative display and interaction

[0482] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[0483] User interaction

[0484] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[0485] Generating answers to questions

[0486] The server generates responses based on the collected product information. The generated response data is sent to the terminal, which then displays it to the user. For example, specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes offers excellent cushioning and is ideal for long-distance running," might be displayed.

[0487] Providing an effective talk

[0488] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal displays this information to the user, offering special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0489] Customer service scenarios with celebrities (optional feature)

[0490] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[0491] Specific example

[0492] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[0493] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[0494] In this way, the system of the present invention provides a practical solution for enhancing purchasing intent by offering a personalized customer service experience that meets the user's needs.

[0495] The following describes the processing flow.

[0496] Step 1:

[0497] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[0498] Step 2:

[0499] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[0500] Step 3:

[0501] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[0502] Step 4:

[0503] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[0504] Step 5:

[0505] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0506] Step 6:

[0507] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[0508] Step 7:

[0509] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[0510] Step 8:

[0511] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[0512] Step 9:

[0513] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0514] Step 10:

[0515] The server sends the generated response data to the terminal. The terminal displays the response to the user.

[0516] Step 11:

[0517] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual customer service representative.

[0518] Step 12:

[0519] The terminal displays exclusive offer information to the user via a virtual customer service representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also recommend socks that go perfectly with these shoes."

[0520] Through these steps, users will be able to have an ideal customer service experience, which will likely increase their desire to purchase.

[0521] (Example 1)

[0522] 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."

[0523] Traditional online shopping systems required users to spend a lot of time finding the products they wanted to buy, and lacked personalized customer service experiences tailored to individual needs. Furthermore, it was difficult for users with specific interests or styles to receive appropriate advice and suggestions. This could lead to decreased purchasing intent and missed sales opportunities. Additionally, it was challenging to adequately accommodate users who desired customer service representatives based on specific celebrities or characters.

[0524] 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.

[0525] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for transmitting the generated virtual customer service representative's text data, talk script, and character information to the terminal, means for displaying the virtual customer service representative, means for transmitting the user's questions from the terminal to the server, means for searching for appropriate product information from a database based on the user's questions, and means for generating answers from a generated AI model based on the collected product information. This makes it possible to provide a personalized customer service experience that matches the user's wishes and interests, thereby increasing their willingness to purchase. Furthermore, it is possible to generate virtual customer service representatives based on specific celebrities or characters and provide personalized responses to users.

[0526] A "user" refers to a person who uses the system to search for and purchase goods and services.

[0527] "Device" refers to electronic devices used by users, such as smartphones, tablets, and personal computers.

[0528] A "server" refers to a central system that processes data, stores information, and handles requests from terminals.

[0529] "Customer service style" refers to the way a virtual customer service representative responds and their attitude, as chosen by the user.

[0530] "Purchase history" refers to a record of products that a user has purchased through the system so far.

[0531] "Preference data" refers to data that includes information about users' preferences and interests.

[0532] A "virtual customer service representative" refers to a virtual character or assistant created using generative AI to provide customer service to users.

[0533] "Interaction" refers to the dialogue and exchange of information that takes place between the user and the virtual customer service representative.

[0534] A "generative AI model" refers to an artificial intelligence model that generates text or responses based on user input or data.

[0535] A "talk script" refers to a script or pattern for a virtual customer service representative to use when speaking to a user.

[0536] "Character information" refers to data about the appearance, personality, and speaking style of the virtual customer service representative.

[0537] A "prompt statement" refers to a command statement that is input to a generating AI model.

[0538] A "database" refers to an entire system used for collecting, storing, and retrieving data.

[0539] "Real-time" refers to processing or responding to user input almost simultaneously.

[0540] A "specific person model" refers to a virtual customer service representative generated based on data of a celebrity or character selected by the user.

[0541] "Limited-time offers" refer to information about special products or discounts that are available only under specific conditions.

[0542] "Cross-selling" refers to a marketing technique where related products are suggested simultaneously when a customer purchases a particular product.

[0543] Modes for carrying out the invention

[0544] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[0545] Hardware and software to be used

[0546] Devices: Smartphones, tablets, PCs, etc.

[0547] Server: A central system that processes data, stores information, and handles requests from each terminal.

[0548] Database: Stores users' past purchase history and preference data (e.g., MySQL).

[0549] Generative AI models: AI models for generating natural language (e.g., OpenAI GPT-4)

[0550] UI framework: A framework for displaying content on a device (e.g., React Native)

[0551] Operating procedures and specific examples

[0552] Users input their preferred customer service style using a device such as a smartphone or computer. For example, they might enter "casual and friendly customer service style" and press the send button. The device then sends this information to the server.

[0553] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server's process begins with executing a SELECT query to extract all data related to the user ID. Next, edge analytics is used to filter the collected data and create a data set that matches the user's requests.

[0554] The server generates a virtual customer service representative using a generative AI (for example, OpenAI's GPT-4 model) based on the collected data and user requests. A prompt is used during the generation process. An example of a prompt might be, "Generate a virtual customer service representative with a casual customer service style based on the user's preferences." The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal in JSON format.

[0555] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. Using a UI framework such as React Native, the virtual customer service representative's avatar is displayed on the screen and greets the user with "Hello! What products are you looking for today?"

[0556] When a user enters the question "What running shoes do you recommend?", the device sends this question to the server. The device receives the user's input and sends it to the server as a POST request in JSON format.

[0557] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to collect the relevant product information. After that, it uses a generative AI model to create an answer. The generated answer data is sent to the terminal in JSON format.

[0558] The device analyzes the received data and displays a response to the user. For example, it can make specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0559] Furthermore, the server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal receives this information, and the sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that would go perfectly with these shoes."

[0560] Furthermore, if a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the device, which then displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?"

[0561] The above describes embodiments of the present invention, which are detailed procedures for providing a personalized customer service experience tailored to user needs and increasing purchasing intent.

[0562] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0563] Step 1: User input

[0564] The user uses their device to input their preferred service style. Specifically, they launch an application on their device and enter a service style selection form, such as "casual and friendly service style." This action causes the device to send the user's input data to the server. An HTTP POST request is used for this transmission. The input data, "casual and friendly service style," is sent to the server.

[0565] Step 2: Consumer Data Collection

[0566] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server first executes a SELECT query to extract all data related to the user ID. This data includes items the user has purchased, browsing history, and favorite categories. The extracted data is then filtered using edge analytics to create a data set that matches the user's requirements. The input is the user ID, and the output is the filtered data set.

[0567] Step 3: Generate a virtual customer service representative

[0568] The server generates a virtual customer service representative using a generative AI (e.g., OpenAI's GPT-4 model) based on the collected data and user requests. In this process, a prompt is input to the generative AI model. For example, "Generate a virtual customer service representative with a casual service style based on the user's preferences" might be the prompt. The AI ​​model analyzes this prompt, generates text data, talk scripts, and character information for a virtual customer service representative that meets the user's preferences, and sends it to the terminal in JSON format. The input is a filtered data set and a prompt, and the output is virtual customer service representative data in JSON format.

[0569] Step 4: Virtual customer service representative display and interaction

[0570] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. A UI framework (e.g., React Native) is used for this. The terminal screen displays an avatar of the virtual customer service representative, who says to the user, "Hello! What products are you looking for today?" The input is virtual customer service representative data in JSON format, and the output is the virtual customer service representative displayed on the screen.

[0571] Step 5: Interaction with the user

[0572] When a user enters the question "What running shoes do you recommend?", the terminal sends this question to the server. The terminal receives the user's input and sends it to the server as a POST request in JSON format. The input is the user's question, and the output is the request to the server.

[0573] Step 6: Generate answers to the questions

[0574] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to extract the relevant product information. After that, it uses a generative AI model to create an answer and sends the generated answer data to the terminal in JSON format. The input is the user's question, and the output is the answer data in JSON format.

[0575] Step 7: Displaying the response to the user

[0576] The device parses the received JSON data and displays a response to the user. For example, it might display specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The input is response data in JSON format, and the output is the response displayed on the screen.

[0577] Step 8: Delivering an effective talk

[0578] The server generates effective chat information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The generated chat information is sent from the server to the terminal. The terminal receives this information, and the virtual sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes." The input is the generated chat information, and the output is the chat information displayed on the screen.

[0579] Step 9: Customer service scenario with a celebrity (optional feature)

[0580] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the terminal, which displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?" The input is the data of the selected person model, and the output is the virtual customer service representative displayed on the screen.

[0581] (Application Example 1)

[0582] 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."

[0583] Current customer service systems offer limited personalized service experiences. Furthermore, there's a challenge in providing effective recommendations to users in physical stores by leveraging their past purchase history and preference data. Additionally, there are no systems that can generate virtual sales representatives tailored to user preferences and provide interactive product suggestions within the store. This results in users being unable to fully develop their purchasing intent and making optimal product selection difficult.

[0584] 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.

[0585] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model. This makes it possible to provide users with a personalized customer service experience even in a physical store and increase their willingness to purchase.

[0586] "Customer service style" refers to the method and attitude of interaction that users prefer.

[0587] "Purchase history" refers to information that records a list of products a user has purchased in the past.

[0588] "Preference data" refers to information that reflects a user's preferences and interests.

[0589] A "virtual customer service representative" refers to a customer service representative created virtually using generative AI, who interacts with the user on screen.

[0590] "Interaction" refers to a two-way exchange between the user and the system.

[0591] "Real-time" refers to responding immediately to user input or actions.

[0592] "Talk information" refers to information that includes the content of conversations with users and the suggestions they made.

[0593] A "physical store" refers to a retail store that provides goods and services in a physical location.

[0594] A "smartphone application" refers to a software program that runs on a smartphone.

[0595] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate text and responses.

[0596] A "prompt message" refers to an instruction message that is input into a generative AI model.

[0597] The system of the present invention includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model.

[0598] Hardware and software

[0599] server

[0600] The server includes a database (e.g., MySQL or PostgreSQL) to store user purchase history and preference data. It also provides computing resources to generate responses from virtual customer service representatives using generative AI models (e.g., GPT-3 or the latest ChatGPT technology). Furthermore, it generates prompt messages and sends the generated virtual customer service representative data to the terminal.

[0601] terminal

[0602] The terminal is a mobile device such as a smartphone or tablet, and interaction between the user and a virtual customer service representative is realized through a dedicated application. The application receives user input and sends it to a server. Based on the data received from the server, the virtual customer service representative is displayed, and real-time interaction takes place.

[0603] System Operation Overview

[0604] User's preference input

[0605] Users enter their preferred customer service style using a smartphone application. For example, they might enter "casual and friendly customer service style."

[0606] Data collection

[0607] The server collects past purchase history and preference data from the database based on the user's ID. This data includes items the user has purchased, browsing history, and favorite categories.

[0608] Generation of virtual customer service representatives

[0609] Based on the collected data and user preferences, the server generates prompt messages, which are then input into a generation AI model to create a virtual customer service representative. Examples of prompt messages are as follows:

[0610] User preference: Casual and friendly customer service style

[0611] User data: Past purchase history: running shoes, sportswear. Favorite category: sporting goods.

[0612] Interaction

[0613] The generated virtual customer service representative data is sent to the terminal and displayed within the smartphone application. The virtual customer service representative begins by saying something like, "Hello! What products are you looking for today?" and interacts with the user.

[0614] Providing real-time answers and talk information.

[0615] When a user asks a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server searches its database for appropriate product information and sends the generated answer back to the terminal. Effective sales talk information, such as limited-time offers and cross-selling suggestions, is also provided in real time.

[0616] This makes it possible to provide users with a personalized customer service experience through interaction with virtual sales representatives, thereby increasing their willingness to purchase.

[0617] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0618] Step 1:

[0619] Users input their desired customer service style into a smartphone application. Specifically, they might enter "casual and friendly customer service style" and submit that information. This input data is then sent from the device to the server.

[0620] Input: User's preferred customer service style (e.g., "Casual and friendly customer service style")

[0621] Output: User's requested input data sent to the server

[0622] Step 2:

[0623] The server uses the user's ID to collect past purchase history and preference data from the database. Specifically, it retrieves information such as products the user has purchased, browsing history, and favorite categories from the database. This collected data is used in subsequent processes.

[0624] Input: User ID

[0625] Output: User's past purchase history and preference data

[0626] Step 3:

[0627] The server generates prompt sentences suitable for the generating AI model based on the collected user data. These prompt sentences include the user's desired customer service style and user data. This is then input into the generating AI model to create a virtual customer service representative.

[0628] Input: User's desired input data, past purchase history, and preference data.

[0629] Output: Prompt text to input into the generating AI model (Example: "User preference: Casual and friendly customer service style. User data: Past purchase history: Running shoes, sportswear. Favorite category: Sports goods.")

[0630] Step 4:

[0631] The server uses a generation AI model to generate a virtual customer service representative based on the prompt text. It then sends the generated virtual customer service representative's text data, talk script, and character information to the terminal.

[0632] Input: Prompt message

[0633] Output: Generated virtual customer service representative data (text data, talk script, character information)

[0634] Step 5:

[0635] The terminal displays the virtual customer service representative's data received from the server. Interaction with the user begins, and the virtual customer service representative starts speaking to the user, saying things like, "Hello! What kind of products are you looking for today?"

[0636] Input: Data of the generated virtual customer service representative

[0637] Output: Virtual customer service representative displayed on the screen

[0638] Step 6:

[0639] The user enters a question into the virtual customer service representative. For example, they might type, "What running shoes would you recommend?" This question is then sent from the terminal to the server.

[0640] Input: User's question (e.g., "What running shoes do you recommend?")

[0641] Output: User questions sent to the server

[0642] Step 7:

[0643] The server analyzes the user's question and searches the database for appropriate product information. Specifically, it searches for recommended running shoe models and features, and generates answer data.

[0644] Input: User's question

[0645] Output: Response data generated based on appropriate product information

[0646] Step 8:

[0647] The server sends the generated response data to the terminal. This data includes specific suggestions, such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0648] Input: Generated response data

[0649] Output: Sending response data to the terminal

[0650] Step 9:

[0651] The terminal displays the received response data to the user. Suggestions from a virtual customer service representative are displayed on the screen, increasing the user's purchasing intent.

[0652] Input: Response data sent from the server

[0653] Output: Response data displayed on the user screen

[0654] Step 10:

[0655] The server generates effective sales pitches, exclusive offers, and cross-selling suggestions in real time and incorporates them into the virtual sales representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also show you socks that would go perfectly with them."

[0656] Input: User's current interaction status, reward information

[0657] Output: Generation of talk data incorporating bonus information

[0658] Step 11:

[0659] The terminal displays updated virtual customer service representatives' conversation information in real time, providing users with effective conversation and special offer information. This can further increase users' purchasing intent.

[0660] Input: Talk data sent from the server

[0661] Output: Talk information from the virtual customer service representative displayed on the screen in real time.

[0662] 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.

[0663] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[0664] System Processing Flow Overview

[0665] User's preference input

[0666] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a selection form, such as "casual and friendly customer service style," and presses the submit button. This information is sent from the device to the server.

[0667] Consumer data collection

[0668] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0669] Virtual Customer Service Representative Generation

[0670] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0671] Virtual customer service representative display and interaction

[0672] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[0673] User emotion recognition

[0674] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. For example, when a user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[0675] Adjusting interactions

[0676] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[0677] User interaction

[0678] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[0679] Generating answers to questions

[0680] The server generates responses based on the collected product information. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated response data is sent to the terminal, which then displays it to the user.

[0681] Providing an effective talk

[0682] The server generates effective sales talk information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0683] Customer service scenarios with celebrities (optional feature)

[0684] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[0685] Specific example

[0686] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[0687] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[0688] During this time, the emotion engine analyzes the user's facial expressions and tone of voice, and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time and look around."

[0689] In this way, the system of the present invention provides a personalized customer service experience that meets the user's needs and further enables responses that take into account the user's emotions in real time, thereby increasing purchasing intent and providing a highly satisfying purchasing experience.

[0690] The following describes the processing flow.

[0691] Step 1:

[0692] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[0693] Step 2:

[0694] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[0695] Step 3:

[0696] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[0697] Step 4:

[0698] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[0699] Step 5:

[0700] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0701] Step 6:

[0702] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[0703] Step 7:

[0704] The device uses its built-in camera and microphone to activate an emotion engine that analyzes the user's facial expressions and voice tone in real time. The emotion engine detects the user's emotional state from their facial expressions and voice tone.

[0705] Step 8:

[0706] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[0707] Step 9:

[0708] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[0709] Step 10:

[0710] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0711] Step 11:

[0712] The server sends the generated response data to the terminal. The terminal displays the response to the user. For example, product features and price information may be displayed.

[0713] Step 12:

[0714] The emotion engine analyzes the user's facial expressions and voice tone to determine whether the user is calm, tired, excited, etc. The emotion engine then sends the identified emotional state to the server.

[0715] Step 13:

[0716] The server adjusts the virtual customer service representative's response and tone based on the identified emotional state. For example, if the server determines that the user is tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time browsing."

[0717] Step 14:

[0718] The server generates effective sales pitches in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal for incorporation into the virtual sales representative. This allows the virtual sales representative to provide special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0719] Step 15:

[0720] The device displays exclusive offers and cross-sell suggestions to the user via a virtual sales representative. If the user is interested in the information, more detailed information is provided to increase their purchase intent.

[0721] In this way, the system of the present invention provides a personalized customer service experience based on the user's wishes and real-time emotional state, thereby increasing the user's desire to purchase. Furthermore, by providing effective conversational information and adjusting responses based on emotion recognition, it realizes a highly satisfying purchasing experience.

[0722] (Example 2)

[0723] 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".

[0724] In modern online shopping, users demand personalized customer service experiences, but current systems fail to meet these demands. In particular, there is a lack of customized customer service styles tailored to user preferences and in-depth responses that consider user emotional states. Furthermore, the creation of virtual customer service representatives using specific person models, as well as real-time limited-time offers and cross-selling suggestions, are not yet realized. A system is needed to address these challenges and provide users with a highly satisfying purchasing experience.

[0725] 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.

[0726] In this invention, the server includes means for inputting the user's desired customer service style, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for displaying the generated virtual customer service representative on the user terminal and allowing interaction, means for recognizing the user's emotional state using an emotion engine, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotional state, means for searching for appropriate product information from a database based on the user's questions and generating answers, and means for providing effective talk information in real time. This enables a personalized customer service experience that meets the user's needs and responses that respond to their emotional state, and also enables the provision of limited offers and cross-selling suggestions in real time.

[0727] A "user" refers to an individual who uses the system to input their preferred customer service style or interact with a virtual customer service representative.

[0728] A "terminal" is a device used by a user to access a system and input or receive information, and includes smartphones, tablets, and personal computers.

[0729] A "server" is a central control unit that collects, processes, and generates data based on user input.

[0730] "Customer service style" refers to the communication characteristics and attitudes of the virtual customer service representative that the user desires.

[0731] "User data" refers to a collection of data related to a user, including their past purchase history and preference data.

[0732] A "virtual customer service representative" refers to a virtual character that is generated based on the user's preferences and data, and that interacts with the user online.

[0733] "Interaction" refers to two-way communication that takes place between the user and the virtual customer service representative.

[0734] An "emotion engine" refers to a technology that analyzes a user's facial expressions, voice tone, and input text to identify the user's emotional state.

[0735] "Emotional state" refers to the user's psychological and emotional condition, and includes things like being tired or excited.

[0736] "Response" refers to the reply or reaction that a virtual customer service representative gives to a user's questions or requests.

[0737] "Tone" refers to the manner in which a virtual customer service representative speaks and their overall demeanor.

[0738] A "database" refers to an information management system used to store user data, product information, and other similar data.

[0739] "Product information" refers to detailed data about products stored within the system, including features, price, and stock status.

[0740] "Talk information" refers to the content and flow of conversations used by virtual customer service representatives in their interactions with users.

[0741] A "limited offer" refers to a discount or benefit that is offered exclusively for the purchase of a specific product.

[0742] "Cross-selling suggestions" refer to suggestions for additional products related to the product a user is considering purchasing.

[0743] A "character model" refers to a collection of data used to generate virtual customer service representatives based on the characteristics of specific celebrities or characters.

[0744] The customer service system of this invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[0745] Users input their preferred customer service style using devices such as smartphones, tablets, or personal computers. For example, a user might enter "casual and friendly customer service style" into a selection form and press the submit button. This information is then sent from the device to the server.

[0746] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0747] The server generates virtual customer service representatives using a generation AI (e.g., the latest chatGPT technology). During this process, the server inputs prompt text into the generation AI, which then generates text data, talk scripts, and character information for the virtual customer service representative. This data is then sent to the terminal.

[0748] The terminal displays a virtual sales representative on the screen and begins interacting with the user. For example, the virtual sales representative might start by saying, "Hello! What kind of product are you looking for today?"

[0749] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. When the user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[0750] The server receives emotional data sent from the terminal and adjusts the virtual customer service representative's response and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[0751] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (for example, recommended running shoe models and their features). Based on the collected product information, the server generates an answer. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated answer data is sent to the terminal, which then displays it to the user.

[0752] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0753] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the terminal. The terminal displays this virtual customer service representative and begins interacting with the user.

[0754] Specific example

[0755] The user selects a "casual and friendly customer service style" in the smartphone app and presses the send button. The device sends this information to the server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a casual, sports-savvy virtual customer service representative and sends that data to the device. The device displays the virtual customer service representative and begins interacting with the user. When the user asks, "What running shoes do you recommend?", the device sends the question to the server, which generates an answer based on appropriate product information and sends it to the device. The device displays the generated answer to the user and also displays more effective talk and limited offers to increase the user's desire to purchase. During this time, an emotion engine analyzes the user's facial expressions and voice tone and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying, "You seem tired today. Please take your time browsing."

[0756] Example of a prompt

[0757] "Specify the user's preferred customer service style as 'casual and friendly,' and generate text recommending running shoes based on their past purchase history and preference data."

[0758] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0759] Step 1:

[0760] The user enters their preferred customer service style.

[0761] The user opens the app on their device (e.g., smartphone, tablet, or computer) and enters their preferred customer service style.

[0762] The user enters "Casual and friendly customer service style" into the customer service style selection form and presses the submit button.

[0763] Input: User's preferred customer service style

[0764] Output: Preferred customer service style information is sent from the terminal to the server.

[0765] Step 2:

[0766] User data collection

[0767] The server accesses the database based on the user's ID.

[0768] The server collects the user's past purchase history and preference data. This data includes items the user has previously purchased, browsing history, and favorite categories.

[0769] The server filters the collected data and creates a dataset that matches the user's requirements.

[0770] Input: User ID, preferred customer service style

[0771] Output: Filtered historical purchase history and preference data

[0772] Step 3:

[0773] Generation of virtual customer service representatives

[0774] The server generates prompt messages based on the collected user data and user requests.

[0775] The server inputs prompt text into a generating AI (e.g., the latest chatGPT technology) to create a virtual customer service representative.

[0776] The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal.

[0777] Input: Filtered user data, preferred customer service style information

[0778] Output: Text data, talk scripts, and character information of the virtual customer service representative.

[0779] Step 4:

[0780] Display of the virtual customer service representative and initiation of interaction.

[0781] The device displays a virtual customer service representative on the screen.

[0782] The virtual customer service representative begins by saying, "Hello! What kind of product are you looking for today?"

[0783] Input: Text data, talk script, and character information of the virtual customer service representative.

[0784] Output: A virtual customer service representative is displayed on the user's terminal, and the interaction begins.

[0785] Step 5:

[0786] User emotion recognition

[0787] The device has a built-in emotion engine that analyzes the user's facial expressions, voice tone, and input text.

[0788] When a user is visible through the camera, the device uses facial recognition technology to identify their emotional state.

[0789] When a user speaks, the device performs voice tone analysis to understand emotional nuances.

[0790] Input: User facial expression data, voice tone data, input text

[0791] Output: The emotional state of the identified user.

[0792] Step 6:

[0793] Response and tone adjustment

[0794] The server receives emotion data sent from the terminal.

[0795] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak in a gentler tone.

[0796] Input: Identified user's emotional state, virtual customer service representative's text data

[0797] Output: Adjusted virtual customer service representative responses and tone

[0798] Step 7:

[0799] Handling user inquiries

[0800] The user enters the question, "What running shoes do you recommend?"

[0801] The terminal sends the question to the server.

[0802] Input: User's question

[0803] Output: Question data sent to the server

[0804] Step 8:

[0805] Generating answers to questions

[0806] The server analyzes the user's question.

[0807] The server searches the database for appropriate product information (e.g., recommended running shoe models and their features).

[0808] The server generates a response based on the collected product information.

[0809] Input: User's question, product information in the database

[0810] Output: Generated response data

[0811] Step 9:

[0812] Submit and display of responses

[0813] The server sends the generated response data to the terminal.

[0814] The device displays the generated response to the user.

[0815] Input: Generated response data

[0816] Output: Answer displayed on the user's terminal

[0817] Step 10:

[0818] Providing effective communication information

[0819] The server generates effective, real-time sales information (e.g., limited-time offers and cross-selling suggestions).

[0820] The server sends data to the terminal to be incorporated into the virtual customer service representative.

[0821] A virtual customer service representative provides users with special offers such as, "These shoes are 20% off right now! We can also recommend socks that would go perfectly with these shoes."

[0822] Input: Real-time conversation information, text data from virtual customer service representatives

[0823] Output: Virtual customer service representative speech with effective conversational information incorporated.

[0824] Step 11:

[0825] Customer service scenarios with celebrities (optional feature)

[0826] When a user selects a customer service style, they can choose a specific person model (for example, a celebrity).

[0827] The server collects data from the selected person model (e.g., voice data, feature data).

[0828] The server generates a virtual customer service representative based on a specific person model and sends it to the terminal.

[0829] The terminal displays this virtual customer service representative and begins interacting with the user.

[0830] Input: User-selected person model, data for the selected person model

[0831] Output: Data and display of a virtual customer service representative generated based on a specific person model.

[0832] Through these steps, the system provides a personalized customer service experience tailored to the user's needs, and further enhances purchasing intent and delivers a highly satisfying shopping experience by enabling real-time responses that take the user's emotions into account.

[0833] (Application Example 2)

[0834] 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."

[0835] Traditional in-store customer service systems faced challenges in providing personalized service tailored to the individual needs and emotions of each user. Furthermore, the inability to make effective real-time suggestions and cross-selling proposals limited the ability to improve customer satisfaction and boost sales. Additionally, there was a lack of customer service systems that could leverage new technologies such as smart glasses.

[0836] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for analyzing the user's facial expressions and voice tone to recognize emotions, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotions, and means for interacting with the virtual customer service representative by wearing smart glasses in a physical store. This enables a personalized customer service experience that responds to the user's needs and emotions, thereby improving customer satisfaction and promoting sales.

[0837] A "virtual customer service representative" is a virtual customer service representative generated based on the user's preferences and collected data.

[0838] "Interaction" refers to two-way communication between the user and the virtual customer service representative.

[0839] "User's past purchase history" refers to a list of products and historical data that the user has purchased so far.

[0840] "Preference data" refers to data related to a user's preferences and interests, and is primarily based on past behavior and preferences.

[0841] "Facial recognition" is a technology that analyzes a user's facial expressions to identify their emotional state at that time.

[0842] "Voice tone" is a technique that identifies a user's emotional state by analyzing the nuances of their voice, including its pitch and volume.

[0843] "Smart glasses" are glasses-type devices that have the function of displaying information from the real world as augmented reality (AR).

[0844] "Real-time response" refers to a response that provides appropriate information immediately in response to a user's question.

[0845] "Effective sales information" refers to information generated to promote sales and make suggestions, tailored to the user's interests and needs.

[0846] "Cross-selling" is a sales strategy that suggests related products that users might be interested in.

[0847] The system program for implementing this invention operates in cooperation between a server and a terminal and is designed to provide users with a personalized customer service experience. This system includes the following main components:

[0848] Component and Processing Overview

[0849] server

[0850] The server is primarily responsible for data collection, execution of generative AI models, and response generation.

[0851] 1. Collection of user data:

[0852] The system collects the user's past purchase history and preference data from a database. This includes a list of products the user has purchased, their browsing history, and their preferences.

[0853] 2. Virtual customer service representative generation:

[0854] Based on the collected data and the user's preferred customer service style, a virtual customer service representative is generated using a generative AI model (e.g., the T5 model). This generation process involves inputting appropriate prompts to the generative AI model and obtaining character information for the virtual customer service representative.

[0855] 3. Real-time response to questions:

[0856] The system analyzes user inquiries and generates appropriate product information and answers. The generated information is then sent to the device.

[0857] terminal

[0858] The device is responsible for user interaction and presents data in real time.

[0859] 1. Emotion recognition:

[0860] The device is equipped with a camera and a voice input device, and recognizes emotions by analyzing the user's facial expressions and voice tone. This analysis utilizes facial recognition technology such as the DeepFace library and voice tone analysis technology.

[0861] 2. Performing an interaction:

[0862] The terminal displays a generated virtual customer service representative and interacts with the user. For example, it provides an environment where the user can converse with a virtual customer service representative through smart glasses.

[0863] 3. Adjusting the response:

[0864] Based on recognized emotion data, the system receives instructions from the server and adjusts the virtual customer service representative's responses and tone. For example, if the user is tired, it will respond in a gentle tone.

[0865] Specific example

[0866] Let's assume the user is wearing smart glasses in a physical store. These smart glasses are equipped with a camera and microphone, and can analyze the user's facial expressions and voice tone in real time.

[0867] When a user requests a "casual and friendly customer service style," the device sends this information to the server. The server collects the user's past purchase history and preference data, and uses a generative AI model to generate an appropriate virtual customer service representative. The generated virtual representative will engage in friendly conversation, for example, saying, "Hello! What products are you looking for today?"

[0868] An example of a prompt is, "Create a virtual assistant with sports and casual preferences and a casual and friendly interaction style." This prompt prompt causes the AI ​​model to generate a virtual customer service representative that aligns with the user's preferences.

[0869] In this way, personalized customer service experiences tailored to user needs and emotions can be provided. This system is highly effective in improving customer satisfaction and boosting sales.

[0870] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0871] Step 1:

[0872] The user inputs their desired customer service style through a device such as smart glasses or a smartphone. The entered information, along with the user's ID, is sent to the server. The input data is in text format and represents the user's preferences.

[0873] Step 2:

[0874] Based on the received user preference data, the server collects the user's past purchase history and preference data from the database. Database queries are used to retrieve a list of products the user has previously purchased, their browsing history, and their preferences. The retrieved data is temporarily stored as a user dataset.

[0875] Step 3:

[0876] The server inputs a prompt message into a generative AI model based on the collected user dataset, generating a virtual customer service representative. The prompt message is "Create a virtual assistant with sports and casual preferences and a casual and friendly interaction style." and the text data output from the generative model (e.g., T5 model) is retrieved. The output text data contains the character information of the virtual customer service representative.

[0877] Step 4:

[0878] The device initiates interaction with the user based on the character information of the virtual customer service representative received from the server. The virtual customer service representative is displayed on the smart glasses' screen and speaks to the user, saying things like, "Hello! What kind of product are you looking for today?" During this process, the displayed text and audio data are generated as interaction data.

[0879] Step 5:

[0880] The device uses its built-in camera and microphone to analyze the user's facial expressions and voice tone, and uses an emotion recognition engine to identify the user's emotions. It analyzes facial expression data using the DeepFace library and voice tone analysis technology to analyze voice data. Detected emotions are stored as emotion data.

[0881] Step 6:

[0882] The server receives emotional data and runs an algorithm that adjusts the virtual customer service representative's responses and tone based on it. For example, if the emotional data identifies the user as "tired," the virtual customer service representative's response will be changed to a gentler tone. This provides the user with a pleasant customer service experience.

[0883] Step 7:

[0884] When a user enters a specific question, the device sends that question to the server. The server analyzes the question and searches its database for appropriate product information. The search results are generated as answer data and sent back to the device.

[0885] Step 8:

[0886] The terminal displays response data received from the server to the user and provides answers through a virtual customer service representative. For example, it might present information such as, "These shoes are lightweight and highly breathable." In this process, the displayed text and audio data are generated as response data.

[0887] Step 9:

[0888] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and sends it to the terminal. The terminal then incorporates this information into the virtual sales representative and presents it to the user. This allows the server to provide users with special offers such as, "These shoes are 20% off right now."

[0889] 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.

[0890] 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.

[0891] 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.

[0892] [Third Embodiment]

[0893] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0894] 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.

[0895] 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).

[0896] 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.

[0897] 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.

[0898] 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).

[0899] 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.

[0900] 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.

[0901] 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.

[0902] 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.

[0903] 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.

[0904] 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".

[0905] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[0906] System Processing Flow Overview

[0907] User's preference input

[0908] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a customer service style selection form, such as "casual and friendly customer service style." This information is sent from the device to the server.

[0909] Consumer data collection

[0910] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[0911] Virtual Customer Service Representative Generation

[0912] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0913] Virtual customer service representative display and interaction

[0914] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[0915] User interaction

[0916] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[0917] Generating answers to questions

[0918] The server generates responses based on the collected product information. The generated response data is sent to the terminal, which then displays it to the user. For example, specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes offers excellent cushioning and is ideal for long-distance running," might be displayed.

[0919] Providing an effective talk

[0920] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal displays this information to the user, offering special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[0921] Customer service scenarios with celebrities (optional feature)

[0922] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[0923] Specific example

[0924] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[0925] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[0926] In this way, the system of the present invention provides a practical solution for enhancing purchasing intent by offering a personalized customer service experience that meets the user's needs.

[0927] The following describes the processing flow.

[0928] Step 1:

[0929] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[0930] Step 2:

[0931] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[0932] Step 3:

[0933] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[0934] Step 4:

[0935] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[0936] Step 5:

[0937] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[0938] Step 6:

[0939] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[0940] Step 7:

[0941] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[0942] Step 8:

[0943] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[0944] Step 9:

[0945] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0946] Step 10:

[0947] The server sends the generated response data to the terminal. The terminal displays the response to the user.

[0948] Step 11:

[0949] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual customer service representative.

[0950] Step 12:

[0951] The terminal displays exclusive offer information to the user via a virtual customer service representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also recommend socks that go perfectly with these shoes."

[0952] Through these steps, users will be able to have an ideal customer service experience, which will likely increase their desire to purchase.

[0953] (Example 1)

[0954] 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."

[0955] Traditional online shopping systems required users to spend a lot of time finding the products they wanted to buy, and lacked personalized customer service experiences tailored to individual needs. Furthermore, it was difficult for users with specific interests or styles to receive appropriate advice and suggestions. This could lead to decreased purchasing intent and missed sales opportunities. Additionally, it was challenging to adequately accommodate users who desired customer service representatives based on specific celebrities or characters.

[0956] 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.

[0957] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for transmitting the generated virtual customer service representative's text data, talk script, and character information to the terminal, means for displaying the virtual customer service representative, means for transmitting the user's questions from the terminal to the server, means for searching for appropriate product information from a database based on the user's questions, and means for generating answers from a generated AI model based on the collected product information. This makes it possible to provide a personalized customer service experience that matches the user's wishes and interests, thereby increasing their willingness to purchase. Furthermore, it is possible to generate virtual customer service representatives based on specific celebrities or characters and provide personalized responses to users.

[0958] A "user" refers to a person who uses the system to search for and purchase goods and services.

[0959] "Device" refers to electronic devices used by users, such as smartphones, tablets, and personal computers.

[0960] A "server" refers to a central system that processes data, stores information, and handles requests from terminals.

[0961] "Customer service style" refers to the way a virtual customer service representative responds and their attitude, as chosen by the user.

[0962] "Purchase history" refers to a record of products that a user has purchased through the system so far.

[0963] "Preference data" refers to data that includes information about users' preferences and interests.

[0964] A "virtual customer service representative" refers to a virtual character or assistant created using generative AI to provide customer service to users.

[0965] "Interaction" refers to the dialogue and exchange of information that takes place between the user and the virtual customer service representative.

[0966] A "generative AI model" refers to an artificial intelligence model that generates text or responses based on user input or data.

[0967] A "talk script" refers to a script or pattern for a virtual customer service representative to use when speaking to a user.

[0968] "Character information" refers to data about the appearance, personality, and speaking style of the virtual customer service representative.

[0969] A "prompt statement" refers to a command statement that is input to a generating AI model.

[0970] A "database" refers to an entire system used for collecting, storing, and retrieving data.

[0971] "Real-time" refers to processing or responding to user input almost simultaneously.

[0972] A "specific person model" refers to a virtual customer service representative generated based on data of a celebrity or character selected by the user.

[0973] "Limited-time offers" refer to information about special products or discounts that are available only under specific conditions.

[0974] "Cross-selling" refers to a marketing technique where related products are suggested simultaneously when a customer purchases a particular product.

[0975] Modes for carrying out the invention

[0976] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[0977] Hardware and software to be used

[0978] Devices: Smartphones, tablets, PCs, etc.

[0979] Server: A central system that processes data, stores information, and handles requests from each terminal.

[0980] Database: Stores users' past purchase history and preference data (e.g., MySQL).

[0981] Generative AI models: AI models for generating natural language (e.g., OpenAI GPT-4)

[0982] UI framework: A framework for displaying content on a device (e.g., React Native)

[0983] Operating procedures and specific examples

[0984] Users input their preferred customer service style using a device such as a smartphone or computer. For example, they might enter "casual and friendly customer service style" and press the send button. The device then sends this information to the server.

[0985] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server's process begins with executing a SELECT query to extract all data related to the user ID. Next, edge analytics is used to filter the collected data and create a data set that matches the user's requests.

[0986] The server generates a virtual customer service representative using a generative AI (for example, OpenAI's GPT-4 model) based on the collected data and user requests. A prompt is used during the generation process. An example of a prompt might be, "Generate a virtual customer service representative with a casual customer service style based on the user's preferences." The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal in JSON format.

[0987] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. Using a UI framework such as React Native, the virtual customer service representative's avatar is displayed on the screen and greets the user with "Hello! What products are you looking for today?"

[0988] When a user enters the question "What running shoes do you recommend?", the device sends this question to the server. The device receives the user's input and sends it to the server as a POST request in JSON format.

[0989] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to collect the relevant product information. After that, it uses a generative AI model to create an answer. The generated answer data is sent to the terminal in JSON format.

[0990] The device analyzes the received data and displays a response to the user. For example, it can make specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[0991] Furthermore, the server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal receives this information, and the sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that would go perfectly with these shoes."

[0992] Furthermore, if a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the device, which then displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?"

[0993] The above describes embodiments of the present invention, which are detailed procedures for providing a personalized customer service experience tailored to user needs and increasing purchasing intent.

[0994] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0995] Step 1: User input

[0996] The user uses their device to input their preferred service style. Specifically, they launch an application on their device and enter a service style selection form, such as "casual and friendly service style." This action causes the device to send the user's input data to the server. An HTTP POST request is used for this transmission. The input data, "casual and friendly service style," is sent to the server.

[0997] Step 2: Consumer Data Collection

[0998] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server first executes a SELECT query to extract all data related to the user ID. This data includes items the user has purchased, browsing history, and favorite categories. The extracted data is then filtered using edge analytics to create a data set that matches the user's requirements. The input is the user ID, and the output is the filtered data set.

[0999] Step 3: Generate a virtual customer service representative

[1000] The server generates a virtual customer service representative using a generative AI (e.g., OpenAI's GPT-4 model) based on the collected data and user requests. In this process, a prompt is input to the generative AI model. For example, "Generate a virtual customer service representative with a casual service style based on the user's preferences" might be the prompt. The AI ​​model analyzes this prompt, generates text data, talk scripts, and character information for a virtual customer service representative that meets the user's preferences, and sends it to the terminal in JSON format. The input is a filtered data set and a prompt, and the output is virtual customer service representative data in JSON format.

[1001] Step 4: Virtual customer service representative display and interaction

[1002] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. A UI framework (e.g., React Native) is used for this. The terminal screen displays an avatar of the virtual customer service representative, who says to the user, "Hello! What products are you looking for today?" The input is virtual customer service representative data in JSON format, and the output is the virtual customer service representative displayed on the screen.

[1003] Step 5: Interaction with the user

[1004] When a user enters the question "What running shoes do you recommend?", the terminal sends this question to the server. The terminal receives the user's input and sends it to the server as a POST request in JSON format. The input is the user's question, and the output is the request to the server.

[1005] Step 6: Generate answers to the questions

[1006] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to extract the relevant product information. After that, it uses a generative AI model to create an answer and sends the generated answer data to the terminal in JSON format. The input is the user's question, and the output is the answer data in JSON format.

[1007] Step 7: Displaying the response to the user

[1008] The device parses the received JSON data and displays a response to the user. For example, it might display specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The input is response data in JSON format, and the output is the response displayed on the screen.

[1009] Step 8: Delivering an effective talk

[1010] The server generates effective chat information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The generated chat information is sent from the server to the terminal. The terminal receives this information, and the virtual sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes." The input is the generated chat information, and the output is the chat information displayed on the screen.

[1011] Step 9: Customer service scenario with a celebrity (optional feature)

[1012] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the terminal, which displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?" The input is the data of the selected person model, and the output is the virtual customer service representative displayed on the screen.

[1013] (Application Example 1)

[1014] 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."

[1015] Current customer service systems offer limited personalized service experiences. Furthermore, there's a challenge in providing effective recommendations to users in physical stores by leveraging their past purchase history and preference data. Additionally, there are no systems that can generate virtual sales representatives tailored to user preferences and provide interactive product suggestions within the store. This results in users being unable to fully develop their purchasing intent and making optimal product selection difficult.

[1016] 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.

[1017] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model. This makes it possible to provide users with a personalized customer service experience even in a physical store and increase their willingness to purchase.

[1018] "Customer service style" refers to the method and attitude of interaction that users prefer.

[1019] "Purchase history" refers to information that records a list of products a user has purchased in the past.

[1020] "Preference data" refers to information that reflects a user's preferences and interests.

[1021] A "virtual customer service representative" refers to a customer service representative created virtually using generative AI, who interacts with the user on screen.

[1022] "Interaction" refers to a two-way exchange between the user and the system.

[1023] "Real-time" refers to responding immediately to user input or actions.

[1024] "Talk information" refers to information that includes the content of conversations with users and the suggestions they made.

[1025] A "physical store" refers to a retail store that provides goods and services in a physical location.

[1026] A "smartphone application" refers to a software program that runs on a smartphone.

[1027] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate text and responses.

[1028] A "prompt message" refers to an instruction message that is input into a generative AI model.

[1029] The system of the present invention includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model.

[1030] Hardware and software

[1031] server

[1032] The server includes a database (e.g., MySQL or PostgreSQL) to store user purchase history and preference data. It also provides computing resources to generate responses from virtual customer service representatives using generative AI models (e.g., GPT-3 or the latest ChatGPT technology). Furthermore, it generates prompt messages and sends the generated virtual customer service representative data to the terminal.

[1033] terminal

[1034] The terminal is a mobile device such as a smartphone or tablet, and interaction between the user and a virtual customer service representative is realized through a dedicated application. The application receives user input and sends it to a server. Based on the data received from the server, the virtual customer service representative is displayed, and real-time interaction takes place.

[1035] System Operation Overview

[1036] User's preference input

[1037] Users enter their preferred customer service style using a smartphone application. For example, they might enter "casual and friendly customer service style."

[1038] Data collection

[1039] The server collects past purchase history and preference data from the database based on the user's ID. This data includes items the user has purchased, browsing history, and favorite categories.

[1040] Generation of virtual customer service representatives

[1041] Based on the collected data and user preferences, the server generates prompt messages, which are then input into a generation AI model to create a virtual customer service representative. Examples of prompt messages are as follows:

[1042] User preference: Casual and friendly customer service style

[1043] User data: Past purchase history: running shoes, sportswear. Favorite category: sporting goods.

[1044] Interaction

[1045] The generated virtual customer service representative data is sent to the terminal and displayed within the smartphone application. The virtual customer service representative begins by saying something like, "Hello! What products are you looking for today?" and interacts with the user.

[1046] Providing real-time answers and talk information.

[1047] When a user asks a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server searches its database for appropriate product information and sends the generated answer back to the terminal. Effective sales talk information, such as limited-time offers and cross-selling suggestions, is also provided in real time.

[1048] This makes it possible to provide users with a personalized customer service experience through interaction with virtual sales representatives, thereby increasing their willingness to purchase.

[1049] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1050] Step 1:

[1051] Users input their desired customer service style into a smartphone application. Specifically, they might enter "casual and friendly customer service style" and submit that information. This input data is then sent from the device to the server.

[1052] Input: User's preferred customer service style (e.g., "Casual and friendly customer service style")

[1053] Output: User's requested input data sent to the server

[1054] Step 2:

[1055] The server uses the user's ID to collect past purchase history and preference data from the database. Specifically, it retrieves information such as products the user has purchased, browsing history, and favorite categories from the database. This collected data is used in subsequent processes.

[1056] Input: User ID

[1057] Output: User's past purchase history and preference data

[1058] Step 3:

[1059] The server generates prompt sentences suitable for the generating AI model based on the collected user data. These prompt sentences include the user's desired customer service style and user data. This is then input into the generating AI model to create a virtual customer service representative.

[1060] Input: User's desired input data, past purchase history, and preference data.

[1061] Output: Prompt text to input into the generating AI model (Example: "User preference: Casual and friendly customer service style. User data: Past purchase history: Running shoes, sportswear. Favorite category: Sports goods.")

[1062] Step 4:

[1063] The server uses a generation AI model to generate a virtual customer service representative based on the prompt text. It then sends the generated virtual customer service representative's text data, talk script, and character information to the terminal.

[1064] Input: Prompt message

[1065] Output: Generated virtual customer service representative data (text data, talk script, character information)

[1066] Step 5:

[1067] The terminal displays the virtual customer service representative's data received from the server. Interaction with the user begins, and the virtual customer service representative starts speaking to the user, saying things like, "Hello! What kind of products are you looking for today?"

[1068] Input: Data of the generated virtual customer service representative

[1069] Output: Virtual customer service representative displayed on the screen

[1070] Step 6:

[1071] The user enters a question into the virtual customer service representative. For example, they might type, "What running shoes would you recommend?" This question is then sent from the terminal to the server.

[1072] Input: User's question (e.g., "What running shoes do you recommend?")

[1073] Output: User questions sent to the server

[1074] Step 7:

[1075] The server analyzes the user's question and searches the database for appropriate product information. Specifically, it searches for recommended running shoe models and features, and generates answer data.

[1076] Input: User's question

[1077] Output: Response data generated based on appropriate product information

[1078] Step 8:

[1079] The server sends the generated response data to the terminal. This data includes specific suggestions, such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[1080] Input: Generated response data

[1081] Output: Sending response data to the terminal

[1082] Step 9:

[1083] The terminal displays the received response data to the user. Suggestions from a virtual customer service representative are displayed on the screen, increasing the user's purchasing intent.

[1084] Input: Response data sent from the server

[1085] Output: Response data displayed on the user screen

[1086] Step 10:

[1087] The server generates effective sales pitches, exclusive offers, and cross-selling suggestions in real time and incorporates them into the virtual sales representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also show you socks that would go perfectly with them."

[1088] Input: User's current interaction status, reward information

[1089] Output: Generation of talk data incorporating bonus information

[1090] Step 11:

[1091] The terminal displays updated virtual customer service representatives' conversation information in real time, providing users with effective conversation and special offer information. This can further increase users' purchasing intent.

[1092] Input: Talk data sent from the server

[1093] Output: Talk information from the virtual customer service representative displayed on the screen in real time.

[1094] 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.

[1095] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[1096] System Processing Flow Overview

[1097] User's preference input

[1098] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a selection form, such as "casual and friendly customer service style," and presses the submit button. This information is sent from the device to the server.

[1099] Consumer data collection

[1100] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[1101] Virtual Customer Service Representative Generation

[1102] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[1103] Virtual customer service representative display and interaction

[1104] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[1105] User emotion recognition

[1106] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. For example, when a user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[1107] Adjusting interactions

[1108] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[1109] User interaction

[1110] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[1111] Generating answers to questions

[1112] The server generates responses based on the collected product information. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated response data is sent to the terminal, which then displays it to the user.

[1113] Providing an effective talk

[1114] The server generates effective sales talk information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[1115] Customer service scenarios with celebrities (optional feature)

[1116] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[1117] Specific example

[1118] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[1119] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[1120] During this time, the emotion engine analyzes the user's facial expressions and tone of voice, and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time and look around."

[1121] In this way, the system of the present invention provides a personalized customer service experience that meets the user's needs and further enables responses that take into account the user's emotions in real time, thereby increasing purchasing intent and providing a highly satisfying purchasing experience.

[1122] The following describes the processing flow.

[1123] Step 1:

[1124] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[1125] Step 2:

[1126] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[1127] Step 3:

[1128] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[1129] Step 4:

[1130] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[1131] Step 5:

[1132] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[1133] Step 6:

[1134] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[1135] Step 7:

[1136] The device uses its built-in camera and microphone to activate an emotion engine that analyzes the user's facial expressions and voice tone in real time. The emotion engine detects the user's emotional state from their facial expressions and voice tone.

[1137] Step 8:

[1138] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[1139] Step 9:

[1140] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[1141] Step 10:

[1142] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[1143] Step 11:

[1144] The server sends the generated response data to the terminal. The terminal displays the response to the user. For example, product features and price information may be displayed.

[1145] Step 12:

[1146] The emotion engine analyzes the user's facial expressions and voice tone to determine whether the user is calm, tired, excited, etc. The emotion engine then sends the identified emotional state to the server.

[1147] Step 13:

[1148] The server adjusts the virtual customer service representative's response and tone based on the identified emotional state. For example, if the server determines that the user is tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time browsing."

[1149] Step 14:

[1150] The server generates effective sales pitches in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal for incorporation into the virtual sales representative. This allows the virtual sales representative to provide special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[1151] Step 15:

[1152] The device displays exclusive offers and cross-sell suggestions to the user via a virtual sales representative. If the user is interested in the information, more detailed information is provided to increase their purchase intent.

[1153] In this way, the system of the present invention provides a personalized customer service experience based on the user's wishes and real-time emotional state, thereby increasing the user's desire to purchase. Furthermore, by providing effective conversational information and adjusting responses based on emotion recognition, it realizes a highly satisfying purchasing experience.

[1154] (Example 2)

[1155] 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."

[1156] In modern online shopping, users demand personalized customer service experiences, but current systems fail to meet these demands. In particular, there is a lack of customized customer service styles tailored to user preferences and in-depth responses that consider user emotional states. Furthermore, the creation of virtual customer service representatives using specific person models, as well as real-time limited-time offers and cross-selling suggestions, are not yet realized. A system is needed to address these challenges and provide users with a highly satisfying purchasing experience.

[1157] 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.

[1158] In this invention, the server includes means for inputting the user's desired customer service style, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for displaying the generated virtual customer service representative on the user terminal and allowing interaction, means for recognizing the user's emotional state using an emotion engine, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotional state, means for searching for appropriate product information from a database based on the user's questions and generating answers, and means for providing effective talk information in real time. This enables a personalized customer service experience that meets the user's needs and responses that respond to their emotional state, and also enables the provision of limited offers and cross-selling suggestions in real time.

[1159] A "user" refers to an individual who uses the system to input their preferred customer service style or interact with a virtual customer service representative.

[1160] A "terminal" is a device used by a user to access a system and input or receive information, and includes smartphones, tablets, and personal computers.

[1161] A "server" is a central control unit that collects, processes, and generates data based on user input.

[1162] "Customer service style" refers to the communication characteristics and attitudes of the virtual customer service representative that the user desires.

[1163] "User data" refers to a collection of data related to a user, including their past purchase history and preference data.

[1164] A "virtual customer service representative" refers to a virtual character that is generated based on the user's preferences and data, and that interacts with the user online.

[1165] "Interaction" refers to two-way communication that takes place between the user and the virtual customer service representative.

[1166] An "emotion engine" refers to a technology that analyzes a user's facial expressions, voice tone, and input text to identify the user's emotional state.

[1167] "Emotional state" refers to the user's psychological and emotional condition, and includes things like being tired or excited.

[1168] "Response" refers to the reply or reaction that a virtual customer service representative gives to a user's questions or requests.

[1169] "Tone" refers to the manner in which a virtual customer service representative speaks and their overall demeanor.

[1170] A "database" refers to an information management system used to store user data, product information, and other similar data.

[1171] "Product information" refers to detailed data about products stored within the system, including features, price, and stock status.

[1172] "Talk information" refers to the content and flow of conversations used by virtual customer service representatives in their interactions with users.

[1173] A "limited offer" refers to a discount or benefit that is offered exclusively for the purchase of a specific product.

[1174] "Cross-selling suggestions" refer to suggestions for additional products related to the product a user is considering purchasing.

[1175] A "character model" refers to a collection of data used to generate virtual customer service representatives based on the characteristics of specific celebrities or characters.

[1176] The customer service system of this invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[1177] Users input their preferred customer service style using devices such as smartphones, tablets, or personal computers. For example, a user might enter "casual and friendly customer service style" into a selection form and press the submit button. This information is then sent from the device to the server.

[1178] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[1179] The server generates virtual customer service representatives using a generation AI (e.g., the latest chatGPT technology). During this process, the server inputs prompt text into the generation AI, which then generates text data, talk scripts, and character information for the virtual customer service representative. This data is then sent to the terminal.

[1180] The terminal displays a virtual sales representative on the screen and begins interacting with the user. For example, the virtual sales representative might start by saying, "Hello! What kind of product are you looking for today?"

[1181] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. When the user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[1182] The server receives emotional data sent from the terminal and adjusts the virtual customer service representative's response and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[1183] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (for example, recommended running shoe models and their features). Based on the collected product information, the server generates an answer. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated answer data is sent to the terminal, which then displays it to the user.

[1184] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[1185] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the terminal. The terminal displays this virtual customer service representative and begins interacting with the user.

[1186] Specific example

[1187] The user selects a "casual and friendly customer service style" in the smartphone app and presses the send button. The device sends this information to the server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a casual, sports-savvy virtual customer service representative and sends that data to the device. The device displays the virtual customer service representative and begins interacting with the user. When the user asks, "What running shoes do you recommend?", the device sends the question to the server, which generates an answer based on appropriate product information and sends it to the device. The device displays the generated answer to the user and also displays more effective talk and limited offers to increase the user's desire to purchase. During this time, an emotion engine analyzes the user's facial expressions and voice tone and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying, "You seem tired today. Please take your time browsing."

[1188] Example of a prompt

[1189] "Specify the user's preferred customer service style as 'casual and friendly,' and generate text recommending running shoes based on their past purchase history and preference data."

[1190] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1191] Step 1:

[1192] The user enters their preferred customer service style.

[1193] The user opens the app on their device (e.g., smartphone, tablet, or computer) and enters their preferred customer service style.

[1194] The user enters "Casual and friendly customer service style" into the customer service style selection form and presses the submit button.

[1195] Input: User's preferred customer service style

[1196] Output: Preferred customer service style information is sent from the terminal to the server.

[1197] Step 2:

[1198] User data collection

[1199] The server accesses the database based on the user's ID.

[1200] The server collects the user's past purchase history and preference data. This data includes items the user has previously purchased, browsing history, and favorite categories.

[1201] The server filters the collected data and creates a dataset that matches the user's requirements.

[1202] Input: User ID, preferred customer service style

[1203] Output: Filtered historical purchase history and preference data

[1204] Step 3:

[1205] Generation of virtual customer service representatives

[1206] The server generates prompt messages based on the collected user data and user requests.

[1207] The server inputs prompt text into a generating AI (e.g., the latest chatGPT technology) to create a virtual customer service representative.

[1208] The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal.

[1209] Input: Filtered user data, preferred customer service style information

[1210] Output: Text data, talk scripts, and character information of the virtual customer service representative.

[1211] Step 4:

[1212] Display of the virtual customer service representative and initiation of interaction.

[1213] The device displays a virtual customer service representative on the screen.

[1214] The virtual customer service representative begins by saying, "Hello! What kind of product are you looking for today?"

[1215] Input: Text data, talk script, and character information of the virtual customer service representative.

[1216] Output: A virtual customer service representative is displayed on the user's terminal, and the interaction begins.

[1217] Step 5:

[1218] User emotion recognition

[1219] The device has a built-in emotion engine that analyzes the user's facial expressions, voice tone, and input text.

[1220] When a user is visible through the camera, the device uses facial recognition technology to identify their emotional state.

[1221] When a user speaks, the device performs voice tone analysis to understand emotional nuances.

[1222] Input: User facial expression data, voice tone data, input text

[1223] Output: The emotional state of the identified user.

[1224] Step 6:

[1225] Response and tone adjustment

[1226] The server receives emotion data sent from the terminal.

[1227] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak in a gentler tone.

[1228] Input: Identified user's emotional state, virtual customer service representative's text data

[1229] Output: Adjusted virtual customer service representative responses and tone

[1230] Step 7:

[1231] Handling user inquiries

[1232] The user enters the question, "What running shoes do you recommend?"

[1233] The terminal sends the question to the server.

[1234] Input: User's question

[1235] Output: Question data sent to the server

[1236] Step 8:

[1237] Generating answers to questions

[1238] The server analyzes the user's question.

[1239] The server searches the database for appropriate product information (e.g., recommended running shoe models and their features).

[1240] The server generates a response based on the collected product information.

[1241] Input: User's question, product information in the database

[1242] Output: Generated response data

[1243] Step 9:

[1244] Submit and display of responses

[1245] The server sends the generated response data to the terminal.

[1246] The device displays the generated response to the user.

[1247] Input: Generated response data

[1248] Output: Answer displayed on the user's terminal

[1249] Step 10:

[1250] Providing effective communication information

[1251] The server generates effective, real-time sales information (e.g., limited-time offers and cross-selling suggestions).

[1252] The server sends data to the terminal to be incorporated into the virtual customer service representative.

[1253] A virtual customer service representative provides users with special offers such as, "These shoes are 20% off right now! We can also recommend socks that would go perfectly with these shoes."

[1254] Input: Real-time conversation information, text data from virtual customer service representatives

[1255] Output: Virtual customer service representative speech with effective conversational information incorporated.

[1256] Step 11:

[1257] Customer service scenarios with celebrities (optional feature)

[1258] When a user selects a customer service style, they can choose a specific person model (for example, a celebrity).

[1259] The server collects data from the selected person model (e.g., voice data, feature data).

[1260] The server generates a virtual customer service representative based on a specific person model and sends it to the terminal.

[1261] The terminal displays this virtual customer service representative and begins interacting with the user.

[1262] Input: User-selected person model, data for the selected person model

[1263] Output: Data and display of a virtual customer service representative generated based on a specific person model.

[1264] Through these steps, the system provides a personalized customer service experience tailored to the user's needs, and further enhances purchasing intent and delivers a highly satisfying shopping experience by enabling real-time responses that take the user's emotions into account.

[1265] (Application Example 2)

[1266] 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."

[1267] Traditional in-store customer service systems faced challenges in providing personalized service tailored to the individual needs and emotions of each user. Furthermore, the inability to make effective real-time suggestions and cross-selling proposals limited the ability to improve customer satisfaction and boost sales. Additionally, there was a lack of customer service systems that could leverage new technologies such as smart glasses.

[1268] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for analyzing the user's facial expressions and voice tone to recognize emotions, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotions, and means for interacting with the virtual customer service representative by wearing smart glasses in a physical store. This enables a personalized customer service experience that responds to the user's needs and emotions, thereby improving customer satisfaction and promoting sales.

[1269] A "virtual customer service representative" is a virtual customer service representative generated based on the user's preferences and collected data.

[1270] "Interaction" refers to two-way communication between the user and the virtual customer service representative.

[1271] "User's past purchase history" refers to a list of products and historical data that the user has purchased so far.

[1272] "Preference data" refers to data related to a user's preferences and interests, and is primarily based on past behavior and preferences.

[1273] "Facial recognition" is a technology that analyzes a user's facial expressions to identify their emotional state at that time.

[1274] "Voice tone" is a technique that identifies a user's emotional state by analyzing the nuances of their voice, including its pitch and volume.

[1275] "Smart glasses" are glasses-type devices that have the function of displaying information from the real world as augmented reality (AR).

[1276] "Real-time response" refers to a response that provides appropriate information immediately in response to a user's question.

[1277] "Effective sales information" refers to information generated to promote sales and make suggestions, tailored to the user's interests and needs.

[1278] "Cross-selling" is a sales strategy that suggests related products that users might be interested in.

[1279] The system program for implementing this invention operates in cooperation between a server and a terminal and is designed to provide users with a personalized customer service experience. This system includes the following main components:

[1280] Component and Processing Overview

[1281] server

[1282] The server is primarily responsible for data collection, execution of generative AI models, and response generation.

[1283] 1. Collection of user data:

[1284] The system collects the user's past purchase history and preference data from a database. This includes a list of products the user has purchased, their browsing history, and their preferences.

[1285] 2. Virtual customer service representative generation:

[1286] Based on the collected data and the user's preferred customer service style, a virtual customer service representative is generated using a generative AI model (e.g., the T5 model). This generation process involves inputting appropriate prompts to the generative AI model and obtaining character information for the virtual customer service representative.

[1287] 3. Real-time response to questions:

[1288] The system analyzes user inquiries and generates appropriate product information and answers. The generated information is then sent to the device.

[1289] terminal

[1290] The device is responsible for user interaction and presents data in real time.

[1291] 1. Emotion recognition:

[1292] The device is equipped with a camera and a voice input device, and recognizes emotions by analyzing the user's facial expressions and voice tone. This analysis utilizes facial recognition technology such as the DeepFace library and voice tone analysis technology.

[1293] 2. Performing an interaction:

[1294] The terminal displays a generated virtual customer service representative and interacts with the user. For example, it provides an environment where the user can converse with a virtual customer service representative through smart glasses.

[1295] 3. Adjusting the response:

[1296] Based on recognized emotion data, the system receives instructions from the server and adjusts the virtual customer service representative's responses and tone. For example, if the user is tired, it will respond in a gentle tone.

[1297] Specific example

[1298] Let's assume the user is wearing smart glasses in a physical store. These smart glasses are equipped with a camera and microphone, and can analyze the user's facial expressions and voice tone in real time.

[1299] When a user requests a "casual and friendly customer service style," the device sends this information to the server. The server collects the user's past purchase history and preference data, and uses a generative AI model to generate an appropriate virtual customer service representative. The generated virtual representative will engage in friendly conversation, for example, saying, "Hello! What products are you looking for today?"

[1300] An example of a prompt is, "Create a virtual assistant with sports and casual preferences and a casual and friendly interaction style." This prompt prompt causes the AI ​​model to generate a virtual customer service representative that aligns with the user's preferences.

[1301] In this way, personalized customer service experiences tailored to user needs and emotions can be provided. This system is highly effective in improving customer satisfaction and boosting sales.

[1302] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1303] Step 1:

[1304] The user inputs their desired customer service style through a device such as smart glasses or a smartphone. The entered information, along with the user's ID, is sent to the server. The input data is in text format and represents the user's preferences.

[1305] Step 2:

[1306] Based on the received user preference data, the server collects the user's past purchase history and preference data from the database. Database queries are used to retrieve a list of products the user has previously purchased, their browsing history, and their preferences. The retrieved data is temporarily stored as a user dataset.

[1307] Step 3:

[1308] The server inputs a prompt message into a generative AI model based on the collected user dataset, generating a virtual customer service representative. The prompt message is "Create a virtual assistant with sports and casual preferences and a casual and friendly interaction style." and the text data output from the generative model (e.g., T5 model) is retrieved. The output text data contains the character information of the virtual customer service representative.

[1309] Step 4:

[1310] The device initiates interaction with the user based on the character information of the virtual customer service representative received from the server. The virtual customer service representative is displayed on the smart glasses' screen and speaks to the user, saying things like, "Hello! What kind of product are you looking for today?" During this process, the displayed text and audio data are generated as interaction data.

[1311] Step 5:

[1312] The device uses its built-in camera and microphone to analyze the user's facial expressions and voice tone, and uses an emotion recognition engine to identify the user's emotions. It analyzes facial expression data using the DeepFace library and voice tone analysis technology to analyze voice data. Detected emotions are stored as emotion data.

[1313] Step 6:

[1314] The server receives emotional data and runs an algorithm that adjusts the virtual customer service representative's responses and tone based on it. For example, if the emotional data identifies the user as "tired," the virtual customer service representative's response will be changed to a gentler tone. This provides the user with a pleasant customer service experience.

[1315] Step 7:

[1316] When a user enters a specific question, the device sends that question to the server. The server analyzes the question and searches its database for appropriate product information. The search results are generated as answer data and sent back to the device.

[1317] Step 8:

[1318] The terminal displays response data received from the server to the user and provides answers through a virtual customer service representative. For example, it might present information such as, "These shoes are lightweight and highly breathable." In this process, the displayed text and audio data are generated as response data.

[1319] Step 9:

[1320] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and sends it to the terminal. The terminal then incorporates this information into the virtual sales representative and presents it to the user. This allows the server to provide users with special offers such as, "These shoes are 20% off right now."

[1321] 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.

[1322] 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.

[1323] 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.

[1324] [Fourth Embodiment]

[1325] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1326] 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.

[1327] 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).

[1328] 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.

[1329] 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.

[1330] 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).

[1331] 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.

[1332] 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.

[1333] 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.

[1334] 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.

[1335] 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.

[1336] 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.

[1337] 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".

[1338] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[1339] System Processing Flow Overview

[1340] User's preference input

[1341] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a customer service style selection form, such as "casual and friendly customer service style." This information is sent from the device to the server.

[1342] Consumer data collection

[1343] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[1344] Virtual Customer Service Representative Generation

[1345] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[1346] Virtual customer service representative display and interaction

[1347] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[1348] User interaction

[1349] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[1350] Generating answers to questions

[1351] The server generates responses based on the collected product information. The generated response data is sent to the terminal, which then displays it to the user. For example, specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes offers excellent cushioning and is ideal for long-distance running," might be displayed.

[1352] Providing an effective talk

[1353] The server generates effective sales information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal displays this information to the user, offering special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[1354] Customer service scenarios with celebrities (optional feature)

[1355] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[1356] Specific example

[1357] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[1358] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[1359] In this way, the system of the present invention provides a practical solution for enhancing purchasing intent by offering a personalized customer service experience that meets the user's needs.

[1360] The following describes the processing flow.

[1361] Step 1:

[1362] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[1363] Step 2:

[1364] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[1365] Step 3:

[1366] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[1367] Step 4:

[1368] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[1369] Step 5:

[1370] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[1371] Step 6:

[1372] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[1373] Step 7:

[1374] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[1375] Step 8:

[1376] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[1377] Step 9:

[1378] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[1379] Step 10:

[1380] The server sends the generated response data to the terminal. The terminal displays the response to the user.

[1381] Step 11:

[1382] The server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual customer service representative.

[1383] Step 12:

[1384] The terminal displays exclusive offer information to the user via a virtual customer service representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also recommend socks that go perfectly with these shoes."

[1385] Through these steps, users will be able to have an ideal customer service experience, which will likely increase their desire to purchase.

[1386] (Example 1)

[1387] 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".

[1388] Traditional online shopping systems required users to spend a lot of time finding the products they wanted to buy, and lacked personalized customer service experiences tailored to individual needs. Furthermore, it was difficult for users with specific interests or styles to receive appropriate advice and suggestions. This could lead to decreased purchasing intent and missed sales opportunities. Additionally, it was challenging to adequately accommodate users who desired customer service representatives based on specific celebrities or characters.

[1389] 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.

[1390] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for transmitting the generated virtual customer service representative's text data, talk script, and character information to the terminal, means for displaying the virtual customer service representative, means for transmitting the user's questions from the terminal to the server, means for searching for appropriate product information from a database based on the user's questions, and means for generating answers from a generated AI model based on the collected product information. This makes it possible to provide a personalized customer service experience that matches the user's wishes and interests, thereby increasing their willingness to purchase. Furthermore, it is possible to generate virtual customer service representatives based on specific celebrities or characters and provide personalized responses to users.

[1391] A "user" refers to a person who uses the system to search for and purchase goods and services.

[1392] "Device" refers to electronic devices used by users, such as smartphones, tablets, and personal computers.

[1393] A "server" refers to a central system that processes data, stores information, and handles requests from terminals.

[1394] "Customer service style" refers to the way a virtual customer service representative responds and their attitude, as chosen by the user.

[1395] "Purchase history" refers to a record of products that a user has purchased through the system so far.

[1396] "Preference data" refers to data that includes information about users' preferences and interests.

[1397] A "virtual customer service representative" refers to a virtual character or assistant created using generative AI to provide customer service to users.

[1398] "Interaction" refers to the dialogue and exchange of information that takes place between the user and the virtual customer service representative.

[1399] A "generative AI model" refers to an artificial intelligence model that generates text or responses based on user input or data.

[1400] A "talk script" refers to a script or pattern for a virtual customer service representative to use when speaking to a user.

[1401] "Character information" refers to data about the appearance, personality, and speaking style of the virtual customer service representative.

[1402] A "prompt statement" refers to a command statement that is input to a generating AI model.

[1403] A "database" refers to an entire system used for collecting, storing, and retrieving data.

[1404] "Real-time" refers to processing or responding to user input almost simultaneously.

[1405] A "specific person model" refers to a virtual customer service representative generated based on data of a celebrity or character selected by the user.

[1406] "Limited-time offers" refer to information about special products or discounts that are available only under specific conditions.

[1407] "Cross-selling" refers to a marketing technique where related products are suggested simultaneously when a customer purchases a particular product.

[1408] Modes for carrying out the invention

[1409] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and enhances the user's willingness to purchase through interaction. This system operates between the user, a terminal, and a server, each playing a specific role.

[1410] Hardware and software to be used

[1411] Devices: Smartphones, tablets, PCs, etc.

[1412] Server: A central system that processes data, stores information, and handles requests from each terminal.

[1413] Database: Stores users' past purchase history and preference data (e.g., MySQL).

[1414] Generative AI models: AI models for generating natural language (e.g., OpenAI GPT-4)

[1415] UI framework: A framework for displaying content on a device (e.g., React Native)

[1416] Operating procedures and specific examples

[1417] Users input their preferred customer service style using a device such as a smartphone or computer. For example, they might enter "casual and friendly customer service style" and press the send button. The device then sends this information to the server.

[1418] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server's process begins with executing a SELECT query to extract all data related to the user ID. Next, edge analytics is used to filter the collected data and create a data set that matches the user's requests.

[1419] The server generates a virtual customer service representative using a generative AI (for example, OpenAI's GPT-4 model) based on the collected data and user requests. A prompt is used during the generation process. An example of a prompt might be, "Generate a virtual customer service representative with a casual customer service style based on the user's preferences." The generated virtual customer service representative's text data, talk script, and character information are sent to the terminal in JSON format.

[1420] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. Using a UI framework such as React Native, the virtual customer service representative's avatar is displayed on the screen and greets the user with "Hello! What products are you looking for today?"

[1421] When a user enters the question "What running shoes do you recommend?", the device sends this question to the server. The device receives the user's input and sends it to the server as a POST request in JSON format.

[1422] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to collect the relevant product information. After that, it uses a generative AI model to create an answer. The generated answer data is sent to the terminal in JSON format.

[1423] The device analyzes the received data and displays a response to the user. For example, it can make specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[1424] Furthermore, the server generates effective conversational information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The terminal receives this information, and the sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that would go perfectly with these shoes."

[1425] Furthermore, if a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the device, which then displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?"

[1426] The above describes embodiments of the present invention, which are detailed procedures for providing a personalized customer service experience tailored to user needs and increasing purchasing intent.

[1427] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1428] Step 1: User input

[1429] The user uses their device to input their preferred service style. Specifically, they launch an application on their device and enter a service style selection form, such as "casual and friendly service style." This action causes the device to send the user's input data to the server. An HTTP POST request is used for this transmission. The input data, "casual and friendly service style," is sent to the server.

[1430] Step 2: Consumer Data Collection

[1431] The server uses the received user ID to access the database and collect the user's past purchase history and preference data. The server first executes a SELECT query to extract all data related to the user ID. This data includes items the user has purchased, browsing history, and favorite categories. The extracted data is then filtered using edge analytics to create a data set that matches the user's requirements. The input is the user ID, and the output is the filtered data set.

[1432] Step 3: Generate a virtual customer service representative

[1433] The server generates a virtual customer service representative using a generative AI (e.g., OpenAI's GPT-4 model) based on the collected data and user requests. In this process, a prompt is input to the generative AI model. For example, "Generate a virtual customer service representative with a casual service style based on the user's preferences" might be the prompt. The AI ​​model analyzes this prompt, generates text data, talk scripts, and character information for a virtual customer service representative that meets the user's preferences, and sends it to the terminal in JSON format. The input is a filtered data set and a prompt, and the output is virtual customer service representative data in JSON format.

[1434] Step 4: Virtual customer service representative display and interaction

[1435] The terminal parses the received JSON data and displays a virtual customer service representative on the screen. A UI framework (e.g., React Native) is used for this. The terminal screen displays an avatar of the virtual customer service representative, who says to the user, "Hello! What products are you looking for today?" The input is virtual customer service representative data in JSON format, and the output is the virtual customer service representative displayed on the screen.

[1436] Step 5: Interaction with the user

[1437] When a user enters the question "What running shoes do you recommend?", the terminal sends this question to the server. The terminal receives the user's input and sends it to the server as a POST request in JSON format. The input is the user's question, and the output is the request to the server.

[1438] Step 6: Generate answers to the questions

[1439] The server analyzes the question from the received JSON request and searches the database for appropriate product information. It then executes another SELECT query to extract the relevant product information. After that, it uses a generative AI model to create an answer and sends the generated answer data to the terminal in JSON format. The input is the user's question, and the output is the answer data in JSON format.

[1440] Step 7: Displaying the response to the user

[1441] The device parses the received JSON data and displays a response to the user. For example, it might display specific suggestions such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The input is response data in JSON format, and the output is the response displayed on the screen.

[1442] Step 8: Delivering an effective talk

[1443] The server generates effective chat information in real time (e.g., limited-time offers and cross-selling suggestions) and incorporates it into the virtual sales representative. The generated chat information is sent from the server to the terminal. The terminal receives this information, and the virtual sales representative displays to the user, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes." The input is the generated chat information, and the output is the chat information displayed on the screen.

[1444] Step 9: Customer service scenario with a celebrity (optional feature)

[1445] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). The server downloads this data via an API and uses a generative AI to generate a virtual customer service representative based on the specific person model. The generated data is sent to the terminal, which displays this virtual customer service representative and begins interacting with the user. Specifically, the customer service representative might say, "Hello, I'm [Name]. What kind of products are you looking for today?" The input is the data of the selected person model, and the output is the virtual customer service representative displayed on the screen.

[1446] (Application Example 1)

[1447] 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".

[1448] Current customer service systems offer limited personalized service experiences. Furthermore, there's a challenge in providing effective recommendations to users in physical stores by leveraging their past purchase history and preference data. Additionally, there are no systems that can generate virtual sales representatives tailored to user preferences and provide interactive product suggestions within the store. This results in users being unable to fully develop their purchasing intent and making optimal product selection difficult.

[1449] 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.

[1450] In this invention, the server includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model. This makes it possible to provide users with a personalized customer service experience even in a physical store and increase their willingness to purchase.

[1451] "Customer service style" refers to the method and attitude of interaction that users prefer.

[1452] "Purchase history" refers to information that records a list of products a user has purchased in the past.

[1453] "Preference data" refers to information that reflects a user's preferences and interests.

[1454] A "virtual customer service representative" refers to a customer service representative created virtually using generative AI, who interacts with the user on screen.

[1455] "Interaction" refers to a two-way exchange between the user and the system.

[1456] "Real-time" refers to responding immediately to user input or actions.

[1457] "Talk information" refers to information that includes the content of conversations with users and the suggestions they made.

[1458] A "physical store" refers to a retail store that provides goods and services in a physical location.

[1459] A "smartphone application" refers to a software program that runs on a smartphone.

[1460] A "generative AI model" refers to an algorithm that uses artificial intelligence technology to generate text and responses.

[1461] A "prompt message" refers to an instruction message that is input into a generative AI model.

[1462] The system of the present invention includes means for inputting the customer service style desired by the user, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative suited to the customer service style based on the collected user data, means for having the generated virtual customer service representative interact with the user, means for answering the user's questions in real time, means for providing effective talk information in real time, means for displaying the virtual customer service representative on a smartphone application for use in a physical store, means for creating responses for the virtual customer service representative using a generation AI model based on user data collected from a database, and means for generating prompt sentences suitable for the generation AI model.

[1463] Hardware and software

[1464] server

[1465] The server includes a database (e.g., MySQL or PostgreSQL) to store user purchase history and preference data. It also provides computing resources to generate responses from virtual customer service representatives using generative AI models (e.g., GPT-3 or the latest ChatGPT technology). Furthermore, it generates prompt messages and sends the generated virtual customer service representative data to the terminal.

[1466] terminal

[1467] The terminal is a mobile device such as a smartphone or tablet, and interaction between the user and a virtual customer service representative is realized through a dedicated application. The application receives user input and sends it to a server. Based on the data received from the server, the virtual customer service representative is displayed, and real-time interaction takes place.

[1468] System Operation Overview

[1469] User's preference input

[1470] Users enter their preferred customer service style using a smartphone application. For example, they might enter "casual and friendly customer service style."

[1471] Data collection

[1472] The server collects past purchase history and preference data from the database based on the user's ID. This data includes items the user has purchased, browsing history, and favorite categories.

[1473] Generation of virtual customer service representatives

[1474] Based on the collected data and user preferences, the server generates prompt messages, which are then input into a generation AI model to create a virtual customer service representative. Examples of prompt messages are as follows:

[1475] User preference: Casual and friendly customer service style

[1476] User data: Past purchase history: running shoes, sportswear. Favorite category: sporting goods.

[1477] Interaction

[1478] The generated virtual customer service representative data is sent to the terminal and displayed within the smartphone application. The virtual customer service representative begins by saying something like, "Hello! What products are you looking for today?" and interacts with the user.

[1479] Providing real-time answers and talk information.

[1480] When a user asks a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server searches its database for appropriate product information and sends the generated answer back to the terminal. Effective sales talk information, such as limited-time offers and cross-selling suggestions, is also provided in real time.

[1481] This makes it possible to provide users with a personalized customer service experience through interaction with virtual sales representatives, thereby increasing their willingness to purchase.

[1482] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1483] Step 1:

[1484] Users input their desired customer service style into a smartphone application. Specifically, they might enter "casual and friendly customer service style" and submit that information. This input data is then sent from the device to the server.

[1485] Input: User's preferred customer service style (e.g., "Casual and friendly customer service style")

[1486] Output: User's requested input data sent to the server

[1487] Step 2:

[1488] The server uses the user's ID to collect past purchase history and preference data from the database. Specifically, it retrieves information such as products the user has purchased, browsing history, and favorite categories from the database. This collected data is used in subsequent processes.

[1489] Input: User ID

[1490] Output: User's past purchase history and preference data

[1491] Step 3:

[1492] The server generates prompt sentences suitable for the generating AI model based on the collected user data. These prompt sentences include the user's desired customer service style and user data. This is then input into the generating AI model to create a virtual customer service representative.

[1493] Input: User's desired input data, past purchase history, and preference data.

[1494] Output: Prompt text to input into the generating AI model (Example: "User preference: Casual and friendly customer service style. User data: Past purchase history: Running shoes, sportswear. Favorite category: Sports goods.")

[1495] Step 4:

[1496] The server uses a generation AI model to generate a virtual customer service representative based on the prompt text. It then sends the generated virtual customer service representative's text data, talk script, and character information to the terminal.

[1497] Input: Prompt message

[1498] Output: Generated virtual customer service representative data (text data, talk script, character information)

[1499] Step 5:

[1500] The terminal displays the virtual customer service representative's data received from the server. Interaction with the user begins, and the virtual customer service representative starts speaking to the user, saying things like, "Hello! What kind of products are you looking for today?"

[1501] Input: Data of the generated virtual customer service representative

[1502] Output: Virtual customer service representative displayed on the screen

[1503] Step 6:

[1504] The user enters a question into the virtual customer service representative. For example, they might type, "What running shoes would you recommend?" This question is then sent from the terminal to the server.

[1505] Input: User's question (e.g., "What running shoes do you recommend?")

[1506] Output: User questions sent to the server

[1507] Step 7:

[1508] The server analyzes the user's question and searches the database for appropriate product information. Specifically, it searches for recommended running shoe models and features, and generates answer data.

[1509] Input: User's question

[1510] Output: Response data generated based on appropriate product information

[1511] Step 8:

[1512] The server sends the generated response data to the terminal. This data includes specific suggestions, such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[1513] Input: Generated response data

[1514] Output: Sending response data to the terminal

[1515] Step 9:

[1516] The terminal displays the received response data to the user. Suggestions from a virtual customer service representative are displayed on the screen, increasing the user's purchasing intent.

[1517] Input: Response data sent from the server

[1518] Output: Response data displayed on the user screen

[1519] Step 10:

[1520] The server generates effective sales pitches, exclusive offers, and cross-selling suggestions in real time and incorporates them into the virtual sales representative. For example, it might offer special offers such as, "These shoes are 20% off right now! We can also show you socks that would go perfectly with them."

[1521] Input: User's current interaction status, reward information

[1522] Output: Generation of talk data incorporating bonus information

[1523] Step 11:

[1524] The terminal displays updated virtual customer service representatives' conversation information in real time, providing users with effective conversation and special offer information. This can further increase users' purchasing intent.

[1525] Input: Talk data sent from the server

[1526] Output: Talk information from the virtual customer service representative displayed on the screen in real time.

[1527] 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.

[1528] The customer service system of the present invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[1529] System Processing Flow Overview

[1530] User's preference input

[1531] The user enters their preferred customer service style using a device (e.g., smartphone, tablet, PC, etc.). Specifically, the user enters a selection form, such as "casual and friendly customer service style," and presses the submit button. This information is sent from the device to the server.

[1532] Consumer data collection

[1533] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user has purchased, browsing history, and favorite categories. The collected data is then filtered to create a data set that matches the user's requests.

[1534] Virtual Customer Service Representative Generation

[1535] Based on the collected data and user requests, the server generates a virtual customer service representative using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[1536] Virtual customer service representative display and interaction

[1537] The terminal displays a virtual sales representative on the screen and begins interacting with the user. Specifically, the virtual sales representative starts by saying something like, "Hello! What kind of product are you looking for today?"

[1538] User emotion recognition

[1539] The device incorporates an emotion engine that analyzes the user's facial expressions, voice tone, and input text. For example, when a user appears on the device via the camera, facial recognition technology identifies their emotional state. Voice tone analysis helps understand the emotional nuances of what the user is saying.

[1540] Adjusting interactions

[1541] The server adjusts the virtual customer service representative's responses and tone based on the identified emotional state. For example, if the user is tired, the virtual customer service representative will speak gently and provide a relaxing experience for the user.

[1542] User interaction

[1543] When a user enters a question such as "What running shoes do you recommend?", the question is sent from the terminal to the server. The server analyzes the question and searches its database for appropriate product information (e.g., recommended running shoe models and their features).

[1544] Generating answers to questions

[1545] The server generates responses based on the collected product information. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running." The generated response data is sent to the terminal, which then displays it to the user.

[1546] Providing an effective talk

[1547] The server generates effective sales talk information in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal to be incorporated into the virtual sales representative. This allows the virtual sales representative to provide users with special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[1548] Customer service scenarios with celebrities (optional feature)

[1549] When a user selects a specific person model (e.g., a celebrity) when choosing a customer service style, the server collects data on the selected person model (e.g., voice data, feature data). Based on the collected data, it generates a virtual customer service representative based on the specific person model and sends it to the device. The device then displays this virtual customer service representative and begins interacting with the user.

[1550] Specific example

[1551] The user selects a "casual and friendly customer service style" on their smartphone app and presses the submit button. The device sends this information to a server, which collects the user's past purchase history and preference data from a database. Based on the collected data, a generating AI creates a virtual customer service representative who is casual and knowledgeable about sports, and sends that data to the device.

[1552] The terminal displays a virtual sales representative and initiates interaction with the user. When the user asks, "What running shoes do you recommend?", the terminal sends the question to the server, which generates an answer based on appropriate product information and sends it back to the terminal. The terminal displays the generated answer to the user and also shows more effective sales pitches and exclusive offers to increase the user's desire to purchase.

[1553] During this time, the emotion engine analyzes the user's facial expressions and tone of voice, and adjusts the virtual customer service representative's response according to the user's emotional state. For example, if the emotion engine determines that the user is a little tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time and look around."

[1554] In this way, the system of the present invention provides a personalized customer service experience that meets the user's needs and further enables responses that take into account the user's emotions in real time, thereby increasing purchasing intent and providing a highly satisfying purchasing experience.

[1555] The following describes the processing flow.

[1556] Step 1:

[1557] The user launches the application and enters their preferred customer service style. Specifically, the user enters "casual and friendly customer service style" in the selection form and presses the submit button.

[1558] Step 2:

[1559] The terminal sends the customer service style preference data entered to the server. The transmitted data includes the user ID and information about the preferred customer service style.

[1560] Step 3:

[1561] The server uses the user ID to access a database and collect data on the user's past purchase history and preferences. This includes data such as the products the user has purchased, their browsing history, and their favorite categories.

[1562] Step 4:

[1563] The server filters the collected data to create a data set that matches the user's requests. For example, if a user has purchased many sporting goods in the past, the server will highlight sport-related product data.

[1564] Step 5:

[1565] Based on the data collected by the server and the user's preferences, a virtual customer service representative is generated using a generation AI (e.g., the latest chatGPT technology). The generated virtual customer service representative's text data, talk script, and character information are then sent to the terminal.

[1566] Step 6:

[1567] The terminal displays a virtual customer service representative on the screen and offers an initial greeting to the user. For example, it might say, "Hello! What kind of product are you looking for today?"

[1568] Step 7:

[1569] The device uses its built-in camera and microphone to activate an emotion engine that analyzes the user's facial expressions and voice tone in real time. The emotion engine detects the user's emotional state from their facial expressions and voice tone.

[1570] Step 8:

[1571] The user enters a question, such as "What running shoes do you recommend?" The entered question is sent from the device to the server.

[1572] Step 9:

[1573] The server analyzes the query and searches the database for appropriate product information. Specifically, it collects information on models and features of running shoes.

[1574] Step 10:

[1575] The server generates responses based on the product information it has collected. For example, it might generate information such as, "These shoes are lightweight and highly breathable. The next pair of shoes has excellent cushioning and is ideal for long-distance running."

[1576] Step 11:

[1577] The server sends the generated response data to the terminal. The terminal displays the response to the user. For example, product features and price information may be displayed.

[1578] Step 12:

[1579] The emotion engine analyzes the user's facial expressions and voice tone to determine whether the user is calm, tired, excited, etc. The emotion engine then sends the identified emotional state to the server.

[1580] Step 13:

[1581] The server adjusts the virtual customer service representative's response and tone based on the identified emotional state. For example, if the server determines that the user is tired, the virtual customer service representative will speak in a gentle tone, saying something like, "You seem tired today. Please take your time browsing."

[1582] Step 14:

[1583] The server generates effective sales pitches in real time (e.g., limited-time offers and cross-selling suggestions) and sends the data to the terminal for incorporation into the virtual sales representative. This allows the virtual sales representative to provide special offers such as, "These shoes are 20% off right now! We can also show you socks that go perfectly with these shoes."

[1584] Step 15:

[1585] The device displays exclusive offers and cross-sell suggestions to the user via a virtual sales representative. If the user is interested in the information, more detailed information is provided to increase their purchase intent.

[1586] In this way, the system of the present invention provides a personalized customer service experience based on the user's wishes and real-time emotional state, thereby increasing the user's desire to purchase. Furthermore, by providing effective conversational information and adjusting responses based on emotion recognition, it realizes a highly satisfying purchasing experience.

[1587] (Example 2)

[1588] 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".

[1589] In modern online shopping, users demand personalized customer service experiences, but current systems fail to meet these demands. In particular, there is a lack of customized customer service styles tailored to user preferences and in-depth responses that consider user emotional states. Furthermore, the creation of virtual customer service representatives using specific person models, as well as real-time limited-time offers and cross-selling suggestions, are not yet realized. A system is needed to address these challenges and provide users with a highly satisfying purchasing experience.

[1590] 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.

[1591] In this invention, the server includes means for inputting the user's desired customer service style, means for collecting the user's past purchase history and preference data, means for generating a virtual customer service representative that matches the customer service style based on the collected user data, means for displaying the generated virtual customer service representative on the user terminal and allowing interaction, means for recognizing the user's emotional state using an emotion engine, means for adjusting the virtual customer service representative's responses and tone based on the recognized emotional state, means for searching for appropriate product information from a database based on the user's questions and generating answers, and means for providing effective talk information in real time. This enables a personalized customer service experience that meets the user's needs and responses that respond to their emotional state, and also enables the provision of limited offers and cross-selling suggestions in real time.

[1592] A "user" refers to an individual who uses the system to input their preferred customer service style or interact with a virtual customer service representative.

[1593] A "terminal" is a device used by a user to access a system and input or receive information, and includes smartphones, tablets, and personal computers.

[1594] A "server" is a central control unit that collects, processes, and generates data based on user input.

[1595] "Customer service style" refers to the communication characteristics and attitudes of the virtual customer service representative that the user desires.

[1596] "User data" refers to a collection of data related to a user, including their past purchase history and preference data.

[1597] A "virtual customer service representative" refers to a virtual character that is generated based on the user's preferences and data, and that interacts with the user online.

[1598] "Interaction" refers to two-way communication that takes place between the user and the virtual customer service representative.

[1599] An "emotion engine" refers to a technology that analyzes a user's facial expressions, voice tone, and input text to identify the user's emotional state.

[1600] "Emotional state" refers to the user's psychological and emotional condition, and includes things like being tired or excited.

[1601] "Response" refers to the reply or reaction that a virtual customer service representative gives to a user's questions or requests.

[1602] "Tone" refers to the manner in which a virtual customer service representative speaks and their overall demeanor.

[1603] A "database" refers to an information management system used to store user data, product information, and other similar data.

[1604] "Product information" refers to detailed data about products stored within the system, including features, price, and stock status.

[1605] "Talk information" refers to the content and flow of conversations used by virtual customer service representatives in their interactions with users.

[1606] A "limited offer" refers to a discount or benefit that is offered exclusively for the purchase of a specific product.

[1607] "Cross-selling suggestions" refer to suggestions for additional products related to the product a user is considering purchasing.

[1608] A "character model" refers to a collection of data used to generate virtual customer service representatives based on the characteristics of specific celebrities or characters.

[1609] The customer service system of this invention generates a personalized virtual customer service representative based on user input and further recognizes the user's emotions to adjust its response. This system operates between the user, a terminal, and a server, each playing a specific role.

[1610] Users input their preferred customer service style using devices such as smartphones, tablets, or personal computers. For example, a user might enter "casual and friendly customer service style" into a selection form and press the submit button. This information is then sent from the device to the server.

[1611] The server uses the user's ID to access a database and collects the user's past purchase history and preference data. This data includes items the user...

Claims

1. A means for users to input their preferred customer service style, A means of collecting users' past purchase history and preference data, A means of generating a virtual customer service representative suited to the customer service style based on collected user data, A means of allowing the generated virtual customer service representative to interact with the user, A means of answering user questions in real time, A means of providing effective talk information in real time, A system that includes this.

2. A means for users to select a specific person model from among the customer service styles they prefer, A means of collecting data on selected human models, Includes means for generating a virtual customer service representative based on a specific person model based on collected data, The system according to claim 1.

3. This includes means of providing users with exclusive offers and cross-selling suggestions in real time. The system according to claim 1.

Images