system

Abstract

We provide the system. [Solution] A means to automatically receive and analyze customer inquiries using internal company data, A means of searching related databases based on the analyzed query, A means of automatically generating and sending a response to the customer based on the search results, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Responding to inquiries from customers is one of the important tasks in enterprises, and there is a demand for its efficiency improvement. However, conventionally, there are many parts that rely on manual labor, and there are problems in response speed and response accuracy. In addition, it is also time-consuming to create estimates based on different price information for each enterprise and to carry out appropriate follow-ups for inquiries, resulting in a high workload. To solve such problems, a system that integrally performs automatic response to inquiries, estimate generation, and schedule management is required.

Means for Solving the Problems

[0005] The present invention provides a system that solves the aforementioned problems by the following means: a system that includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, and means for automatically generating and sending responses to customers based on the search results. It also includes means for searching each company's pricing database and automatically generating quotations in a format accessible to customers, and means for generating and providing customers with a downloadable link to the generated quotations. Furthermore, it includes means for tracking the processing status of customer inquiries, setting dates and times when follow-up is required, and sending push notifications as reminders to customers and staff based on the set dates and times. This significantly improves the efficiency of all customer service operations.

[0006] "Internal company data" refers to various types of information and databases managed within a company, including customer information, product information, inventory information, etc.

[0007] "Customer inquiries" refer to information such as questions and requests sent by customers, including inquiries about product inventory status and requests for quotation issuance.

[0008] "Automated response" refers to a function that automatically generates and sends responses to customer inquiries using a pre-configured program.

[0009] "Analysis" is the process of processing received information based on an analysis algorithm to extract necessary keywords and related information.

[0010] "Related databases" refer to databases that store information related to customer inquiries, and include inventory information databases and pricing information databases.

[0011] A "quotation" refers to a document containing price information provided to a customer, and typically includes details of the price of a product or service, the quantity, and the total amount.

[0012] A "download link" refers to a URL or hyperlink that allows users to obtain a file via the internet. By clicking on it, users can download files such as quotations.

[0013] "Follow-up" refers to the process of providing additional support or confirmation after a period of time has elapsed since a customer made an inquiry or request, and includes, for example, a follow-up contact after sending a sample.

[0014] A "reminder" refers to a notification or alarm designed to prompt a scheduled action at a specific date and time, and includes those sent via push notifications.

[0015] "Push notifications" refer to messages or notifications that are automatically sent from a server to a user's device, prompting the recipient to take a specific action promptly. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8]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

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described according to the accompanying drawings.

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

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

[0020] 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.

[0021] 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.

[0022] 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).

[0023] 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."

[0024] [First Embodiment]

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

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0031] 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.

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

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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".

[0037] This invention is a system that utilizes internal company data to centrally manage automated responses to customer inquiries, automated generation of quotations, and follow-up schedules.

[0038] 1. Automated response to customer inquiries

[0039] User

[0040] 1. The user enters their inquiry details on their device and clicks the send button.

[0041] 2. For example, a user enters "I would like to know the stock status of product A."

[0042] terminal

[0043] 1. The terminal saves the user's input and sends the inquiry data to the server.

[0044] server

[0045] 1. The server analyzes the received query data.

[0046] 2. Based on the analysis results, search for relevant databases (e.g., inventory databases).

[0047] 3. Automatically generate a response message based on the search results (e.g., availability of product A).

[0048] 4. Send the generated response message to the user's terminal.

[0049] 2. Automatic generation of quotations

[0050] User

[0051] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[0052] 2. For example, you could request, "Please issue a quotation for product B."

[0053] terminal

[0054] 1. The terminal saves the request data and sends it to the server.

[0055] server

[0056] 1. The server parses the received request.

[0057] 2. Search the approval database for price data of the target company and product.

[0058] 3. Automatically generate a quotation based on the search results. (Example: PDF format)

[0059] 4. Save the generated quotation and create a download link.

[0060] 5. Send the download link to the user's device.

[0061] User

[0062] 1. The user clicks the received download link to download the quotation.

[0063] 3. Schedule management for follow-up

[0064] User

[0065] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[0066] terminal

[0067] 1. The device sends requests requiring follow-up to the server.

[0068] server

[0069] 1. The server analyzes the received request and determines whether follow-up is necessary.

[0070] 2. Set the date and time for follow-up and register it in the schedule management database.

[0071] 3. Generate a push notification as a reminder based on the set date and time.

[0072] 4. Send push notifications to the user's and the person in charge's devices.

[0073] terminal

[0074] 1. Users and staff will receive push notifications and take appropriate action.

[0075] Specific example

[0076] For example, consider a case where a user submits an inquiry requesting "Please send me a sample of product C." In this case, the server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a schedule for follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and, based on the instructions, confirm the sample shipment and follow up.

[0077] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and schedule management to automate and streamline customer service operations, thereby achieving fast and accurate customer service.

[0078] The following describes the processing flow.

[0079] Automated response to customer inquiries

[0080] Step 1:

[0081] The user enters their inquiry details on their device and clicks the send button.

[0082] Step 2:

[0083] The terminal saves the user's input and sends the inquiry data to the server.

[0084] Step 3:

[0085] The server analyzes the received query data and extracts keywords and related information.

[0086] Step 4:

[0087] Based on the analysis results, the server searches relevant databases (e.g., inventory databases).

[0088] Step 5:

[0089] The server generates an appropriate response message based on the search results.

[0090] Step 6:

[0091] The server sends the generated response message to the terminal.

[0092] Step 7:

[0093] The terminal receives a response from the server and displays it to the user.

[0094] Automatic generation of quotations

[0095] Step 1:

[0096] The user enters a request to request a quote and clicks the submit button.

[0097] Step 2:

[0098] The terminal saves the request data and sends it to the server.

[0099] Step 3:

[0100] The server analyzes the received request and extracts price data for the target company and product.

[0101] Step 4:

[0102] The server retrieves price information from the approval database.

[0103] Step 5:

[0104] The server automatically generates a quote based on the price information (e.g., in PDF format).

[0105] Step 6:

[0106] The server saves the generated quote and creates a download link.

[0107] Step 7:

[0108] The server sends a download link to the device.

[0109] Step 8:

[0110] The device receives the download link and displays it to the user.

[0111] Step 9:

[0112] The user clicks the download link to download the quote.

[0113] Follow-up schedule management

[0114] Step 1:

[0115] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[0116] Step 2:

[0117] The device sends requests to the server that require follow-up.

[0118] Step 3:

[0119] The server analyzes the received request and determines whether follow-up is necessary.

[0120] Step 4:

[0121] The server sets the date and time for follow-up and registers it in the schedule management database.

[0122] Step 5:

[0123] The server generates a push notification as a reminder when the scheduled date and time approach.

[0124] Step 6:

[0125] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[0126] Step 7:

[0127] The device receives push notifications and displays them to the user and the person in charge.

[0128] Step 8:

[0129] Users and their representatives will take appropriate follow-up actions based on push notifications.

[0130] (Example 1)

[0131] 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."

[0132] In corporate customer service operations, traditional manual responses are time-consuming and labor-intensive, leading to decreased customer satisfaction. In particular, efficiently managing inquiries, issuing quotations, and follow-ups is difficult, and there is a demand for quick and accurate responses. Furthermore, the lack of a unified management system for these tasks necessitates information integration and automation.

[0133] 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.

[0134] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for generating response messages using a generation AI model. This enables rapid and accurate automated responses to customer inquiries.

[0135] In this invention, the server includes means for searching a database of prices offered by each company and automatically generating a quotation in a format accessible to the customer; means for generating a downloadable link for the generated quotation and providing it to the customer; means for downloading the quotation via the provided link; and means for adjusting the content of the quotation using a generation AI model. This enables the rapid and automatic creation and provision of quotations.

[0136] In this invention, the server includes means for tracking the processing status of customer inquiries and setting a date and time when follow-up is required; means for sending push notifications as reminders to customers and personnel based on the set date and time; means for receiving push notifications; and means for generating the content of the reminder using a prompt statement. This enables efficient scheduling of follow-ups and timely responses to be taken.

[0137] These measures enable companies to automate and streamline customer service operations, contributing to improved customer satisfaction.

[0138] "Internal company data" refers to information and documents generated and stored within a company.

[0139] "Customer inquiries" refer to information about questions and requests that customers make to a company.

[0140] "Means of receiving and analyzing data" refers to the technology that allows a system to take in data from an external source and understand its contents.

[0141] A "related database" refers to a database that stores information related to a specific query.

[0142] "Searching methods" refer to techniques for finding information within a database based on specific criteria.

[0143] "Means of generating responses" refers to technologies that automatically create appropriate answers to inquiries.

[0144] A "generative AI model" refers to an algorithm that uses artificial intelligence to generate new information from data.

[0145] A "downloadable link" refers to a web address that allows a user to download a specific file by clicking on it.

[0146] A "quote" refers to a document that lists the price, details, and terms and conditions of a product or service.

[0147] "Means of tracking" refers to technologies that track and record the progress of a specific process or event.

[0148] "Follow-up" refers to additional actions or confirmations taken after the initial response or action.

[0149] "Push notifications" refer to a technology that sends information from a server to a device in real time.

[0150] A "prompt statement" refers to an input statement used to instruct a generative AI model to generate information.

[0151] This invention is a system that automates and streamlines customer service operations by utilizing internal company data. The program processing of this system is described in detail below.

[0152] Automated response to customer inquiries

[0153] 1. User

[0154] The user enters their inquiry details on their device and clicks the "Send" button.

[0155] For example, you would enter an inquiry such as, "Please tell me the stock status of product A."

[0156] 2. Terminal

[0157] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[0158] 3. Server

[0159] The server parses the received query data and uses natural language processing libraries (e.g., spaCy or NLTK) to analyze the information the user is seeking.

[0160] Based on the analysis results, the system searches relevant databases (e.g., inventory databases) and issues SQL queries.

[0161] Based on the search results, a prompt is sent to a generative AI model (e.g., OpenAI®'s GPT-4®) to generate a response message. Example prompt: "How much of product A is currently in stock?"

[0162] The generated response message is sent to the user's terminal in JSON format.

[0163] 4. Terminal

[0164] The terminal displays the received response message on the user interface.

[0165] Automatic generation of quotations

[0166] 1. User

[0167] The user enters the request for a quote and clicks the "Submit" button.

[0168] For example, you might enter, "Please issue a quotation for product B."

[0169] 2. Terminal

[0170] The terminal saves the request data to a local database and then sends it to the server.

[0171] 3. Server

[0172] The server analyzes the request data and extracts the necessary information using a natural language processing library.

[0173] Search the price database and issue an SQL query. Example SQL query: "SELECT price FROM products WHERE product_name='product B'"

[0174] Based on the search results, the system automatically generates a quotation in PDF format using libraries such as Apache® PDFBox.

[0175] The generated quote is saved to cloud storage (e.g., AWS® S3), and a downloadable link is generated and sent to the user's device.

[0176] 4. User

[0177] The user clicks the provided link and downloads the quote.

[0178] Follow-up schedule management

[0179] 1. User

[0180] The user enters a request that requires follow-up and clicks the "Send" button.

[0181] For example, you might enter, "Please send me a sample of product C."

[0182] 2. Terminal

[0183] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[0184] 3. Server

[0185] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[0186] Set the necessary follow-up dates and register them in the schedule management database (e.g., MICROSOFT® SQ®L Server).

[0187] Using Firebase Cloud Messaging, push notifications are generated at a set date and time and sent to the user's and the assigned person's devices.

[0188] 4. Terminal

[0189] Users and staff members receive push notifications and follow up accordingly.

[0190] Hardware and software used

[0191] Server: Cloud server (e.g., AWS EC2, Microsoft Azure®)

[0192] Database management systems: MySQL®, Oracle, Microsoft SQL Server

[0193] Natural language processing libraries: spaCy, NLTK

[0194] Generative AI models: OpenAI GPT-4, etc.

[0195] Cloud storage: AWS S3

[0196] PDF generation library: Apache PDFBox

[0197] Push notification service: Firebase Cloud Messaging

[0198] Specific example

[0199] For example, if a user submits an inquiry requesting "Please send me a sample of product C," the following process takes place: The server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a follow-up appointment in the schedule management database for three days later and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and confirm and carry out the sample shipment based on the instructions.

[0200] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and follow-up schedule management to automate and streamline customer service operations, thereby achieving prompt and accurate customer service.

[0201] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0202] Processing steps for automated responses to customer inquiries

[0203] Step 1:

[0204] User

[0205] The user enters their inquiry details on their device and clicks the "Send" button.

[0206] Input: User inquiry (e.g., "Please tell me the stock status of product A")

[0207] Output: A request is generated to send query data to the terminal.

[0208] Step 2:

[0209] terminal

[0210] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[0211] Input: User inquiry data

[0212] Output: Send query data to the server in JSON format.

[0213] Step 3:

[0214] server

[0215] The server analyzes the received query data and uses a natural language processing library to understand the intent of the query.

[0216] Software used: spaCy, NLTK

[0217] Input: Query data in JSON format

[0218] Data processing: Text analysis using natural language processing

[0219] Output: Analysis results of the inquiry

[0220] Step 4:

[0221] server

[0222] Based on the analysis results, SQL queries are issued to relevant databases (e.g., inventory databases) to perform searches.

[0223] Database management system used: MySQL

[0224] Input: Analysis results of the inquiry content

[0225] Data Calculation: Searching Related Databases

[0226] Output: Database search results (e.g., "Inventory quantity of product A")

[0227] Step 5:

[0228] server

[0229] A response message is generated using an AI model based on the database search results.

[0230] Generative AI model used: OpenAI GPT-4

[0231] Input: Database search results

[0232] Data processing: Generating response messages using a generative AI model.

[0233] Output: Response message (Example: "Product A is in stock")

[0234] Step 6:

[0235] server

[0236] The server sends the generated response message to the user's terminal in JSON format.

[0237] Input: Response message

[0238] Output: Send a JSON-formatted response message to the user's terminal.

[0239] Step 7:

[0240] terminal

[0241] The terminal displays the received response message on the user interface.

[0242] Input: Response message in JSON format

[0243] Output: Display in the user interface

[0244] Process steps for automatic quotation generation

[0245] Step 1:

[0246] User

[0247] The user enters the request for a quote and clicks the "Submit" button.

[0248] Input: Product name and quotation request (Example: "Please issue a quotation for product B")

[0249] Output: Data is input to the terminal, and a transmission command is generated.

[0250] Step 2:

[0251] terminal

[0252] The terminal saves the request data to a local database and then sends it to the server.

[0253] Input: User's request data

[0254] Output: Send request data to the server in JSON format.

[0255] Step 3:

[0256] server

[0257] The server parses the request data and searches the database to retrieve relevant pricing information.

[0258] Software used: Natural language processing libraries (spaCy, NLTK)

[0259] Input: Request data

[0260] Data processing: Text analysis using natural language processing

[0261] Output: Analyzed request content

[0262] Step 4:

[0263] server

[0264] Based on the parsed request, an SQL query is issued to the price database to retrieve the necessary price information.

[0265] Database management system used: Oracle

[0266] Input: Parsed request content

[0267] Data Calculation: Searching Price Databases

[0268] Output: Price data (Example: "Price information for product B")

[0269] Step 5:

[0270] server

[0271] Based on the acquired price data, an AI model is used to automatically generate a quotation.

[0272] Generative AI model used: OpenAI GPT-4

[0273] PDF generation library used: Apache PDFBox

[0274] Input: Price data

[0275] Data calculation: Generating a PDF quotation

[0276] Output: Generated quotation (PDF format)

[0277] Step 6:

[0278] server

[0279] Save the generated quotation in cloud storage and generate a downloadable link to send to the user's terminal.

[0280] Cloud storage to be used: AWS S3

[0281] Input: Generated quotation (in PDF format)

[0282] Output: Download link

[0283] Step 7:

[0284] User

[0285] The user clicks the provided link to download the quotation.

[0286] Input: Download link

[0287] Output: Downloaded quotation (in PDF format)

[0288] Processing steps for follow-up schedule management

[0289] Step 1:

[0290] User

[0291] The user inputs a request that requires follow-up and clicks the "Send button".

[0292] Input: Follow-up request (e.g., "Please send a sample of Product C")

[0293] Output: Data is input into the terminal and a transmission instruction is generated.

[0294] Step 2:

[0295] Terminal

[0296] The terminal saves requests that require follow-up in the local database and sends them to the server.

[0297] Input: Request data that requires follow-up

[0298] Output: Send the request data to the server in JSON format

[0299] Step 3:

[0300] Server

[0301] The server analyzes the follow-up request and determines whether follow-up is necessary using a natural language processing library.

[0302] Software to be used: Natural language processing libraries (spaCy, NLTK)

[0303] Input: Follow-up request data

[0304] Data processing: Text analysis by natural language processing <00,00961> Output: Judgment result of the necessity of follow-up

[0306] Step 4:

[0307] Server

[0308] Set the date and time when follow-up is necessary and register it in the schedule management database. <,

[0309] Database management system to be used: Microsoft SQL Server

[0310] Input: Judgment result of the necessity of follow-up

[0311] Data calculation: Setting and registration of follow-up date and time

[0312] Output: Configured follow-up schedule

[0313] Step 5:

[0314] server

[0315] Based on the configured date and time, push notifications are generated using Firebase Cloud Messaging and sent to the user's and assigned personnel's devices.

[0316] Push notification service used: Firebase Cloud Messaging

[0317] Input: Configured follow-up schedule

[0318] Data processing: Generating and sending push notifications

[0319] Output: Push notifications to user and staff terminals

[0320] Step 6:

[0321] terminal

[0322] Users and staff members receive push notifications and follow up accordingly.

[0323] Input: Push notification

[0324] Output: Follow-up will be performed.

[0325] Through the specific processing steps outlined above, the system can automate and streamline customer service operations.

[0326] (Application Example 1)

[0327] 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."

[0328] Traditional e-commerce sites struggled to respond quickly and accurately to customer inquiries, particularly with manual processes like checking inventory and issuing quotes, resulting in inefficiencies. Furthermore, managing follow-up schedules was cumbersome, often leading to delays in notifications to both customers and staff. This resulted in decreased customer satisfaction and missed sales opportunities. There is a need to solve these problems and automate and streamline customer service.

[0329] 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.

[0330] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data; means for searching relevant databases based on the analyzed inquiries; means for automatically generating and sending responses to customers based on the search results; means for analyzing customer inquiries regarding specific product information or inventory status using natural language processing; means for performing natural language processing using a generative AI model to automatically generate appropriate responses; means for searching each company's pricing database and automatically generating quotations in a format accessible to customers; means for generating and providing downloadable links for the generated quotations to customers; means for receiving quotation requests, analyzing product information and pricing data using a generative AI model, and automatically generating quotations; means for tracking the processing status of customer inquiries and setting dates and times for follow-up; means for sending push notifications to customers and personnel as reminders based on the set dates and times; and means for using a schedule management database to automatically generate reminders when follow-up is required. This enables not only a quick and accurate response to customer inquiries but also efficient automatic quotation generation and follow-up schedule management.

[0331] "Internal company data" refers to the collection of all information and data managed within a company.

[0332] An "inquiry" refers to a question or request sent by a customer to confirm product information, stock availability, etc.

[0333] "Analysis" is the process of breaking down the content of an received inquiry and interpreting its meaning and intent.

[0334] A "database" is a system that stores and manages related information in an organization.

[0335] "Searching" is the process of finding specific information within a database.

[0336] "Response" refers to the answers or information provided in response to customer inquiries.

[0337] "Natural language processing" is the technology that enables computers to understand and process human language.

[0338] A "generative AI model" refers to an algorithm or framework that uses artificial intelligence to generate new text or data.

[0339] A "price database" is a database where product price information is stored in an organ.

[0340] A "quotation" is a document that shows the price and conditions of a specific product or service.

[0341] "Follow-up" refers to additional confirmations or follow-ups conducted after the initial response.

[0342] A "schedule management database" is a system that organizes and stores follow-up dates and appointments.

[0343] A "reminder" is a message sent to notify you of an action that needs to be taken at a specific date and time.

[0344] A "push notification" is a notification message that is sent to a user's device in real time.

[0345] The specific system for implementing the present invention is capable of efficiently and quickly processing customer inquiries and easily managing the automatic generation of quotations and follow-up schedules. This system is configured as follows.

[0346] Automated response to customer inquiries

[0347] The server automatically receives and analyzes customer inquiries using internal company data. It receives the inquiry content entered by the customer on their device and sends it to the server. The server uses Google® Cloud AI to perform natural language processing and analyze the inquiry content. It then searches relevant databases (e.g., product inventory database) and automatically generates an appropriate response. The generated response is sent to the customer's device.

[0348] Specific example

[0349] For example, if a customer asks, "What is the stock status of product A?", the server analyzes the inquiry, retrieves the stock status of product A from the product inventory database, and automatically generates a response message saying, "Product A is currently in stock."

[0350] Example of a prompt:

[0351] Inquiry: "Please tell me the stock status of product A."

[0352] Generated AI prompt: "Check the inventory status of product A and respond to the customer whether it is in stock or not."

[0353] Automatic generation of quotations

[0354] When a customer requests a quote for a specific product, the request data is sent from the terminal to the server. The server receives and analyzes the request. Next, it retrieves the price data for the product from the price database and automatically generates a quote using a generation AI model. The generated quote is saved in PDF format, and a download link is sent to the customer.

[0355] Specific example

[0356] When a customer requests "Please issue a quote for product B," the server analyzes the request, retrieves the price information for product B from the price database, and automatically generates a quote. A download link for the generated quote is then sent to the customer's device.

[0357] Example of a prompt:

[0358] Quote Request: "Please issue a quote for product B."

[0359] Generated AI prompt: "Retrieve pricing data for product B, generate a quote, and provide it to the customer."

[0360] Follow-up schedule management

[0361] If a customer requires follow-up on an inquiry, they send a follow-up request from their device to the server. The server parses this request and sets a date and time for the follow-up. This information is registered in the schedule management database, and a push notification is generated as a reminder based on the set date and time. This push notification is sent to the customer's and the agent's devices.

[0362] Specific example

[0363] For example, if a customer requests a sample of product C, the server analyzes the request and determines that follow-up is necessary. It sets a schedule for a follow-up three days later and registers it in the schedule management database. Three days later, a push notification is automatically sent as a reminder.

[0364] Example of a prompt:

[0365] Follow-up request: "Please send me a sample of product C."

[0366] Generated AI prompt: "Please schedule a follow-up for sending a sample of Product C in 3 days and send a reminder."

[0367] As described above, the present invention is a system that enables rapid and accurate responses to customer inquiries, automatic generation of quotations, and efficient scheduling of follow-ups. This improves customer satisfaction and streamlines operations.

[0368] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0369] Processing steps for automated responses to customer inquiries

[0370] Step 1:

[0371] The user enters their inquiry details on their device and clicks the send button. An example of input data is, "Please tell me the stock status of product A." Input: Inquiry details. Output: Sending of inquiry data.

[0372] Step 2:

[0373] The terminal saves user input and sends query data to the server. Network communication is performed for data transfer. Input: Query data. Output: Data sent to the server.

[0374] Step 3:

[0375] The server receives query data, which is then processed using natural language processing and analysis with Google Cloud AI. Input: Query data. Output: Analysis results.

[0376] Step 4:

[0377] The server searches relevant databases (e.g., inventory databases) based on the analysis results and retrieves the necessary information. Input: Analysis results. Output: Search results.

[0378] Step 5:

[0379] The server automatically generates response messages based on search results, using a generation AI model. Input: Search results. Output: Response message.

[0380] Step 6:

[0381] The server generates a response message and sends it to the user's terminal. Input: Response message. Output: Message sent to the user's terminal.

[0382] Process steps for automatic quotation generation

[0383] Step 1:

[0384] The user requests a quotation and clicks the send button on their terminal. Example input: "Please issue a quotation for product B." Input: Quotation request. Output: Quotation request data sent.

[0385] Step 2:

[0386] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Estimate request data. Output: Data sent to the server.

[0387] Step 3:

[0388] The server analyzes the request and uses a generated AI model to retrieve product information and pricing data. Input: Quotation request data. Output: Analysis results and pricing information.

[0389] Step 4:

[0390] The server automatically generates a quotation based on the analysis results and saves it in PDF format. Input: Analysis results and price information. Output: Quotation in PDF format.

[0391] Step 5:

[0392] The server generates a download link for the generated quotation and sends it to the user's device. Input: Quotation in PDF format. Output: Sending of download link.

[0393] Follow-up scheduling process steps

[0394] Step 1:

[0395] The user enters a request requiring follow-up and clicks the send button on their device. Example input: "Please send a sample of product C." Input: Follow-up request. Output: Sending of follow-up request data.

[0396] Step 2:

[0397] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Follow-up request data. Output: Data sent to the server.

[0398] Step 3:

[0399] The server analyzes the received request data and sets the date and time for follow-up. Input: Follow-up request data. Output: Follow-up schedule.

[0400] Step 4:

[0401] The server registers the follow-up date and time in the schedule management database. Input: Follow-up schedule. Output: Schedule registration information.

[0402] Step 5:

[0403] The server generates a push notification as a reminder based on the configured date and time. Input: Follow-up schedule. Output: Push notification.

[0404] Step 6:

[0405] The server generates push notifications and sends them to the user's and the assigned personnel's devices. Input: Push notification. Output: Notification sent to the user's and the assigned personnel's devices.

[0406] 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.

[0407] This invention is a system that utilizes internal company data to automatically respond to customer inquiries, automatically generate quotations, manage follow-up schedules, and combines this with an emotion engine that recognizes user emotions.

[0408] 1. Automated response to customer inquiries

[0409] User

[0410] 1. The user enters their inquiry details on their device and clicks the send button.

[0411] 2. For example, a user enters "I would like to know the stock status of product A."

[0412] terminal

[0413] 1. The terminal saves the user's input and sends the inquiry data to the server.

[0414] server

[0415] 1. The server analyzes the received query data and extracts keywords and related information.

[0416] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[0417] 3. Based on the recognized emotions, generate data to adjust the response message.

[0418] 4. Based on the analysis results and sentiment data, search relevant databases (e.g., inventory databases).

[0419] 5. Based on the search results, generate appropriate response messages that reflect sentiment data.

[0420] 6. Send the generated response message to the user's terminal.

[0421] 2. Automatic generation of quotations

[0422] User

[0423] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[0424] 2. For example, you could request, "Please issue a quotation for product B."

[0425] terminal

[0426] 1. The terminal saves the request data and sends it to the server.

[0427] server

[0428] 1. The server analyzes the received request and extracts price data for the target company and product.

[0429] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[0430] 3. Based on the recognized emotion data, generate data to adjust the contents of the quotation.

[0431] 4. Retrieve price information from the approval database.

[0432] 5. Automatically generate quotes based on price information and sentiment data (e.g., in PDF format).

[0433] 6. Save the generated quotation and create a download link.

[0434] 7. Send the download link to the user's device.

[0435] User

[0436] 1. The user clicks the received download link to download the quotation.

[0437] 3. Schedule management for follow-up

[0438] User

[0439] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[0440] terminal

[0441] 1. The device sends requests requiring follow-up to the server.

[0442] server

[0443] 1. The server analyzes the received request and determines whether follow-up is necessary.

[0444] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[0445] 3. Based on the recognized emotional data, adjust the priorities and methods of follow-up.

[0446] 4. Set the date and time for follow-up and register it in the schedule management database.

[0447] 5. When the set date and time approach, generate a push notification as a reminder.

[0448] 6. Send the generated push notification to the user's and the person in charge's devices.

[0449] terminal

[0450] 1. Users and staff will receive push notifications and take appropriate action.

[0451] Specific example

[0452] For example, consider a case where a user submits an inquiry requesting a sample of product C. In this case, the server analyzes the inquiry and determines that follow-up is necessary to send the sample. It also uses an emotion engine to recognize if the user feels a sense of urgency. The server sets a schedule for a follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge then confirm the sample delivery and follow up based on these instructions.

[0453] As described above, the system of the present invention can automate and streamline customer service operations, and provide a better customer experience through responses and methods that take into account the user's emotions.

[0454] The following describes the processing flow.

[0455] Automated response to customer inquiries

[0456] Step 1:

[0457] The user enters their inquiry details on their device and clicks the send button.

[0458] Step 2:

[0459] The terminal saves the user's input and sends the inquiry data to the server.

[0460] Step 3:

[0461] The server analyzes the received query data and extracts keywords and related information.

[0462] Step 4:

[0463] The server uses an emotion engine to recognize the user's emotions.

[0464] Step 5:

[0465] The server searches relevant databases (e.g., inventory databases) based on the analysis results and sentiment data.

[0466] Step 6:

[0467] The server generates a response message based on the search results.

[0468] Step 7:

[0469] The server incorporates sentiment data into the generated response message and makes appropriate adjustments.

[0470] Step 8:

[0471] The server sends the generated response message to the terminal.

[0472] Step 9:

[0473] The terminal receives a response from the server and displays it to the user.

[0474] Automatic generation of quotations

[0475] Step 1:

[0476] The user enters a request to request a quote and clicks the submit button.

[0477] Step 2:

[0478] The terminal saves the request data and sends it to the server.

[0479] Step 3:

[0480] The server analyzes the received request and extracts price data for the target company and product.

[0481] Step 4:

[0482] The server uses an emotion engine to recognize the user's emotions.

[0483] Step 5:

[0484] The server retrieves price information from the approval database.

[0485] Step 6:

[0486] The server automatically generates quotes (e.g., in PDF format) based on price information and sentiment data.

[0487] Step 7:

[0488] The server saves the generated quote and creates a download link.

[0489] Step 8:

[0490] The server sends a download link to the device.

[0491] Step 9:

[0492] The device receives the download link and displays it to the user.

[0493] Step 10:

[0494] The user clicks the download link to download the quote.

[0495] Follow-up schedule management

[0496] Step 1:

[0497] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[0498] Step 2:

[0499] The device sends requests to the server that require follow-up.

[0500] Step 3:

[0501] The server analyzes the received request and determines whether follow-up is necessary.

[0502] Step 4:

[0503] The server uses an emotion engine to recognize the user's emotions.

[0504] Step 5:

[0505] The server sets the date and time for follow-up and registers it in the schedule management database.

[0506] Step 6:

[0507] The server generates a push notification as a reminder when the scheduled date and time approach.

[0508] Step 7:

[0509] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[0510] Step 8:

[0511] The device receives push notifications and displays them to the user and the person in charge.

[0512] Step 9:

[0513] Users and their representatives will take appropriate follow-up actions based on push notifications.

[0514] Specifically, the emotion engine analyzes the emotional data of user inquiries and requests, and the server operates to quickly and accurately adjust the response content if, for example, the user is in a hurry, thereby increasing the priority of follow-up. This enables more personalized customer service.

[0515] (Example 2)

[0516] 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".

[0517] Traditional in-house customer service systems have suffered from low accuracy in automated responses to customer inquiries, and have been unable to consider customer emotions when generating quotations or managing follow-up schedules. This can lead to decreased customer satisfaction and delays in response times, potentially negatively impacting a company's credibility and performance. Therefore, there is a need for highly accurate automated responses, quotation generation, and follow-up management that take customer emotions into consideration.

[0518] 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.

[0519] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for adjusting the analysis results using an emotion engine that recognizes the user's emotions. This enables a quick and appropriate response that takes customer emotions into consideration.

[0520] "Internal company data" refers to all data collected and stored within a company, including customer information, product information, inquiry history, inventory information, etc.

[0521] An "inquiry" refers to a question or request made by a customer to a company, and includes inquiries about product information, pricing, and services.

[0522] "Analysis" is the process of processing received inquiries and data using analytical tools and algorithms to extract necessary information, keywords, sentiments, and so on.

[0523] A "related database" is a database that stores data searchable based on specific criteria, and may include information on products and services, inventory information, and so on.

[0524] A "response" refers to an answer or message generated based on the analysis results and search results from related databases, and is sent to the customer in the form of email, chat message, push notification, etc.

[0525] An "emotion engine" refers to a technology that recognizes and analyzes emotions from user input text, utilizing natural language processing to determine the user's mood and tone.

[0526] A "quotation" is a document that contains price information for products or services offered to a customer, and is usually automatically generated in PDF format.

[0527] "Follow-up" refers to additional responses to the initial inquiry or request, and includes sending product samples and responding to further inquiries.

[0528] "Push notifications" are notification messages sent to a user's device in real time and are used to provide reminders and important updates.

[0529] This invention is a system that utilizes internal company data to automate responses to customer inquiries, automate the generation of quotations, manage follow-up schedules, and combines an emotion engine that recognizes user emotions. The embodiments of this system are described in detail below.

[0530] Automated response to customer inquiries

[0531] User

[0532] The user enters their inquiry from their device and clicks the send button. For example, the user might enter, "I would like to know the stock status of product A."

[0533] terminal

[0534] The terminal saves user input and sends query data to the server. This communication uses an API endpoint via an internet connection.

[0535] server

[0536] The server analyzes the received query data. This analysis uses text analysis tools such as AWS Textract. Furthermore, it uses IBM Watson® natural language processing (NLP) services to recognize the user's emotions. Based on the recognized emotion data, the content of the response message is adjusted. Next, the server searches a database such as MySQL to retrieve the necessary information. For example, it searches the inventory status of product A in the inventory database. Based on the search results, an appropriate response message is generated. The generated response message, reflecting the emotion data, is sent to the user's terminal.

[0537] Automatic generation of quotations

[0538] User

[0539] The user enters a request to issue a quote and clicks the submit button. For example, they might request, "Please issue a quote for product B."

[0540] terminal

[0541] The terminal saves the request data and sends it to the server.

[0542] server

[0543] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product. The server uses IBM Watson's NLP service to recognize sentiment and adjusts the content of the quote based on the sentiment data. The server retrieves price information from the approval database using SQL queries and generates a quote in PDF format using the ReportLab library. The generated quote is saved on the server, and a download link is created. The download link is sent to the user's terminal.

[0544] User

[0545] The user clicks the received download link to download the quote.

[0546] Follow-up schedule management

[0547] User

[0548] When a user submits an inquiry, if follow-up is needed, for example, they might request, "Please send me a sample."

[0549] terminal

[0550] The device saves requests that require follow-up and sends them to the server.

[0551] server

[0552] The server analyzes the received request and determines the need for follow-up. Furthermore, it uses IBM Watson's sentiment engine to recognize the user's emotions. Based on the recognized emotion data, it adjusts the priority and method of follow-up. The server uses the Google Calendar API to set the date and time for follow-up and registers it in the schedule management database. As the set date and time approach, it generates a push notification using Firebase Cloud Messaging and sends it to the user's and assigned personnel's devices.

[0553] terminal

[0554] Users and staff will receive push notifications and take appropriate action.

[0555] Specific example

[0556] For example, if a user sends an inquiry saying, "Please send me a sample of product C," the server analyzes the inquiry and determines that follow-up is necessary to send the sample. Furthermore, if the emotion engine detects that the user is feeling urgent, the server will set an emergency follow-up for 3 days later at 3 PM and send a push notification as a reminder. Based on these instructions, the user and the person in charge will confirm the sample delivery and follow up.

[0557] Example of a prompt

[0558] The following are examples of prompts to input into the generating AI model.

[0559] You are an engineer developing a system to automate customer service within a company. This system will include automated responses to customer inquiries, automated quotation generation, and follow-up scheduling. Please describe the appropriate processing flow and the technologies to be used to meet the following requirements.

[0560] Request: A user has submitted an inquiry about the stock status of product A. Please describe the system flow for handling this inquiry.

[0561] As described above, the present invention automates and streamlines customer service operations, enabling responses that take user emotions into consideration. As a result, higher customer satisfaction can be achieved, and the company's credibility can be improved.

[0562] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0563] Automated response to customer inquiries

[0564] Step 1:

[0565] User

[0566] The user enters their inquiry details on their device and clicks the send button.

[0567] Input: "Please tell me the stock status of product A."

[0568] Output: Query data

[0569] Step 2:

[0570] terminal

[0571] The terminal saves the user's input and sends the inquiry data to the server.

[0572] Input: Inquiry data

[0573] Output: Request to send to server

[0574] Step 3:

[0575] server

[0576] The server analyzes the received query data. AWS Textract is used for this analysis, extracting keywords from the text.

[0577] Input: Submitted inquiry data

[0578] Output: Analysis results (e.g., "Product A", "Inventory Status")

[0579] Step 4:

[0580] server

[0581] The server uses IBM Watson's NLP service to recognize the emotions contained in the query.

[0582] Input: Analysis results

[0583] Output: Sentiment data (e.g., confusion, interest)

[0584] Step 5:

[0585] server

[0586] The server generates data to adjust response messages based on recognized emotion data.

[0587] Input: Sentiment data

[0588] Output: Adjustment data

[0589] Step 6:

[0590] server

[0591] The server searches databases such as MySQL and retrieves relevant information (such as inventory information). It then executes SQL queries to extract the necessary data.

[0592] Input: Query to search for "Inventory information for product A"

[0593] Output: Inventory information (e.g., In stock)

[0594] Step 7:

[0595] server

[0596] The server generates an appropriate response message based on inventory information and sentiment data.

[0597] Input: Inventory information, adjustment data

[0598] Output: Response message (Example: "Thank you. Product A is currently in stock.")

[0599] Step 8:

[0600] server

[0601] The server sends the generated response message to the user's terminal.

[0602] Input: Response message

[0603] Output: Notification to user terminal

[0604] Automatic generation of quotations

[0605] Step 1:

[0606] User

[0607] The user enters a request to request a quote and clicks the submit button.

[0608] Input: "Please issue a quotation for product B."

[0609] Output: Request data

[0610] Step 2:

[0611] terminal

[0612] The terminal saves the request data and sends it to the server.

[0613] Input: Request data

[0614] Output: Request to send to server

[0615] Step 3:

[0616] server

[0617] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product.

[0618] Input: Request data

[0619] Output: Analysis results (e.g., target company, product information)

[0620] Step 4:

[0621] server

[0622] The server uses IBM Watson's NLP service to recognize the user's emotions.

[0623] Input: Analysis results

[0624] Output: Sentiment data (e.g., interest, expectation)

[0625] Step 5:

[0626] server

[0627] The server generates data to adjust the contents of the estimate based on the recognized emotion data.

[0628] Input: Sentiment data

[0629] Output: Adjustment data

[0630] Step 6:

[0631] server

[0632] The server retrieves price information from the approval database using SQL queries.

[0633] Input: Query to search for price information for "Product B"

[0634] Output: Price information (e.g., ¥100,000)

[0635] Step 7:

[0636] server

[0637] The server automatically generates a quotation in PDF format using the ReportLab library.

[0638] Input: Price information, adjustment data

[0639] Output: Quotation PDF (Example: quotate_2023.pdf)

[0640] Step 8:

[0641] server

[0642] The generated quotation is saved to the server, and a download link is generated.

[0643] Input: Quotation PDF

[0644] Output: Download link (Example: https: / / example.com / downloads / quote_2023.pdf)

[0645] Step 9:

[0646] server

[0647] The server sends the generated download link to the user's device.

[0648] Input: Download link

[0649] Output: Notification to user terminal

[0650] Step 10:

[0651] User

[0652] The user clicks the received download link to download the quote.

[0653] Input: Download link

[0654] Output: Download quotation

[0655] Follow-up schedule management

[0656] Step 1:

[0657] User

[0658] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[0659] Input: "Please send a sample."

[0660] Output: Follow-up request

[0661] Step 2:

[0662] terminal

[0663] The device saves requests that require follow-up and sends them to the server.

[0664] Input: Follow-up request

[0665] Output: Request to send to server

[0666] Step 3:

[0667] server

[0668] The server analyzes the received request and determines whether follow-up is necessary.

[0669] Input: Follow-up request

[0670] Output: Analysis results (e.g., "Sample submission required")

[0671] Step 4:

[0672] server

[0673] The server uses IBM Watson's emotion engine to recognize the user's emotions.

[0674] Input: Analysis results

[0675] Output: Sentimental data (e.g., urgency)

[0676] Step 5:

[0677] server

[0678] The server adjusts the priority and method of follow-up based on the recognized sentiment data.

[0679] Input: Sentiment data

[0680] Output: Adjustment data

[0681] Step 6:

[0682] server

[0683] The server uses the Google Calendar API to set the date and time that requires follow-up and registers it in the schedule management database.

[0684] Input: Adjustment data

[0685] Output: Follow-up schedule (Example: 3 days from now, 3 PM)

[0686] Step 7:

[0687] server

[0688] As the scheduled date and time approach, Firebase Cloud Messaging is used to generate a push notification as a reminder, which is then sent to the user and the responsible party.

[0689] Input: Follow-up schedule

[0690] Output: Push notification

[0691] Step 8:

[0692] terminal

[0693] Users and staff will receive push notifications and take appropriate action.

[0694] Input: Push notification

[0695] Output: Follow-up available

[0696] (Application Example 2)

[0697] 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".

[0698] In modern society, companies are required to improve the efficiency and quality of their customer service operations. In particular, electronic payment services demand prompt and appropriate customer service, while simultaneously making it difficult to consider customer emotions. Conventional systems could only provide mechanical responses to inquiries, making it difficult to increase customer satisfaction. Furthermore, tasks such as generating quotations and managing follow-ups required manual handling, resulting in a significant workload. This invention aims to solve these problems by providing a system that recognizes customer emotions and responds accordingly.

[0699] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries and user sentiment data, means for automatically generating and sending a response that reflects the sentiment based on the search results and sentiment data to the customer, and means for automatically generating an electronic receipt and providing the customer with a downloadable link. This makes it possible to provide a quick and appropriate response while taking customer sentiment into consideration.

[0700] "Internal company data" refers to all information generated and stored within a company, including customer data, product data, inventory data, and sales history.

[0701] "Customer inquiries" refer to requests for information, questions, and feedback that customers send to a company, and include those sent via telephone, email, web forms, etc.

[0702] "Analysis" refers to the process of deciphering and understanding received data, and performing operations to grasp its intent and meaning.

[0703] A "related database" refers to a collection of data that stores specific information and is structured in a way that allows for later searching and referencing.

[0704] "Emotional data" refers to data that identifies a user's emotional state (e.g., joy, anger, surprise, etc.) using natural language processing and emotion recognition technologies, and expresses it as numerical values ​​or tags.

[0705] "Automatically generating responses" refers to a system generating reply text without human intervention, based on pre-configured algorithms and rules.

[0706] An "electronic receipt" refers to a receipt issued in digital format, usually generated in PDF format, and provided via email or a download link.

[0707] A "quotation" is a document that shows the price and conditions of the products or services to be offered in advance, and serves as a basis for customers to consider purchasing.

[0708] A "downloadable link" refers to a URL (web address) from which a specific file or information can be obtained via the internet.

[0709] "Follow-up" refers to additional support and verification work carried out continuously after the initial response, and is a process aimed at maintaining and improving customer satisfaction.

[0710] "Push notifications as reminders" refers to displaying short messages on a user's smartphone or other device to alert them based on a specified date, time, or conditions.

[0711] This invention relates to a system for automating and streamlining customer service in electronic payment services. In particular, it aims to improve customer satisfaction by recognizing customer emotions and providing appropriate responses and follow-ups.

[0712] Hardware configuration

[0713] This invention is carried out using the following hardware:

[0714] Smartphone (iOS or Android®)

[0715] Cloud servers (e.g., AWS or Google Cloud)

[0716] Software Configuration

[0717] This invention is carried out using the following software:

[0718] Smartphone applications (iOS: Swift, Android: Kotlin)

[0719] Server-side: Node.js, Python

[0720] Database: MySQL or PostgreSQL

[0721] Emotion recognition engine: IBM Watson Tone Analyzer or Google Cloud Natural Language API

[0722] Push notification service: Firebase Cloud Messaging (FCM)

[0723] Overview of Data Processing and Data Calculation

[0724] 1. User submits inquiry:

[0725] The user enters an inquiry about electronic payments via a smartphone app and clicks the submit button. The inquiry data is sent from the device to the server.

[0726] 2. Reception and analysis on the server:

[0727] The server receives the query data and extracts keywords using natural language processing techniques. Furthermore, it analyzes the user's sentiment data using an emotion recognition engine (IBM Watson Tone Analyzer or Google Cloud Natural Language API).

[0728] 3. Generating and sending response messages:

[0729] The server searches the electronic payment database for corresponding information based on the analysis results and sentiment data. It then automatically generates a response message that reflects the sentiment and sends it to the user's smartphone.

[0730] 4. Automatic generation of electronic receipts:

[0731] When a user completes a purchase, the server automatically generates an electronic receipt in PDF format. It then generates a download link for the electronic receipt and provides it to the user.

[0732] 5. Follow-up schedule management:

[0733] The server prioritizes follow-ups based on user sentiment data. The cloud server manages the dates and times when follow-ups are needed, and push notifications are sent as reminders to the user's and the assigned staff member's devices as the scheduled time approaches.

[0734] Specific example

[0735] For example, if a user sends an inquiry asking, "Please tell me my recent transaction history," the server uses its emotion recognition engine to recognize the user's emotion as "high stress." Next, the server retrieves the corresponding transaction history from the database and generates and sends a response message in polite language to alleviate the stress, such as, "Here is your transaction history. Please feel free to contact us anytime if you have any further questions."

[0736] Example of a prompt

[0737] "Please tell me your most recent credit card transaction history."

[0738] "Please resend the most recent receipt via email."

[0739] "I'd like to confirm the payment details."

[0740] This invention makes it possible to recognize user emotions and respond accordingly. This system not only increases customer satisfaction but also streamlines a company's customer service operations.

[0741] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0742] Step 1:

[0743] The user enters their inquiry and clicks the submit button. The "inquiry details" are retrieved from the user's device as input data. This inquiry is then sent to the server via the smartphone.

[0744] Step 2:

[0745] The system analyzes query data received by the server. The input is the query data, and keywords are extracted using a natural language processing tool (e.g., Python's NLTK). The output is the extracted keywords.

[0746] Step 3:

[0747] The server uses an emotion recognition engine to analyze the user's emotional data. The input is "data from the inquiry," and the emotions are analyzed using an emotion recognition tool (e.g., IBM Watson Tone Analyzer). The output is "recognized emotional data."

[0748] Step 4:

[0749] The server performs a search of relevant databases based on the analyzed keywords and sentiment data. The inputs are "extracted keywords" and "recognized sentiment data," and the server queries the databases. The output is the "search results."

[0750] Step 5:

[0751] The server generates a response message that reflects emotions based on search results and sentiment data. The inputs are "search results" and "recognized sentiment data," and the server generates the response message using natural language generation technology (e.g., the GPT model). The output is the "response message."

[0752] Step 6:

[0753] The server sends the generated response message to the user's terminal. The input is the "response message," and the output is the "response message sent to the user's terminal."

[0754] Step 7:

[0755] When a user completes a purchase, the server automatically generates an electronic receipt. The input is "purchase data," and the electronic receipt is generated using a PDF generation tool. The output is a "PDF file of the electronic receipt."

[0756] Step 8:

[0757] The server generates a download link for the electronic receipt and provides it to the user. The input is a "PDF file of the electronic receipt," and the output is a "download link for the electronic receipt."

[0758] Step 9:

[0759] The server manages the schedule for follow-ups when necessary. Inputs are "follow-up request data" and "recognized sentiment data," and the server sets the follow-up priority. The output is the "configured follow-up schedule."

[0760] Step 10:

[0761] The server generates a push notification as a reminder as the scheduled date and time approaches. The input is the "follow-up schedule data," and the output is the "generated push notification." The notification is sent to the user's and the assigned person's devices.

[0762] 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.

[0763] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0764] 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.

[0765] [Second Embodiment]

[0766] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0767] 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.

[0768] 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).

[0769] 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.

[0770] 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.

[0771] 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).

[0772] 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.

[0773] 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.

[0774] 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.

[0775] 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.

[0776] 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.

[0777] 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".

[0778] This invention is a system that utilizes internal company data to centrally manage automated responses to customer inquiries, automated generation of quotations, and follow-up schedules.

[0779] 1. Automated response to customer inquiries

[0780] User

[0781] 1. The user enters their inquiry details on their device and clicks the send button.

[0782] 2. For example, a user enters "I would like to know the stock status of product A."

[0783] terminal

[0784] 1. The terminal saves the user's input and sends the inquiry data to the server.

[0785] server

[0786] 1. The server analyzes the received query data.

[0787] 2. Based on the analysis results, search for relevant databases (e.g., inventory databases).

[0788] 3. Automatically generate a response message based on the search results (e.g., availability of product A).

[0789] 4. Send the generated response message to the user's terminal.

[0790] 2. Automatic generation of quotations

[0791] User

[0792] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[0793] 2. For example, you could request, "Please issue a quotation for product B."

[0794] terminal

[0795] 1. The terminal saves the request data and sends it to the server.

[0796] server

[0797] 1. The server parses the received request.

[0798] 2. Search the approval database for price data of the target company and product.

[0799] 3. Automatically generate a quotation based on the search results. (Example: PDF format)

[0800] 4. Save the generated quotation and create a download link.

[0801] 5. Send the download link to the user's device.

[0802] User

[0803] 1. The user clicks the received download link to download the quotation.

[0804] 3. Schedule management for follow-up

[0805] User

[0806] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[0807] terminal

[0808] 1. The device sends requests requiring follow-up to the server.

[0809] server

[0810] 1. The server analyzes the received request and determines whether follow-up is necessary.

[0811] 2. Set the date and time for follow-up and register it in the schedule management database.

[0812] 3. Generate a push notification as a reminder based on the set date and time.

[0813] 4. Send push notifications to the user's and the person in charge's devices.

[0814] terminal

[0815] 1. Users and staff will receive push notifications and take appropriate action.

[0816] Specific example

[0817] For example, consider a case where a user submits an inquiry requesting "Please send me a sample of product C." In this case, the server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a schedule for follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and, based on the instructions, confirm the sample shipment and follow up.

[0818] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and schedule management to automate and streamline customer service operations, thereby achieving fast and accurate customer service.

[0819] The following describes the processing flow.

[0820] Automated response to customer inquiries

[0821] Step 1:

[0822] The user enters their inquiry details on their device and clicks the send button.

[0823] Step 2:

[0824] The terminal saves the user's input and sends the inquiry data to the server.

[0825] Step 3:

[0826] The server analyzes the received query data and extracts keywords and related information.

[0827] Step 4:

[0828] Based on the analysis results, the server searches relevant databases (e.g., inventory databases).

[0829] Step 5:

[0830] The server generates an appropriate response message based on the search results.

[0831] Step 6:

[0832] The server sends the generated response message to the terminal.

[0833] Step 7:

[0834] The terminal receives a response from the server and displays it to the user.

[0835] Automatic generation of quotations

[0836] Step 1:

[0837] The user enters a request to request a quote and clicks the submit button.

[0838] Step 2:

[0839] The terminal saves the request data and sends it to the server.

[0840] Step 3:

[0841] The server analyzes the received request and extracts price data for the target company and product.

[0842] Step 4:

[0843] The server retrieves price information from the approval database.

[0844] Step 5:

[0845] The server automatically generates a quote based on the price information (e.g., in PDF format).

[0846] Step 6:

[0847] The server saves the generated quote and creates a download link.

[0848] Step 7:

[0849] The server sends a download link to the device.

[0850] Step 8:

[0851] The device receives the download link and displays it to the user.

[0852] Step 9:

[0853] The user clicks the download link to download the quote.

[0854] Follow-up schedule management

[0855] Step 1:

[0856] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[0857] Step 2:

[0858] The device sends requests to the server that require follow-up.

[0859] Step 3:

[0860] The server analyzes the received request and determines whether follow-up is necessary.

[0861] Step 4:

[0862] The server sets the date and time for follow-up and registers it in the schedule management database.

[0863] Step 5:

[0864] The server generates a push notification as a reminder when the scheduled date and time approach.

[0865] Step 6:

[0866] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[0867] Step 7:

[0868] The device receives push notifications and displays them to the user and the person in charge.

[0869] Step 8:

[0870] Users and their representatives will take appropriate follow-up actions based on push notifications.

[0871] (Example 1)

[0872] 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".

[0873] In corporate customer service operations, traditional manual responses are time-consuming and labor-intensive, leading to decreased customer satisfaction. In particular, efficiently managing inquiries, issuing quotations, and follow-ups is difficult, and there is a demand for quick and accurate responses. Furthermore, the lack of a unified management system for these tasks necessitates information integration and automation.

[0874] 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.

[0875] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for generating response messages using a generation AI model. This enables rapid and accurate automated responses to customer inquiries.

[0876] In this invention, the server includes means for searching a database of prices offered by each company and automatically generating a quotation in a format accessible to the customer; means for generating a downloadable link for the generated quotation and providing it to the customer; means for downloading the quotation via the provided link; and means for adjusting the content of the quotation using a generation AI model. This enables the rapid and automatic creation and provision of quotations.

[0877] In this invention, the server includes means for tracking the processing status of customer inquiries and setting a date and time when follow-up is required; means for sending push notifications as reminders to customers and personnel based on the set date and time; means for receiving push notifications; and means for generating the content of the reminder using a prompt statement. This enables efficient scheduling of follow-ups and timely responses to be taken.

[0878] These measures enable companies to automate and streamline customer service operations, contributing to improved customer satisfaction.

[0879] "Internal company data" refers to information and documents generated and stored within a company.

[0880] "Customer inquiries" refer to information about questions and requests that customers make to a company.

[0881] "Means of receiving and analyzing data" refers to the technology that allows a system to take in data from an external source and understand its contents.

[0882] A "related database" refers to a database that stores information related to a specific query.

[0883] "Searching methods" refer to techniques for finding information within a database based on specific criteria.

[0884] "Means of generating responses" refers to technologies that automatically create appropriate answers to inquiries.

[0885] A "generative AI model" refers to an algorithm that uses artificial intelligence to generate new information from data.

[0886] A "downloadable link" refers to a web address that allows a user to download a specific file by clicking on it.

[0887] A "quote" refers to a document that lists the price, details, and terms and conditions of a product or service.

[0888] "Means of tracking" refers to technologies that track and record the progress of a specific process or event.

[0889] "Follow-up" refers to additional actions or confirmations taken after the initial response or action.

[0890] "Push notifications" refer to a technology that sends information from a server to a device in real time.

[0891] A "prompt statement" refers to an input statement used to instruct a generative AI model to generate information.

[0892] This invention is a system that automates and streamlines customer service operations by utilizing internal company data. The program processing of this system is described in detail below.

[0893] Automated response to customer inquiries

[0894] 1. User

[0895] The user enters their inquiry details on their device and clicks the "Send" button.

[0896] For example, you would enter an inquiry such as, "Please tell me the stock status of product A."

[0897] 2. Terminal

[0898] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[0899] 3. Server

[0900] The server parses the received query data and uses natural language processing libraries (e.g., spaCy or NLTK) to analyze the information the user is seeking.

[0901] Based on the analysis results, the system searches relevant databases (e.g., inventory databases) and issues SQL queries.

[0902] Based on the search results, a prompt is sent to a generative AI model (e.g., OpenAI's GPT-4) to generate a response message. Example prompt: "How much stock do you currently have of product A?"

[0903] The generated response message is sent to the user's terminal in JSON format.

[0904] 4. Terminal

[0905] The terminal displays the received response message on the user interface.

[0906] Automatic generation of quotations

[0907] 1. User

[0908] The user enters the request for a quote and clicks the "Submit" button.

[0909] For example, you might enter, "Please issue a quotation for product B."

[0910] 2. Terminal

[0911] The terminal saves the request data to a local database and then sends it to the server.

[0912] 3. Server

[0913] The server analyzes the request data and extracts the necessary information using a natural language processing library.

[0914] Search the price database and issue an SQL query. Example SQL query: "SELECT price FROM products WHERE product_name='product B'"

[0915] Based on the search results, the system automatically generates a quotation in PDF format using libraries such as Apache PDFBox.

[0916] The generated quote is saved to cloud storage (e.g., AWS S3), and a downloadable link is generated and sent to the user's device.

[0917] 4. User

[0918] The user clicks the provided link and downloads the quote.

[0919] Follow-up schedule management

[0920] 1. User

[0921] The user enters a request that requires follow-up and clicks the "Send" button.

[0922] For example, you might enter, "Please send me a sample of product C."

[0923] 2. Terminal

[0924] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[0925] 3. Server

[0926] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[0927] Set the necessary follow-up dates and register them in the schedule management database (e.g., Microsoft SQL Server).

[0928] Using Firebase Cloud Messaging, push notifications are generated at a set date and time and sent to the user's and the assigned person's devices.

[0929] 4. Terminal

[0930] Users and staff members receive push notifications and follow up accordingly.

[0931] Hardware and software used

[0932] Server: Cloud server (e.g., AWS EC2, Microsoft Azure)

[0933] Database management systems: MySQL, Oracle, Microsoft SQL Server

[0934] Natural language processing libraries: spaCy, NLTK

[0935] Generative AI models: OpenAI GPT-4, etc.

[0936] Cloud storage: AWS S3

[0937] PDF generation library: Apache PDFBox

[0938] Push notification service: Firebase Cloud Messaging

[0939] Specific example

[0940] For example, if a user submits an inquiry requesting "Please send me a sample of product C," the following process takes place: The server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a follow-up appointment in the schedule management database for three days later and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and confirm and carry out the sample shipment based on the instructions.

[0941] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and follow-up schedule management to automate and streamline customer service operations, thereby achieving prompt and accurate customer service.

[0942] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0943] Processing steps for automated responses to customer inquiries

[0944] Step 1:

[0945] User

[0946] The user enters their inquiry details on their device and clicks the "Send" button.

[0947] Input: User inquiry (e.g., "Please tell me the stock status of product A")

[0948] Output: A request is generated to send query data to the terminal.

[0949] Step 2:

[0950] terminal

[0951] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[0952] Input: User inquiry data

[0953] Output: Send query data to the server in JSON format.

[0954] Step 3:

[0955] server

[0956] The server analyzes the received query data and uses a natural language processing library to understand the intent of the query.

[0957] Software used: spaCy, NLTK

[0958] Input: Query data in JSON format

[0959] Data processing: Text analysis using natural language processing

[0960] Output: Analysis results of the inquiry

[0961] Step 4:

[0962] server

[0963] Based on the analysis results, SQL queries are issued to relevant databases (e.g., inventory databases) to perform searches.

[0964] Database management system used: MySQL

[0965] Input: Analysis results of the inquiry content

[0966] Data Calculation: Searching Related Databases

[0967] Output: Database search results (e.g., "Inventory quantity of product A")

[0968] Step 5:

[0969] server

[0970] A response message is generated using an AI model based on the database search results.

[0971] Generative AI model used: OpenAI GPT-4

[0972] Input: Database search results

[0973] Data processing: Generating response messages using a generative AI model.

[0974] Output: Response message (Example: "Product A is in stock")

[0975] Step 6:

[0976] server

[0977] The server sends the generated response message to the user's terminal in JSON format.

[0978] Input: Response message

[0979] Output: Send a JSON-formatted response message to the user's terminal.

[0980] Step 7:

[0981] terminal

[0982] The terminal displays the received response message on the user interface.

[0983] Input: Response message in JSON format

[0984] Output: Display in the user interface

[0985] Process steps for automatic quotation generation

[0986] Step 1:

[0987] User

[0988] The user enters the request for a quote and clicks the "Submit" button.

[0989] Input: Product name and quotation request (Example: "Please issue a quotation for product B")

[0990] Output: Data is input to the terminal, and a transmission command is generated.

[0991] Step 2:

[0992] terminal

[0993] The terminal saves the request data to a local database and then sends it to the server.

[0994] Input: User's request data

[0995] Output: Send request data to the server in JSON format.

[0996] Step 3:

[0997] server

[0998] The server parses the request data and searches the database to retrieve relevant pricing information.

[0999] Software used: Natural language processing libraries (spaCy, NLTK)

[1000] Input: Request data

[1001] Data processing: Text analysis using natural language processing

[1002] Output: Analyzed request content

[1003] Step 4:

[1004] server

[1005] Based on the parsed request, an SQL query is issued to the price database to retrieve the necessary price information.

[1006] Database management system used: Oracle

[1007] Input: Parsed request content

[1008] Data Calculation: Searching Price Databases

[1009] Output: Price data (Example: "Price information for product B")

[1010] Step 5:

[1011] server

[1012] Based on the acquired price data, an AI model is used to automatically generate a quotation.

[1013] Generative AI model used: OpenAI GPT-4

[1014] PDF generation library used: Apache PDFBox

[1015] Input: Price data

[1016] Data calculation: Generating a PDF quotation

[1017] Output: Generated quotation (PDF format)

[1018] Step 6:

[1019] server

[1020] The generated quote is saved to cloud storage, and a downloadable link is created and sent to the user's device.

[1021] Cloud storage to use: AWS S3

[1022] Input: Generated quotation (PDF format)

[1023] Output: Download link

[1024] Step 7:

[1025] User

[1026] The user clicks the provided link and downloads the quote.

[1027] Input: Download link

[1028] Output: Downloaded quotation (PDF format)

[1029] Follow-up scheduling process steps

[1030] Step 1:

[1031] User

[1032] The user enters a request that requires follow-up and clicks the "Send" button.

[1033] Input: Follow-up request (Example: "Please send a sample of product C")

[1034] Output: Data is input to the terminal, and a transmission command is generated.

[1035] Step 2:

[1036] terminal

[1037] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[1038] Input: Request data requiring follow-up

[1039] Output: Send request data to the server in JSON format.

[1040] Step 3:

[1041] server

[1042] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[1043] Software used: Natural language processing libraries (spaCy, NLTK)

[1044] Input: Follow-up request data

[1045] Data processing: Text analysis using natural language processing

[1046] Output: Result of the assessment of the need for follow-up.

[1047] Step 4:

[1048] server

[1049] Set the date and time for follow-up and register it in the schedule management database.

[1050] Database management system used: Microsoft SQL Server

[1051] Input: Result of the assessment of the need for follow-up.

[1052] Data processing: Setting and registering follow-up dates and times.

[1053] Output: Configured follow-up schedule

[1054] Step 5:

[1055] server

[1056] Based on the configured date and time, push notifications are generated using Firebase Cloud Messaging and sent to the user's and assigned personnel's devices.

[1057] Push notification service used: Firebase Cloud Messaging

[1058] Input: Configured follow-up schedule

[1059] Data processing: Generating and sending push notifications

[1060] Output: Push notifications to user and staff terminals

[1061] Step 6:

[1062] terminal

[1063] Users and staff members receive push notifications and follow up accordingly.

[1064] Input: Push notification

[1065] Output: Follow-up will be performed.

[1066] Through the specific processing steps outlined above, the system can automate and streamline customer service operations.

[1067] (Application Example 1)

[1068] 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."

[1069] Traditional e-commerce sites struggled to respond quickly and accurately to customer inquiries, particularly with manual processes like checking inventory and issuing quotes, resulting in inefficiencies. Furthermore, managing follow-up schedules was cumbersome, often leading to delays in notifications to both customers and staff. This resulted in decreased customer satisfaction and missed sales opportunities. There is a need to solve these problems and automate and streamline customer service.

[1070] 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.

[1071] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data; means for searching relevant databases based on the analyzed inquiries; means for automatically generating and sending responses to customers based on the search results; means for analyzing customer inquiries regarding specific product information or inventory status using natural language processing; means for performing natural language processing using a generative AI model to automatically generate appropriate responses; means for searching each company's pricing database and automatically generating quotations in a format accessible to customers; means for generating and providing downloadable links for the generated quotations to customers; means for receiving quotation requests, analyzing product information and pricing data using a generative AI model, and automatically generating quotations; means for tracking the processing status of customer inquiries and setting dates and times for follow-up; means for sending push notifications to customers and personnel as reminders based on the set dates and times; and means for using a schedule management database to automatically generate reminders when follow-up is required. This enables not only a quick and accurate response to customer inquiries but also efficient automatic quotation generation and follow-up schedule management.

[1072] "Internal company data" refers to the collection of all information and data managed within a company.

[1073] An "inquiry" refers to a question or request sent by a customer to confirm product information, stock availability, etc.

[1074] "Analysis" is the process of breaking down the content of an received inquiry and interpreting its meaning and intent.

[1075] A "database" is a system that stores and manages related information in an organization.

[1076] "Searching" is the process of finding specific information within a database.

[1077] "Response" refers to the answers or information provided in response to customer inquiries.

[1078] "Natural language processing" is the technology that enables computers to understand and process human language.

[1079] A "generative AI model" refers to an algorithm or framework that uses artificial intelligence to generate new text or data.

[1080] A "price database" is a database where product price information is stored in an organ.

[1081] A "quotation" is a document that shows the price and conditions of a specific product or service.

[1082] "Follow-up" refers to additional confirmations or follow-ups conducted after the initial response.

[1083] A "schedule management database" is a system that organizes and stores follow-up dates and appointments.

[1084] A "reminder" is a message sent to notify you of an action that needs to be taken at a specific date and time.

[1085] A "push notification" is a notification message that is sent to a user's device in real time.

[1086] The specific system for implementing the present invention is capable of efficiently and quickly processing customer inquiries and easily managing the automatic generation of quotations and follow-up schedules. This system is configured as follows.

[1087] Automated response to customer inquiries

[1088] The server automatically receives and analyzes customer inquiries using internal company data. It receives the inquiry content entered by the customer on their device and sends it to the server. The server uses Google Cloud AI to perform natural language processing and analyze the inquiry content. It then searches relevant databases (e.g., product inventory database) and automatically generates an appropriate response. The generated response is sent to the customer's device.

[1089] Specific example

[1090] For example, if a customer asks, "What is the stock status of product A?", the server analyzes the inquiry, retrieves the stock status of product A from the product inventory database, and automatically generates a response message saying, "Product A is currently in stock."

[1091] Example of a prompt:

[1092] Inquiry: "Please tell me the stock status of product A."

[1093] Generated AI prompt: "Check the inventory status of product A and respond to the customer whether it is in stock or not."

[1094] Automatic generation of quotations

[1095] When a customer requests a quote for a specific product, the request data is sent from the terminal to the server. The server receives and analyzes the request. Next, it retrieves the price data for the product from the price database and automatically generates a quote using a generation AI model. The generated quote is saved in PDF format, and a download link is sent to the customer.

[1096] Specific example

[1097] When a customer requests "Please issue a quote for product B," the server analyzes the request, retrieves the price information for product B from the price database, and automatically generates a quote. A download link for the generated quote is then sent to the customer's device.

[1098] Example of a prompt:

[1099] Quote Request: "Please issue a quote for product B."

[1100] Generated AI prompt: "Retrieve pricing data for product B, generate a quote, and provide it to the customer."

[1101] Follow-up schedule management

[1102] If a customer requires follow-up on an inquiry, they send a follow-up request from their device to the server. The server parses this request and sets a date and time for the follow-up. This information is registered in the schedule management database, and a push notification is generated as a reminder based on the set date and time. This push notification is sent to the customer's and the agent's devices.

[1103] Specific example

[1104] For example, if a customer requests a sample of product C, the server analyzes the request and determines that follow-up is necessary. It sets a schedule for a follow-up three days later and registers it in the schedule management database. Three days later, a push notification is automatically sent as a reminder.

[1105] Example of a prompt:

[1106] Follow-up request: "Please send me a sample of product C."

[1107] Generated AI prompt: "Please schedule a follow-up for sending a sample of Product C in 3 days and send a reminder."

[1108] As described above, the present invention is a system that enables rapid and accurate responses to customer inquiries, automatic generation of quotations, and efficient scheduling of follow-ups. This improves customer satisfaction and streamlines operations.

[1109] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1110] Processing steps for automated responses to customer inquiries

[1111] Step 1:

[1112] The user enters their inquiry details on their device and clicks the send button. An example of input data is, "Please tell me the stock status of product A." Input: Inquiry details. Output: Sending of inquiry data.

[1113] Step 2:

[1114] The terminal saves user input and sends query data to the server. Network communication is performed for data transfer. Input: Query data. Output: Data sent to the server.

[1115] Step 3:

[1116] The server receives query data, which is then processed using natural language processing and analysis with Google Cloud AI. Input: Query data. Output: Analysis results.

[1117] Step 4:

[1118] The server searches relevant databases (e.g., inventory databases) based on the analysis results and retrieves the necessary information. Input: Analysis results. Output: Search results.

[1119] Step 5:

[1120] The server automatically generates response messages based on search results, using a generation AI model. Input: Search results. Output: Response message.

[1121] Step 6:

[1122] The server generates a response message and sends it to the user's terminal. Input: Response message. Output: Message sent to the user's terminal.

[1123] Process steps for automatic quotation generation

[1124] Step 1:

[1125] The user requests a quotation and clicks the send button on their terminal. Example input: "Please issue a quotation for product B." Input: Quotation request. Output: Quotation request data sent.

[1126] Step 2:

[1127] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Estimate request data. Output: Data sent to the server.

[1128] Step 3:

[1129] The server analyzes the request and uses a generated AI model to retrieve product information and pricing data. Input: Quotation request data. Output: Analysis results and pricing information.

[1130] Step 4:

[1131] The server automatically generates a quotation based on the analysis results and saves it in PDF format. Input: Analysis results and price information. Output: Quotation in PDF format.

[1132] Step 5:

[1133] The server generates a download link for the generated quotation and sends it to the user's device. Input: Quotation in PDF format. Output: Sending of download link.

[1134] Follow-up scheduling process steps

[1135] Step 1:

[1136] The user enters a request requiring follow-up and clicks the send button on their device. Example input: "Please send a sample of product C." Input: Follow-up request. Output: Sending of follow-up request data.

[1137] Step 2:

[1138] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Follow-up request data. Output: Data sent to the server.

[1139] Step 3:

[1140] The server analyzes the received request data and sets the date and time for follow-up. Input: Follow-up request data. Output: Follow-up schedule.

[1141] Step 4:

[1142] The server registers the follow-up date and time in the schedule management database. Input: Follow-up schedule. Output: Schedule registration information.

[1143] Step 5:

[1144] The server generates a push notification as a reminder based on the configured date and time. Input: Follow-up schedule. Output: Push notification.

[1145] Step 6:

[1146] The server generates push notifications and sends them to the user's and the assigned personnel's devices. Input: Push notification. Output: Notification sent to the user's and the assigned personnel's devices.

[1147] 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.

[1148] This invention is a system that utilizes internal company data to automatically respond to customer inquiries, automatically generate quotations, manage follow-up schedules, and combines this with an emotion engine that recognizes user emotions.

[1149] 1. Automated response to customer inquiries

[1150] User

[1151] 1. The user enters their inquiry details on their device and clicks the send button.

[1152] 2. For example, a user enters "I would like to know the stock status of product A."

[1153] terminal

[1154] 1. The terminal saves the user's input and sends the inquiry data to the server.

[1155] server

[1156] 1. The server analyzes the received query data and extracts keywords and related information.

[1157] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[1158] 3. Based on the recognized emotions, generate data to adjust the response message.

[1159] 4. Based on the analysis results and sentiment data, search relevant databases (e.g., inventory databases).

[1160] 5. Based on the search results, generate appropriate response messages that reflect sentiment data.

[1161] 6. Send the generated response message to the user's terminal.

[1162] 2. Automatic generation of quotations

[1163] User

[1164] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[1165] 2. For example, you could request, "Please issue a quotation for product B."

[1166] terminal

[1167] 1. The terminal saves the request data and sends it to the server.

[1168] server

[1169] 1. The server analyzes the received request and extracts price data for the target company and product.

[1170] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[1171] 3. Based on the recognized emotion data, generate data to adjust the contents of the quotation.

[1172] 4. Retrieve price information from the approval database.

[1173] 5. Automatically generate quotes based on price information and sentiment data (e.g., in PDF format).

[1174] 6. Save the generated quotation and create a download link.

[1175] 7. Send the download link to the user's device.

[1176] User

[1177] 1. The user clicks the received download link to download the quotation.

[1178] 3. Schedule management for follow-up

[1179] User

[1180] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[1181] terminal

[1182] 1. The device sends requests requiring follow-up to the server.

[1183] server

[1184] 1. The server analyzes the received request and determines whether follow-up is necessary.

[1185] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[1186] 3. Based on the recognized emotional data, adjust the priorities and methods of follow-up.

[1187] 4. Set the date and time for follow-up and register it in the schedule management database.

[1188] 5. When the set date and time approach, generate a push notification as a reminder.

[1189] 6. Send the generated push notification to the user's and the person in charge's devices.

[1190] terminal

[1191] 1. Users and staff will receive push notifications and take appropriate action.

[1192] Specific example

[1193] For example, consider a case where a user submits an inquiry requesting a sample of product C. In this case, the server analyzes the inquiry and determines that follow-up is necessary to send the sample. It also uses an emotion engine to recognize if the user feels a sense of urgency. The server sets a schedule for a follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge then confirm the sample delivery and follow up based on these instructions.

[1194] As described above, the system of the present invention can automate and streamline customer service operations, and provide a better customer experience through responses and methods that take into account the user's emotions.

[1195] The following describes the processing flow.

[1196] Automated response to customer inquiries

[1197] Step 1:

[1198] The user enters their inquiry details on their device and clicks the send button.

[1199] Step 2:

[1200] The terminal saves the user's input and sends the inquiry data to the server.

[1201] Step 3:

[1202] The server analyzes the received query data and extracts keywords and related information.

[1203] Step 4:

[1204] The server uses an emotion engine to recognize the user's emotions.

[1205] Step 5:

[1206] The server searches relevant databases (e.g., inventory databases) based on the analysis results and sentiment data.

[1207] Step 6:

[1208] The server generates a response message based on the search results.

[1209] Step 7:

[1210] The server incorporates sentiment data into the generated response message and makes appropriate adjustments.

[1211] Step 8:

[1212] The server sends the generated response message to the terminal.

[1213] Step 9:

[1214] The terminal receives a response from the server and displays it to the user.

[1215] Automatic generation of quotations

[1216] Step 1:

[1217] The user enters a request to request a quote and clicks the submit button.

[1218] Step 2:

[1219] The terminal saves the request data and sends it to the server.

[1220] Step 3:

[1221] The server analyzes the received request and extracts price data for the target company and product.

[1222] Step 4:

[1223] The server uses an emotion engine to recognize the user's emotions.

[1224] Step 5:

[1225] The server retrieves price information from the approval database.

[1226] Step 6:

[1227] The server automatically generates quotes (e.g., in PDF format) based on price information and sentiment data.

[1228] Step 7:

[1229] The server saves the generated quote and creates a download link.

[1230] Step 8:

[1231] The server sends a download link to the device.

[1232] Step 9:

[1233] The device receives the download link and displays it to the user.

[1234] Step 10:

[1235] The user clicks the download link to download the quote.

[1236] Follow-up schedule management

[1237] Step 1:

[1238] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[1239] Step 2:

[1240] The device sends requests to the server that require follow-up.

[1241] Step 3:

[1242] The server analyzes the received request and determines whether follow-up is necessary.

[1243] Step 4:

[1244] The server uses an emotion engine to recognize the user's emotions.

[1245] Step 5:

[1246] The server sets the date and time for follow-up and registers it in the schedule management database.

[1247] Step 6:

[1248] The server generates a push notification as a reminder when the scheduled date and time approach.

[1249] Step 7:

[1250] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[1251] Step 8:

[1252] The device receives push notifications and displays them to the user and the person in charge.

[1253] Step 9:

[1254] Users and their representatives will take appropriate follow-up actions based on push notifications.

[1255] Specifically, the emotion engine analyzes the emotional data of user inquiries and requests, and the server operates to quickly and accurately adjust the response content if, for example, the user is in a hurry, thereby increasing the priority of follow-up. This enables more personalized customer service.

[1256] (Example 2)

[1257] 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".

[1258] Traditional in-house customer service systems have suffered from low accuracy in automated responses to customer inquiries, and have been unable to consider customer emotions when generating quotations or managing follow-up schedules. This can lead to decreased customer satisfaction and delays in response times, potentially negatively impacting a company's credibility and performance. Therefore, there is a need for highly accurate automated responses, quotation generation, and follow-up management that take customer emotions into consideration.

[1259] 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.

[1260] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for adjusting the analysis results using an emotion engine that recognizes the user's emotions. This enables a quick and appropriate response that takes customer emotions into consideration.

[1261] "Internal company data" refers to all data collected and stored within a company, including customer information, product information, inquiry history, inventory information, etc.

[1262] An "inquiry" refers to a question or request made by a customer to a company, and includes inquiries about product information, pricing, and services.

[1263] "Analysis" is the process of processing received inquiries and data using analytical tools and algorithms to extract necessary information, keywords, sentiments, and so on.

[1264] A "related database" is a database that stores data searchable based on specific criteria, and may include information on products and services, inventory information, and so on.

[1265] A "response" refers to an answer or message generated based on the analysis results and search results from related databases, and is sent to the customer in the form of email, chat message, push notification, etc.

[1266] An "emotion engine" refers to a technology that recognizes and analyzes emotions from user input text, utilizing natural language processing to determine the user's mood and tone.

[1267] A "quotation" is a document that contains price information for products or services offered to a customer, and is usually automatically generated in PDF format.

[1268] "Follow-up" refers to additional responses to the initial inquiry or request, and includes sending product samples and responding to further inquiries.

[1269] "Push notifications" are notification messages sent to a user's device in real time and are used to provide reminders and important updates.

[1270] This invention is a system that utilizes internal company data to automate responses to customer inquiries, automate the generation of quotations, manage follow-up schedules, and combines an emotion engine that recognizes user emotions. The embodiments of this system are described in detail below.

[1271] Automated response to customer inquiries

[1272] User

[1273] The user enters their inquiry from their device and clicks the send button. For example, the user might enter, "I would like to know the stock status of product A."

[1274] terminal

[1275] The terminal saves user input and sends query data to the server. This communication uses an API endpoint via an internet connection.

[1276] server

[1277] The server analyzes the received query data. This analysis uses text analysis tools such as AWS Textract. Furthermore, it utilizes IBM Watson's natural language processing (NLP) service to recognize the user's sentiment. Based on the recognized sentiment data, the content of the response message is adjusted. Next, the server searches a database such as MySQL to retrieve the necessary information. For example, it searches the inventory status of product A in the inventory database. Based on the search results, an appropriate response message is generated. The generated response message, reflecting the sentiment data, is sent to the user's terminal.

[1278] Automatic generation of quotations

[1279] User

[1280] The user enters a request to issue a quote and clicks the submit button. For example, they might request, "Please issue a quote for product B."

[1281] terminal

[1282] The terminal saves the request data and sends it to the server.

[1283] server

[1284] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product. The server uses IBM Watson's NLP service to recognize sentiment and adjusts the content of the quote based on the sentiment data. The server retrieves price information from the approval database using SQL queries and generates a quote in PDF format using the ReportLab library. The generated quote is saved on the server, and a download link is created. The download link is sent to the user's terminal.

[1285] User

[1286] The user clicks the received download link to download the quote.

[1287] Follow-up schedule management

[1288] User

[1289] When a user submits an inquiry, if follow-up is needed, for example, they might request, "Please send me a sample."

[1290] terminal

[1291] The device saves requests that require follow-up and sends them to the server.

[1292] server

[1293] The server analyzes the received request and determines the need for follow-up. Furthermore, it uses IBM Watson's sentiment engine to recognize the user's emotions. Based on the recognized emotion data, it adjusts the priority and method of follow-up. The server uses the Google Calendar API to set the date and time for follow-up and registers it in the schedule management database. As the set date and time approach, it generates a push notification using Firebase Cloud Messaging and sends it to the user's and assigned personnel's devices.

[1294] terminal

[1295] Users and staff will receive push notifications and take appropriate action.

[1296] Specific example

[1297] For example, if a user sends an inquiry saying, "Please send me a sample of product C," the server analyzes the inquiry and determines that follow-up is necessary to send the sample. Furthermore, if the emotion engine detects that the user is feeling urgent, the server will set an emergency follow-up for 3 days later at 3 PM and send a push notification as a reminder. Based on these instructions, the user and the person in charge will confirm the sample delivery and follow up.

[1298] Example of a prompt

[1299] The following are examples of prompts to input into the generating AI model.

[1300] You are an engineer developing a system to automate customer service within a company. This system will include automated responses to customer inquiries, automated quotation generation, and follow-up scheduling. Please describe the appropriate processing flow and the technologies to be used to meet the following requirements.

[1301] Request: A user has submitted an inquiry about the stock status of product A. Please describe the system flow for handling this inquiry.

[1302] As described above, the present invention automates and streamlines customer service operations, enabling responses that take user emotions into consideration. As a result, higher customer satisfaction can be achieved, and the company's credibility can be improved.

[1303] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1304] Automated response to customer inquiries

[1305] Step 1:

[1306] User

[1307] The user enters their inquiry details on their device and clicks the send button.

[1308] Input: "Please tell me the stock status of product A."

[1309] Output: Query data

[1310] Step 2:

[1311] terminal

[1312] The terminal saves the user's input and sends the inquiry data to the server.

[1313] Input: Inquiry data

[1314] Output: Request to send to server

[1315] Step 3:

[1316] server

[1317] The server analyzes the received query data. AWS Textract is used for this analysis, extracting keywords from the text.

[1318] Input: Submitted inquiry data

[1319] Output: Analysis results (e.g., "Product A", "Inventory Status")

[1320] Step 4:

[1321] server

[1322] The server uses IBM Watson's NLP service to recognize the emotions contained in the query.

[1323] Input: Analysis results

[1324] Output: Sentiment data (e.g., confusion, interest)

[1325] Step 5:

[1326] server

[1327] The server generates data to adjust response messages based on recognized emotion data.

[1328] Input: Sentiment data

[1329] Output: Adjustment data

[1330] Step 6:

[1331] server

[1332] The server searches databases such as MySQL and retrieves relevant information (such as inventory information). It then executes SQL queries to extract the necessary data.

[1333] Input: Query to search for "Inventory information for product A"

[1334] Output: Inventory information (e.g., In stock)

[1335] Step 7:

[1336] server

[1337] The server generates an appropriate response message based on inventory information and sentiment data.

[1338] Input: Inventory information, adjustment data

[1339] Output: Response message (Example: "Thank you. Product A is currently in stock.")

[1340] Step 8:

[1341] server

[1342] The server sends the generated response message to the user's terminal.

[1343] Input: Response message

[1344] Output: Notification to user terminal

[1345] Automatic generation of quotations

[1346] Step 1:

[1347] User

[1348] The user enters a request to request a quote and clicks the submit button.

[1349] Input: "Please issue a quotation for product B."

[1350] Output: Request data

[1351] Step 2:

[1352] terminal

[1353] The terminal saves the request data and sends it to the server.

[1354] Input: Request data

[1355] Output: Request to send to server

[1356] Step 3:

[1357] server

[1358] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product.

[1359] Input: Request data

[1360] Output: Analysis results (e.g., target company, product information)

[1361] Step 4:

[1362] server

[1363] The server uses IBM Watson's NLP service to recognize the user's emotions.

[1364] Input: Analysis results

[1365] Output: Sentiment data (e.g., interest, expectation)

[1366] Step 5:

[1367] server

[1368] The server generates data to adjust the contents of the estimate based on the recognized emotion data.

[1369] Input: Sentiment data

[1370] Output: Adjustment data

[1371] Step 6:

[1372] server

[1373] The server retrieves price information from the approval database using SQL queries.

[1374] Input: Query to search for price information for "Product B"

[1375] Output: Price information (e.g., ¥100,000)

[1376] Step 7:

[1377] server

[1378] The server automatically generates a quotation in PDF format using the ReportLab library.

[1379] Input: Price information, adjustment data

[1380] Output: Quotation PDF (Example: quotate_2023.pdf)

[1381] Step 8:

[1382] server

[1383] The generated quotation is saved to the server, and a download link is generated.

[1384] Input: Quotation PDF

[1385] Output: Download link (Example: https: / / example.com / downloads / quote_2023.pdf)

[1386] Step 9:

[1387] server

[1388] The server sends the generated download link to the user's device.

[1389] Input: Download link

[1390] Output: Notification to user terminal

[1391] Step 10:

[1392] User

[1393] The user clicks the received download link to download the quote.

[1394] Input: Download link

[1395] Output: Download quotation

[1396] Follow-up schedule management

[1397] Step 1:

[1398] User

[1399] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[1400] Input: "Please send a sample."

[1401] Output: Follow-up request

[1402] Step 2:

[1403] terminal

[1404] The device saves requests that require follow-up and sends them to the server.

[1405] Input: Follow-up request

[1406] Output: Request to send to server

[1407] Step 3:

[1408] server

[1409] The server analyzes the received request and determines whether follow-up is necessary.

[1410] Input: Follow-up request

[1411] Output: Analysis results (e.g., "Sample submission required")

[1412] Step 4:

[1413] server

[1414] The server uses IBM Watson's emotion engine to recognize the user's emotions.

[1415] Input: Analysis results

[1416] Output: Sentimental data (e.g., urgency)

[1417] Step 5:

[1418] server

[1419] The server adjusts the priority and method of follow-up based on the recognized sentiment data.

[1420] Input: Sentiment data

[1421] Output: Adjustment data

[1422] Step 6:

[1423] server

[1424] The server uses the Google Calendar API to set the date and time that requires follow-up and registers it in the schedule management database.

[1425] Input: Adjustment data

[1426] Output: Follow-up schedule (Example: 3 days from now, 3 PM)

[1427] Step 7:

[1428] server

[1429] As the scheduled date and time approach, Firebase Cloud Messaging is used to generate a push notification as a reminder, which is then sent to the user and the responsible party.

[1430] Input: Follow-up schedule

[1431] Output: Push notification

[1432] Step 8:

[1433] terminal

[1434] Users and staff will receive push notifications and take appropriate action.

[1435] Input: Push notification

[1436] Output: Follow-up available

[1437] (Application Example 2)

[1438] 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."

[1439] In modern society, companies are required to improve the efficiency and quality of their customer service operations. In particular, electronic payment services demand prompt and appropriate customer service, while simultaneously making it difficult to consider customer emotions. Conventional systems could only provide mechanical responses to inquiries, making it difficult to increase customer satisfaction. Furthermore, tasks such as generating quotations and managing follow-ups required manual handling, resulting in a significant workload. This invention aims to solve these problems by providing a system that recognizes customer emotions and responds accordingly.

[1440] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries and user sentiment data, means for automatically generating and sending a response that reflects the sentiment based on the search results and sentiment data to the customer, and means for automatically generating an electronic receipt and providing the customer with a downloadable link. This makes it possible to provide a quick and appropriate response while taking customer sentiment into consideration.

[1441] "Internal company data" refers to all information generated and stored within a company, including customer data, product data, inventory data, and sales history.

[1442] "Customer inquiries" refer to requests for information, questions, and feedback that customers send to a company, and include those sent via telephone, email, web forms, etc.

[1443] "Analysis" refers to the process of deciphering and understanding received data, and performing operations to grasp its intent and meaning.

[1444] A "related database" refers to a collection of data that stores specific information and is structured in a way that allows for later searching and referencing.

[1445] "Emotional data" refers to data that identifies a user's emotional state (e.g., joy, anger, surprise, etc.) using natural language processing and emotion recognition technologies, and expresses it as numerical values ​​or tags.

[1446] "Automatically generating responses" refers to a system generating reply text without human intervention, based on pre-configured algorithms and rules.

[1447] An "electronic receipt" refers to a receipt issued in digital format, usually generated in PDF format, and provided via email or a download link.

[1448] A "quotation" is a document that shows the price and conditions of the products or services to be offered in advance, and serves as a basis for customers to consider purchasing.

[1449] A "downloadable link" refers to a URL (web address) from which a specific file or information can be obtained via the internet.

[1450] "Follow-up" refers to additional support and verification work carried out continuously after the initial response, and is a process aimed at maintaining and improving customer satisfaction.

[1451] "Push notifications as reminders" refers to displaying short messages on a user's smartphone or other device to alert them based on a specified date, time, or conditions.

[1452] This invention relates to a system for automating and streamlining customer service in electronic payment services. In particular, it aims to improve customer satisfaction by recognizing customer emotions and providing appropriate responses and follow-ups.

[1453] Hardware configuration

[1454] This invention is carried out using the following hardware:

[1455] Smartphone (iOS or Android)

[1456] Cloud servers (e.g., AWS or Google Cloud)

[1457] Software Configuration

[1458] This invention is carried out using the following software:

[1459] Smartphone applications (iOS: Swift, Android: Kotlin)

[1460] Server-side: Node.js, Python

[1461] Database: MySQL or PostgreSQL

[1462] Emotion recognition engine: IBM Watson Tone Analyzer or Google Cloud Natural Language API

[1463] Push notification service: Firebase Cloud Messaging (FCM)

[1464] Overview of Data Processing and Data Calculation

[1465] 1. User submits inquiry:

[1466] The user enters an inquiry about electronic payments via a smartphone app and clicks the submit button. The inquiry data is sent from the device to the server.

[1467] 2. Reception and analysis on the server:

[1468] The server receives the query data and extracts keywords using natural language processing techniques. Furthermore, it analyzes the user's sentiment data using an emotion recognition engine (IBM Watson Tone Analyzer or Google Cloud Natural Language API).

[1469] 3. Generating and sending response messages:

[1470] The server searches the electronic payment database for corresponding information based on the analysis results and sentiment data. It then automatically generates a response message that reflects the sentiment and sends it to the user's smartphone.

[1471] 4. Automatic generation of electronic receipts:

[1472] When a user completes a purchase, the server automatically generates an electronic receipt in PDF format. It then generates a download link for the electronic receipt and provides it to the user.

[1473] 5. Follow-up schedule management:

[1474] The server prioritizes follow-ups based on user sentiment data. The cloud server manages the dates and times when follow-ups are needed, and push notifications are sent as reminders to the user's and the assigned staff member's devices as the scheduled time approaches.

[1475] Specific example

[1476] For example, if a user sends an inquiry asking, "Please tell me my recent transaction history," the server uses its emotion recognition engine to recognize the user's emotion as "high stress." Next, the server retrieves the corresponding transaction history from the database and generates and sends a response message in polite language to alleviate the stress, such as, "Here is your transaction history. Please feel free to contact us anytime if you have any further questions."

[1477] Example of a prompt

[1478] "Please tell me your most recent credit card transaction history."

[1479] "Please resend the most recent receipt via email."

[1480] "I'd like to confirm the payment details."

[1481] This invention makes it possible to recognize user emotions and respond accordingly. This system not only increases customer satisfaction but also streamlines a company's customer service operations.

[1482] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1483] Step 1:

[1484] The user enters their inquiry and clicks the submit button. The "inquiry details" are retrieved from the user's device as input data. This inquiry is then sent to the server via the smartphone.

[1485] Step 2:

[1486] The system analyzes query data received by the server. The input is the query data, and keywords are extracted using a natural language processing tool (e.g., Python's NLTK). The output is the extracted keywords.

[1487] Step 3:

[1488] The server uses an emotion recognition engine to analyze the user's emotional data. The input is "data from the inquiry," and the emotions are analyzed using an emotion recognition tool (e.g., IBM Watson Tone Analyzer). The output is "recognized emotional data."

[1489] Step 4:

[1490] The server performs a search of relevant databases based on the analyzed keywords and sentiment data. The inputs are "extracted keywords" and "recognized sentiment data," and the server queries the databases. The output is the "search results."

[1491] Step 5:

[1492] The server generates a response message that reflects emotions based on search results and sentiment data. The inputs are "search results" and "recognized sentiment data," and the server generates the response message using natural language generation technology (e.g., the GPT model). The output is the "response message."

[1493] Step 6:

[1494] The server sends the generated response message to the user's terminal. The input is the "response message," and the output is the "response message sent to the user's terminal."

[1495] Step 7:

[1496] When a user completes a purchase, the server automatically generates an electronic receipt. The input is "purchase data," and the electronic receipt is generated using a PDF generation tool. The output is a "PDF file of the electronic receipt."

[1497] Step 8:

[1498] The server generates a download link for the electronic receipt and provides it to the user. The input is a "PDF file of the electronic receipt," and the output is a "download link for the electronic receipt."

[1499] Step 9:

[1500] The server manages the schedule for follow-ups when necessary. Inputs are "follow-up request data" and "recognized sentiment data," and the server sets the follow-up priority. The output is the "configured follow-up schedule."

[1501] Step 10:

[1502] The server generates a push notification as a reminder as the scheduled date and time approaches. The input is the "follow-up schedule data," and the output is the "generated push notification." The notification is sent to the user's and the assigned person's devices.

[1503] 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.

[1504] 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.

[1505] 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.

[1506] [Third Embodiment]

[1507] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1508] 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.

[1509] 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).

[1510] 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.

[1511] 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.

[1512] 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).

[1513] 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.

[1514] 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.

[1515] 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.

[1516] 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.

[1517] 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.

[1518] 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".

[1519] This invention is a system that utilizes internal company data to centrally manage automated responses to customer inquiries, automated generation of quotations, and follow-up schedules.

[1520] 1. Automated response to customer inquiries

[1521] User

[1522] 1. The user enters their inquiry details on their device and clicks the send button.

[1523] 2. For example, a user enters "I would like to know the stock status of product A."

[1524] terminal

[1525] 1. The terminal saves the user's input and sends the inquiry data to the server.

[1526] server

[1527] 1. The server analyzes the received query data.

[1528] 2. Based on the analysis results, search for relevant databases (e.g., inventory databases).

[1529] 3. Automatically generate a response message based on the search results (e.g., availability of product A).

[1530] 4. Send the generated response message to the user's terminal.

[1531] 2. Automatic generation of quotations

[1532] User

[1533] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[1534] 2. For example, you could request, "Please issue a quotation for product B."

[1535] terminal

[1536] 1. The terminal saves the request data and sends it to the server.

[1537] server

[1538] 1. The server parses the received request.

[1539] 2. Search the approval database for price data of the target company and product.

[1540] 3. Automatically generate a quotation based on the search results. (Example: PDF format)

[1541] 4. Save the generated quotation and create a download link.

[1542] 5. Send the download link to the user's device.

[1543] User

[1544] 1. The user clicks the received download link to download the quotation.

[1545] 3. Schedule management for follow-up

[1546] User

[1547] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[1548] terminal

[1549] 1. The device sends requests requiring follow-up to the server.

[1550] server

[1551] 1. The server analyzes the received request and determines whether follow-up is necessary.

[1552] 2. Set the date and time for follow-up and register it in the schedule management database.

[1553] 3. Generate a push notification as a reminder based on the set date and time.

[1554] 4. Send push notifications to the user's and the person in charge's devices.

[1555] terminal

[1556] 1. Users and staff will receive push notifications and take appropriate action.

[1557] Specific example

[1558] For example, consider a case where a user submits an inquiry requesting "Please send me a sample of product C." In this case, the server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a schedule for follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and, based on the instructions, confirm the sample shipment and follow up.

[1559] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and schedule management to automate and streamline customer service operations, thereby achieving fast and accurate customer service.

[1560] The following describes the processing flow.

[1561] Automated response to customer inquiries

[1562] Step 1:

[1563] The user enters their inquiry details on their device and clicks the send button.

[1564] Step 2:

[1565] The terminal saves the user's input and sends the inquiry data to the server.

[1566] Step 3:

[1567] The server analyzes the received query data and extracts keywords and related information.

[1568] Step 4:

[1569] Based on the analysis results, the server searches relevant databases (e.g., inventory databases).

[1570] Step 5:

[1571] The server generates an appropriate response message based on the search results.

[1572] Step 6:

[1573] The server sends the generated response message to the terminal.

[1574] Step 7:

[1575] The terminal receives a response from the server and displays it to the user.

[1576] Automatic generation of quotations

[1577] Step 1:

[1578] The user enters a request to request a quote and clicks the submit button.

[1579] Step 2:

[1580] The terminal saves the request data and sends it to the server.

[1581] Step 3:

[1582] The server analyzes the received request and extracts price data for the target company and product.

[1583] Step 4:

[1584] The server retrieves price information from the approval database.

[1585] Step 5:

[1586] The server automatically generates a quote based on the price information (e.g., in PDF format).

[1587] Step 6:

[1588] The server saves the generated quote and creates a download link.

[1589] Step 7:

[1590] The server sends a download link to the device.

[1591] Step 8:

[1592] The device receives the download link and displays it to the user.

[1593] Step 9:

[1594] The user clicks the download link to download the quote.

[1595] Follow-up schedule management

[1596] Step 1:

[1597] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[1598] Step 2:

[1599] The device sends requests to the server that require follow-up.

[1600] Step 3:

[1601] The server analyzes the received request and determines whether follow-up is necessary.

[1602] Step 4:

[1603] The server sets the date and time for follow-up and registers it in the schedule management database.

[1604] Step 5:

[1605] The server generates a push notification as a reminder when the scheduled date and time approach.

[1606] Step 6:

[1607] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[1608] Step 7:

[1609] The device receives push notifications and displays them to the user and the person in charge.

[1610] Step 8:

[1611] Users and their representatives will take appropriate follow-up actions based on push notifications.

[1612] (Example 1)

[1613] 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."

[1614] In corporate customer service operations, traditional manual responses are time-consuming and labor-intensive, leading to decreased customer satisfaction. In particular, efficiently managing inquiries, issuing quotations, and follow-ups is difficult, and there is a demand for quick and accurate responses. Furthermore, the lack of a unified management system for these tasks necessitates information integration and automation.

[1615] 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.

[1616] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for generating response messages using a generation AI model. This enables rapid and accurate automated responses to customer inquiries.

[1617] In this invention, the server includes means for searching a database of prices offered by each company and automatically generating a quotation in a format accessible to the customer; means for generating a downloadable link for the generated quotation and providing it to the customer; means for downloading the quotation via the provided link; and means for adjusting the content of the quotation using a generation AI model. This enables the rapid and automatic creation and provision of quotations.

[1618] In this invention, the server includes means for tracking the processing status of customer inquiries and setting a date and time when follow-up is required; means for sending push notifications as reminders to customers and personnel based on the set date and time; means for receiving push notifications; and means for generating the content of the reminder using a prompt statement. This enables efficient scheduling of follow-ups and timely responses to be taken.

[1619] These measures enable companies to automate and streamline customer service operations, contributing to improved customer satisfaction.

[1620] "Internal company data" refers to information and documents generated and stored within a company.

[1621] "Customer inquiries" refer to information about questions and requests that customers make to a company.

[1622] "Means of receiving and analyzing data" refers to the technology that allows a system to take in data from an external source and understand its contents.

[1623] A "related database" refers to a database that stores information related to a specific query.

[1624] "Searching methods" refer to techniques for finding information within a database based on specific criteria.

[1625] "Means of generating responses" refers to technologies that automatically create appropriate answers to inquiries.

[1626] A "generative AI model" refers to an algorithm that uses artificial intelligence to generate new information from data.

[1627] A "downloadable link" refers to a web address that allows a user to download a specific file by clicking on it.

[1628] A "quote" refers to a document that lists the price, details, and terms and conditions of a product or service.

[1629] "Means of tracking" refers to technologies that track and record the progress of a specific process or event.

[1630] "Follow-up" refers to additional actions or confirmations taken after the initial response or action.

[1631] "Push notifications" refer to a technology that sends information from a server to a device in real time.

[1632] A "prompt statement" refers to an input statement used to instruct a generative AI model to generate information.

[1633] This invention is a system that automates and streamlines customer service operations by utilizing internal company data. The program processing of this system is described in detail below.

[1634] Automated response to customer inquiries

[1635] 1. User

[1636] The user enters their inquiry details on their device and clicks the "Send" button.

[1637] For example, you would enter an inquiry such as, "Please tell me the stock status of product A."

[1638] 2. Terminal

[1639] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[1640] 3. Server

[1641] The server parses the received query data and uses natural language processing libraries (e.g., spaCy or NLTK) to analyze the information the user is seeking.

[1642] Based on the analysis results, the system searches relevant databases (e.g., inventory databases) and issues SQL queries.

[1643] Based on the search results, a prompt is sent to a generative AI model (e.g., OpenAI's GPT-4) to generate a response message. Example prompt: "How much stock do you currently have of product A?"

[1644] The generated response message is sent to the user's terminal in JSON format.

[1645] 4. Terminal

[1646] The terminal displays the received response message on the user interface.

[1647] Automatic generation of quotations

[1648] 1. User

[1649] The user enters the request for a quote and clicks the "Submit" button.

[1650] For example, you might enter, "Please issue a quotation for product B."

[1651] 2. Terminal

[1652] The terminal saves the request data to a local database and then sends it to the server.

[1653] 3. Server

[1654] The server analyzes the request data and extracts the necessary information using a natural language processing library.

[1655] Search the price database and issue an SQL query. Example SQL query: "SELECT price FROM products WHERE product_name='product B'"

[1656] Based on the search results, the system automatically generates a quotation in PDF format using libraries such as Apache PDFBox.

[1657] The generated quote is saved to cloud storage (e.g., AWS S3), and a downloadable link is generated and sent to the user's device.

[1658] 4. User

[1659] The user clicks the provided link and downloads the quote.

[1660] Follow-up schedule management

[1661] 1. User

[1662] The user enters a request that requires follow-up and clicks the "Send" button.

[1663] For example, you might enter, "Please send me a sample of product C."

[1664] 2. Terminal

[1665] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[1666] 3. Server

[1667] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[1668] Set the necessary follow-up dates and register them in the schedule management database (e.g., Microsoft SQL Server).

[1669] Using Firebase Cloud Messaging, push notifications are generated at a set date and time and sent to the user's and the assigned person's devices.

[1670] 4. Terminal

[1671] Users and staff members receive push notifications and follow up accordingly.

[1672] Hardware and software used

[1673] Server: Cloud server (e.g., AWS EC2, Microsoft Azure)

[1674] Database management systems: MySQL, Oracle, Microsoft SQL Server

[1675] Natural language processing libraries: spaCy, NLTK

[1676] Generative AI models: OpenAI GPT-4, etc.

[1677] Cloud storage: AWS S3

[1678] PDF generation library: Apache PDFBox

[1679] Push notification service: Firebase Cloud Messaging

[1680] Specific example

[1681] For example, if a user submits an inquiry requesting "Please send me a sample of product C," the following process takes place: The server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a follow-up appointment in the schedule management database for three days later and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and confirm and carry out the sample shipment based on the instructions.

[1682] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and follow-up schedule management to automate and streamline customer service operations, thereby achieving prompt and accurate customer service.

[1683] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1684] Processing steps for automated responses to customer inquiries

[1685] Step 1:

[1686] User

[1687] The user enters their inquiry details on their device and clicks the "Send" button.

[1688] Input: User inquiry (e.g., "Please tell me the stock status of product A")

[1689] Output: A request is generated to send query data to the terminal.

[1690] Step 2:

[1691] terminal

[1692] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[1693] Input: User inquiry data

[1694] Output: Send query data to the server in JSON format.

[1695] Step 3:

[1696] server

[1697] The server analyzes the received query data and uses a natural language processing library to understand the intent of the query.

[1698] Software used: spaCy, NLTK

[1699] Input: Query data in JSON format

[1700] Data processing: Text analysis using natural language processing

[1701] Output: Analysis results of the inquiry

[1702] Step 4:

[1703] server

[1704] Based on the analysis results, SQL queries are issued to relevant databases (e.g., inventory databases) to perform searches.

[1705] Database management system used: MySQL

[1706] Input: Analysis results of the inquiry content

[1707] Data Calculation: Searching Related Databases

[1708] Output: Database search results (e.g., "Inventory quantity of product A")

[1709] Step 5:

[1710] server

[1711] A response message is generated using an AI model based on the database search results.

[1712] Generative AI model used: OpenAI GPT-4

[1713] Input: Database search results

[1714] Data processing: Generating response messages using a generative AI model.

[1715] Output: Response message (Example: "Product A is in stock")

[1716] Step 6:

[1717] server

[1718] The server sends the generated response message to the user's terminal in JSON format.

[1719] Input: Response message

[1720] Output: Send a JSON-formatted response message to the user's terminal.

[1721] Step 7:

[1722] terminal

[1723] The terminal displays the received response message on the user interface.

[1724] Input: Response message in JSON format

[1725] Output: Display in the user interface

[1726] Process steps for automatic quotation generation

[1727] Step 1:

[1728] User

[1729] The user enters the request for a quote and clicks the "Submit" button.

[1730] Input: Product name and quotation request (Example: "Please issue a quotation for product B")

[1731] Output: Data is input to the terminal, and a transmission command is generated.

[1732] Step 2:

[1733] terminal

[1734] The terminal saves the request data to a local database and then sends it to the server.

[1735] Input: User's request data

[1736] Output: Send request data to the server in JSON format.

[1737] Step 3:

[1738] server

[1739] The server parses the request data and searches the database to retrieve relevant pricing information.

[1740] Software used: Natural language processing libraries (spaCy, NLTK)

[1741] Input: Request data

[1742] Data processing: Text analysis using natural language processing

[1743] Output: Analyzed request content

[1744] Step 4:

[1745] server

[1746] Based on the parsed request, an SQL query is issued to the price database to retrieve the necessary price information.

[1747] Database management system used: Oracle

[1748] Input: Parsed request content

[1749] Data Calculation: Searching Price Databases

[1750] Output: Price data (Example: "Price information for product B")

[1751] Step 5:

[1752] server

[1753] Based on the acquired price data, an AI model is used to automatically generate a quotation.

[1754] Generative AI model used: OpenAI GPT-4

[1755] PDF generation library used: Apache PDFBox

[1756] Input: Price data

[1757] Data calculation: Generating a PDF quotation

[1758] Output: Generated quotation (PDF format)

[1759] Step 6:

[1760] server

[1761] The generated quote is saved to cloud storage, and a downloadable link is created and sent to the user's device.

[1762] Cloud storage to use: AWS S3

[1763] Input: Generated quotation (PDF format)

[1764] Output: Download link

[1765] Step 7:

[1766] User

[1767] The user clicks the provided link and downloads the quote.

[1768] Input: Download link

[1769] Output: Downloaded quotation (PDF format)

[1770] Follow-up scheduling process steps

[1771] Step 1:

[1772] User

[1773] The user enters a request that requires follow-up and clicks the "Send" button.

[1774] Input: Follow-up request (Example: "Please send a sample of product C")

[1775] Output: Data is input to the terminal, and a transmission command is generated.

[1776] Step 2:

[1777] terminal

[1778] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[1779] Input: Request data requiring follow-up

[1780] Output: Send request data to the server in JSON format.

[1781] Step 3:

[1782] server

[1783] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[1784] Software used: Natural language processing libraries (spaCy, NLTK)

[1785] Input: Follow-up request data

[1786] Data processing: Text analysis using natural language processing

[1787] Output: Result of the assessment of the need for follow-up.

[1788] Step 4:

[1789] server

[1790] Set the date and time for follow-up and register it in the schedule management database.

[1791] Database management system used: Microsoft SQL Server

[1792] Input: Result of the assessment of the need for follow-up.

[1793] Data processing: Setting and registering follow-up dates and times.

[1794] Output: Configured follow-up schedule

[1795] Step 5:

[1796] server

[1797] Based on the configured date and time, push notifications are generated using Firebase Cloud Messaging and sent to the user's and assigned personnel's devices.

[1798] Push notification service used: Firebase Cloud Messaging

[1799] Input: Configured follow-up schedule

[1800] Data processing: Generating and sending push notifications

[1801] Output: Push notifications to user and staff terminals

[1802] Step 6:

[1803] terminal

[1804] Users and staff members receive push notifications and follow up accordingly.

[1805] Input: Push notification

[1806] Output: Follow-up will be performed.

[1807] Through the specific processing steps outlined above, the system can automate and streamline customer service operations.

[1808] (Application Example 1)

[1809] 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."

[1810] Traditional e-commerce sites struggled to respond quickly and accurately to customer inquiries, particularly with manual processes like checking inventory and issuing quotes, resulting in inefficiencies. Furthermore, managing follow-up schedules was cumbersome, often leading to delays in notifications to both customers and staff. This resulted in decreased customer satisfaction and missed sales opportunities. There is a need to solve these problems and automate and streamline customer service.

[1811] 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.

[1812] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data; means for searching relevant databases based on the analyzed inquiries; means for automatically generating and sending responses to customers based on the search results; means for analyzing customer inquiries regarding specific product information or inventory status using natural language processing; means for performing natural language processing using a generative AI model to automatically generate appropriate responses; means for searching each company's pricing database and automatically generating quotations in a format accessible to customers; means for generating and providing downloadable links for the generated quotations to customers; means for receiving quotation requests, analyzing product information and pricing data using a generative AI model, and automatically generating quotations; means for tracking the processing status of customer inquiries and setting dates and times for follow-up; means for sending push notifications to customers and personnel as reminders based on the set dates and times; and means for using a schedule management database to automatically generate reminders when follow-up is required. This enables not only a quick and accurate response to customer inquiries but also efficient automatic quotation generation and follow-up schedule management.

[1813] "Internal company data" refers to the collection of all information and data managed within a company.

[1814] An "inquiry" refers to a question or request sent by a customer to confirm product information, stock availability, etc.

[1815] "Analysis" is the process of breaking down the content of an received inquiry and interpreting its meaning and intent.

[1816] A "database" is a system that stores and manages related information in an organization.

[1817] "Searching" is the process of finding specific information within a database.

[1818] "Response" refers to the answers or information provided in response to customer inquiries.

[1819] "Natural language processing" is the technology that enables computers to understand and process human language.

[1820] A "generative AI model" refers to an algorithm or framework that uses artificial intelligence to generate new text or data.

[1821] A "price database" is a database where product price information is stored in an organ.

[1822] A "quotation" is a document that shows the price and conditions of a specific product or service.

[1823] "Follow-up" refers to additional confirmations or follow-ups conducted after the initial response.

[1824] A "schedule management database" is a system that organizes and stores follow-up dates and appointments.

[1825] A "reminder" is a message sent to notify you of an action that needs to be taken at a specific date and time.

[1826] A "push notification" is a notification message that is sent to a user's device in real time.

[1827] The specific system for implementing the present invention is capable of efficiently and quickly processing customer inquiries and easily managing the automatic generation of quotations and follow-up schedules. This system is configured as follows.

[1828] Automated response to customer inquiries

[1829] The server automatically receives and analyzes customer inquiries using internal company data. It receives the inquiry content entered by the customer on their device and sends it to the server. The server uses Google Cloud AI to perform natural language processing and analyze the inquiry content. It then searches relevant databases (e.g., product inventory database) and automatically generates an appropriate response. The generated response is sent to the customer's device.

[1830] Specific example

[1831] For example, if a customer asks, "What is the stock status of product A?", the server analyzes the inquiry, retrieves the stock status of product A from the product inventory database, and automatically generates a response message saying, "Product A is currently in stock."

[1832] Example of a prompt:

[1833] Inquiry: "Please tell me the stock status of product A."

[1834] Generated AI prompt: "Check the inventory status of product A and respond to the customer whether it is in stock or not."

[1835] Automatic generation of quotations

[1836] When a customer requests a quote for a specific product, the request data is sent from the terminal to the server. The server receives and analyzes the request. Next, it retrieves the price data for the product from the price database and automatically generates a quote using a generation AI model. The generated quote is saved in PDF format, and a download link is sent to the customer.

[1837] Specific example

[1838] When a customer requests "Please issue a quote for product B," the server analyzes the request, retrieves the price information for product B from the price database, and automatically generates a quote. A download link for the generated quote is then sent to the customer's device.

[1839] Example of a prompt:

[1840] Quote Request: "Please issue a quote for product B."

[1841] Generated AI prompt: "Retrieve pricing data for product B, generate a quote, and provide it to the customer."

[1842] Follow-up schedule management

[1843] If a customer requires follow-up on an inquiry, they send a follow-up request from their device to the server. The server parses this request and sets a date and time for the follow-up. This information is registered in the schedule management database, and a push notification is generated as a reminder based on the set date and time. This push notification is sent to the customer's and the agent's devices.

[1844] Specific example

[1845] For example, if a customer requests a sample of product C, the server analyzes the request and determines that follow-up is necessary. It sets a schedule for a follow-up three days later and registers it in the schedule management database. Three days later, a push notification is automatically sent as a reminder.

[1846] Example of a prompt:

[1847] Follow-up request: "Please send me a sample of product C."

[1848] Generated AI prompt: "Please schedule a follow-up for sending a sample of Product C in 3 days and send a reminder."

[1849] As described above, the present invention is a system that enables rapid and accurate responses to customer inquiries, automatic generation of quotations, and efficient scheduling of follow-ups. This improves customer satisfaction and streamlines operations.

[1850] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1851] Processing steps for automated responses to customer inquiries

[1852] Step 1:

[1853] The user enters their inquiry details on their device and clicks the send button. An example of input data is, "Please tell me the stock status of product A." Input: Inquiry details. Output: Sending of inquiry data.

[1854] Step 2:

[1855] The terminal saves user input and sends query data to the server. Network communication is performed for data transfer. Input: Query data. Output: Data sent to the server.

[1856] Step 3:

[1857] The server receives query data, which is then processed using natural language processing and analysis with Google Cloud AI. Input: Query data. Output: Analysis results.

[1858] Step 4:

[1859] The server searches relevant databases (e.g., inventory databases) based on the analysis results and retrieves the necessary information. Input: Analysis results. Output: Search results.

[1860] Step 5:

[1861] The server automatically generates response messages based on search results, using a generation AI model. Input: Search results. Output: Response message.

[1862] Step 6:

[1863] The server generates a response message and sends it to the user's terminal. Input: Response message. Output: Message sent to the user's terminal.

[1864] Process steps for automatic quotation generation

[1865] Step 1:

[1866] The user requests a quotation and clicks the send button on their terminal. Example input: "Please issue a quotation for product B." Input: Quotation request. Output: Quotation request data sent.

[1867] Step 2:

[1868] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Estimate request data. Output: Data sent to the server.

[1869] Step 3:

[1870] The server analyzes the request and uses a generated AI model to retrieve product information and pricing data. Input: Quotation request data. Output: Analysis results and pricing information.

[1871] Step 4:

[1872] The server automatically generates a quotation based on the analysis results and saves it in PDF format. Input: Analysis results and price information. Output: Quotation in PDF format.

[1873] Step 5:

[1874] The server generates a download link for the generated quotation and sends it to the user's device. Input: Quotation in PDF format. Output: Sending of download link.

[1875] Follow-up scheduling process steps

[1876] Step 1:

[1877] The user enters a request requiring follow-up and clicks the send button on their device. Example input: "Please send a sample of product C." Input: Follow-up request. Output: Sending of follow-up request data.

[1878] Step 2:

[1879] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Follow-up request data. Output: Data sent to the server.

[1880] Step 3:

[1881] The server analyzes the received request data and sets the date and time for follow-up. Input: Follow-up request data. Output: Follow-up schedule.

[1882] Step 4:

[1883] The server registers the follow-up date and time in the schedule management database. Input: Follow-up schedule. Output: Schedule registration information.

[1884] Step 5:

[1885] The server generates a push notification as a reminder based on the configured date and time. Input: Follow-up schedule. Output: Push notification.

[1886] Step 6:

[1887] The server generates push notifications and sends them to the user's and the assigned personnel's devices. Input: Push notification. Output: Notification sent to the user's and the assigned personnel's devices.

[1888] 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.

[1889] This invention is a system that utilizes internal company data to automatically respond to customer inquiries, automatically generate quotations, manage follow-up schedules, and combines this with an emotion engine that recognizes user emotions.

[1890] 1. Automated response to customer inquiries

[1891] User

[1892] 1. The user enters their inquiry details on their device and clicks the send button.

[1893] 2. For example, a user enters "I would like to know the stock status of product A."

[1894] terminal

[1895] 1. The terminal saves the user's input and sends the inquiry data to the server.

[1896] server

[1897] 1. The server analyzes the received query data and extracts keywords and related information.

[1898] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[1899] 3. Based on the recognized emotions, generate data to adjust the response message.

[1900] 4. Based on the analysis results and sentiment data, search relevant databases (e.g., inventory databases).

[1901] 5. Based on the search results, generate appropriate response messages that reflect sentiment data.

[1902] 6. Send the generated response message to the user's terminal.

[1903] 2. Automatic generation of quotations

[1904] User

[1905] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[1906] 2. For example, you could request, "Please issue a quotation for product B."

[1907] terminal

[1908] 1. The terminal saves the request data and sends it to the server.

[1909] server

[1910] 1. The server analyzes the received request and extracts price data for the target company and product.

[1911] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[1912] 3. Based on the recognized emotion data, generate data to adjust the contents of the quotation.

[1913] 4. Retrieve price information from the approval database.

[1914] 5. Automatically generate quotes based on price information and sentiment data (e.g., in PDF format).

[1915] 6. Save the generated quotation and create a download link.

[1916] 7. Send the download link to the user's device.

[1917] User

[1918] 1. The user clicks the received download link to download the quotation.

[1919] 3. Schedule management for follow-up

[1920] User

[1921] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[1922] terminal

[1923] 1. The device sends requests requiring follow-up to the server.

[1924] server

[1925] 1. The server analyzes the received request and determines whether follow-up is necessary.

[1926] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[1927] 3. Based on the recognized emotional data, adjust the priorities and methods of follow-up.

[1928] 4. Set the date and time for follow-up and register it in the schedule management database.

[1929] 5. When the set date and time approach, generate a push notification as a reminder.

[1930] 6. Send the generated push notification to the user's and the person in charge's devices.

[1931] terminal

[1932] 1. Users and staff will receive push notifications and take appropriate action.

[1933] Specific example

[1934] For example, consider a case where a user submits an inquiry requesting a sample of product C. In this case, the server analyzes the inquiry and determines that follow-up is necessary to send the sample. It also uses an emotion engine to recognize if the user feels a sense of urgency. The server sets a schedule for a follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge then confirm the sample delivery and follow up based on these instructions.

[1935] As described above, the system of the present invention can automate and streamline customer service operations, and provide a better customer experience through responses and methods that take into account the user's emotions.

[1936] The following describes the processing flow.

[1937] Automated response to customer inquiries

[1938] Step 1:

[1939] The user enters their inquiry details on their device and clicks the send button.

[1940] Step 2:

[1941] The terminal saves the user's input and sends the inquiry data to the server.

[1942] Step 3:

[1943] The server analyzes the received query data and extracts keywords and related information.

[1944] Step 4:

[1945] The server uses an emotion engine to recognize the user's emotions.

[1946] Step 5:

[1947] The server searches relevant databases (e.g., inventory databases) based on the analysis results and sentiment data.

[1948] Step 6:

[1949] The server generates a response message based on the search results.

[1950] Step 7:

[1951] The server incorporates sentiment data into the generated response message and makes appropriate adjustments.

[1952] Step 8:

[1953] The server sends the generated response message to the terminal.

[1954] Step 9:

[1955] The terminal receives a response from the server and displays it to the user.

[1956] Automatic generation of quotations

[1957] Step 1:

[1958] The user enters a request to request a quote and clicks the submit button.

[1959] Step 2:

[1960] The terminal saves the request data and sends it to the server.

[1961] Step 3:

[1962] The server analyzes the received request and extracts price data for the target company and product.

[1963] Step 4:

[1964] The server uses an emotion engine to recognize the user's emotions.

[1965] Step 5:

[1966] The server retrieves price information from the approval database.

[1967] Step 6:

[1968] The server automatically generates quotes (e.g., in PDF format) based on price information and sentiment data.

[1969] Step 7:

[1970] The server saves the generated quote and creates a download link.

[1971] Step 8:

[1972] The server sends a download link to the device.

[1973] Step 9:

[1974] The device receives the download link and displays it to the user.

[1975] Step 10:

[1976] The user clicks the download link to download the quote.

[1977] Follow-up schedule management

[1978] Step 1:

[1979] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[1980] Step 2:

[1981] The device sends requests to the server that require follow-up.

[1982] Step 3:

[1983] The server analyzes the received request and determines whether follow-up is necessary.

[1984] Step 4:

[1985] The server uses an emotion engine to recognize the user's emotions.

[1986] Step 5:

[1987] The server sets the date and time for follow-up and registers it in the schedule management database.

[1988] Step 6:

[1989] The server generates a push notification as a reminder when the scheduled date and time approach.

[1990] Step 7:

[1991] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[1992] Step 8:

[1993] The device receives push notifications and displays them to the user and the person in charge.

[1994] Step 9:

[1995] Users and their representatives will take appropriate follow-up actions based on push notifications.

[1996] Specifically, the emotion engine analyzes the emotional data of user inquiries and requests, and the server operates to quickly and accurately adjust the response content if, for example, the user is in a hurry, thereby increasing the priority of follow-up. This enables more personalized customer service.

[1997] (Example 2)

[1998] 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."

[1999] Traditional in-house customer service systems have suffered from low accuracy in automated responses to customer inquiries, and have been unable to consider customer emotions when generating quotations or managing follow-up schedules. This can lead to decreased customer satisfaction and delays in response times, potentially negatively impacting a company's credibility and performance. Therefore, there is a need for highly accurate automated responses, quotation generation, and follow-up management that take customer emotions into consideration.

[2000] 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.

[2001] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for adjusting the analysis results using an emotion engine that recognizes the user's emotions. This enables a quick and appropriate response that takes customer emotions into consideration.

[2002] "Internal company data" refers to all data collected and stored within a company, including customer information, product information, inquiry history, inventory information, etc.

[2003] An "inquiry" refers to a question or request made by a customer to a company, and includes inquiries about product information, pricing, and services.

[2004] "Analysis" is the process of processing received inquiries and data using analytical tools and algorithms to extract necessary information, keywords, sentiments, and so on.

[2005] A "related database" is a database that stores data searchable based on specific criteria, and may include information on products and services, inventory information, and so on.

[2006] A "response" refers to an answer or message generated based on the analysis results and search results from related databases, and is sent to the customer in the form of email, chat message, push notification, etc.

[2007] An "emotion engine" refers to a technology that recognizes and analyzes emotions from user input text, utilizing natural language processing to determine the user's mood and tone.

[2008] A "quotation" is a document that contains price information for products or services offered to a customer, and is usually automatically generated in PDF format.

[2009] "Follow-up" refers to additional responses to the initial inquiry or request, and includes sending product samples and responding to further inquiries.

[2010] "Push notifications" are notification messages sent to a user's device in real time and are used to provide reminders and important updates.

[2011] This invention is a system that utilizes internal company data to automate responses to customer inquiries, automate the generation of quotations, manage follow-up schedules, and combines an emotion engine that recognizes user emotions. The embodiments of this system are described in detail below.

[2012] Automated response to customer inquiries

[2013] User

[2014] The user enters their inquiry from their device and clicks the send button. For example, the user might enter, "I would like to know the stock status of product A."

[2015] terminal

[2016] The terminal saves user input and sends query data to the server. This communication uses an API endpoint via an internet connection.

[2017] server

[2018] The server analyzes the received query data. This analysis uses text analysis tools such as AWS Textract. Furthermore, it utilizes IBM Watson's natural language processing (NLP) service to recognize the user's sentiment. Based on the recognized sentiment data, the content of the response message is adjusted. Next, the server searches a database such as MySQL to retrieve the necessary information. For example, it searches the inventory status of product A in the inventory database. Based on the search results, an appropriate response message is generated. The generated response message, reflecting the sentiment data, is sent to the user's terminal.

[2019] Automatic generation of quotations

[2020] User

[2021] The user enters a request to issue a quote and clicks the submit button. For example, they might request, "Please issue a quote for product B."

[2022] terminal

[2023] The terminal saves the request data and sends it to the server.

[2024] server

[2025] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product. The server uses IBM Watson's NLP service to recognize sentiment and adjusts the content of the quote based on the sentiment data. The server retrieves price information from the approval database using SQL queries and generates a quote in PDF format using the ReportLab library. The generated quote is saved on the server, and a download link is created. The download link is sent to the user's terminal.

[2026] User

[2027] The user clicks the received download link to download the quote.

[2028] Follow-up schedule management

[2029] User

[2030] When a user submits an inquiry, if follow-up is needed, for example, they might request, "Please send me a sample."

[2031] terminal

[2032] The device saves requests that require follow-up and sends them to the server.

[2033] server

[2034] The server analyzes the received request and determines the need for follow-up. Furthermore, it uses IBM Watson's sentiment engine to recognize the user's emotions. Based on the recognized emotion data, it adjusts the priority and method of follow-up. The server uses the Google Calendar API to set the date and time for follow-up and registers it in the schedule management database. As the set date and time approach, it generates a push notification using Firebase Cloud Messaging and sends it to the user's and assigned personnel's devices.

[2035] terminal

[2036] Users and staff will receive push notifications and take appropriate action.

[2037] Specific example

[2038] For example, if a user sends an inquiry saying, "Please send me a sample of product C," the server analyzes the inquiry and determines that follow-up is necessary to send the sample. Furthermore, if the emotion engine detects that the user is feeling urgent, the server will set an emergency follow-up for 3 days later at 3 PM and send a push notification as a reminder. Based on these instructions, the user and the person in charge will confirm the sample delivery and follow up.

[2039] Example of a prompt

[2040] The following are examples of prompts to input into the generating AI model.

[2041] You are an engineer developing a system to automate customer service within a company. This system will include automated responses to customer inquiries, automated quotation generation, and follow-up scheduling. Please describe the appropriate processing flow and the technologies to be used to meet the following requirements.

[2042] Request: A user has submitted an inquiry about the stock status of product A. Please describe the system flow for handling this inquiry.

[2043] As described above, the present invention automates and streamlines customer service operations, enabling responses that take user emotions into consideration. As a result, higher customer satisfaction can be achieved, and the company's credibility can be improved.

[2044] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2045] Automated response to customer inquiries

[2046] Step 1:

[2047] User

[2048] The user enters their inquiry details on their device and clicks the send button.

[2049] Input: "Please tell me the stock status of product A."

[2050] Output: Query data

[2051] Step 2:

[2052] terminal

[2053] The terminal saves the user's input and sends the inquiry data to the server.

[2054] Input: Inquiry data

[2055] Output: Request to send to server

[2056] Step 3:

[2057] server

[2058] The server analyzes the received query data. AWS Textract is used for this analysis, extracting keywords from the text.

[2059] Input: Submitted inquiry data

[2060] Output: Analysis results (e.g., "Product A", "Inventory Status")

[2061] Step 4:

[2062] server

[2063] The server uses IBM Watson's NLP service to recognize the emotions contained in the query.

[2064] Input: Analysis results

[2065] Output: Sentiment data (e.g., confusion, interest)

[2066] Step 5:

[2067] server

[2068] The server generates data to adjust response messages based on recognized emotion data.

[2069] Input: Sentiment data

[2070] Output: Adjustment data

[2071] Step 6:

[2072] server

[2073] The server searches databases such as MySQL and retrieves relevant information (such as inventory information). It then executes SQL queries to extract the necessary data.

[2074] Input: Query to search for "Inventory information for product A"

[2075] Output: Inventory information (e.g., In stock)

[2076] Step 7:

[2077] server

[2078] The server generates an appropriate response message based on inventory information and sentiment data.

[2079] Input: Inventory information, adjustment data

[2080] Output: Response message (Example: "Thank you. Product A is currently in stock.")

[2081] Step 8:

[2082] server

[2083] The server sends the generated response message to the user's terminal.

[2084] Input: Response message

[2085] Output: Notification to user terminal

[2086] Automatic generation of quotations

[2087] Step 1:

[2088] User

[2089] The user enters a request to request a quote and clicks the submit button.

[2090] Input: "Please issue a quotation for product B."

[2091] Output: Request data

[2092] Step 2:

[2093] terminal

[2094] The terminal saves the request data and sends it to the server.

[2095] Input: Request data

[2096] Output: Request to send to server

[2097] Step 3:

[2098] server

[2099] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product.

[2100] Input: Request data

[2101] Output: Analysis results (e.g., target company, product information)

[2102] Step 4:

[2103] server

[2104] The server uses IBM Watson's NLP service to recognize the user's emotions.

[2105] Input: Analysis results

[2106] Output: Sentiment data (e.g., interest, expectation)

[2107] Step 5:

[2108] server

[2109] The server generates data to adjust the contents of the estimate based on the recognized emotion data.

[2110] Input: Sentiment data

[2111] Output: Adjustment data

[2112] Step 6:

[2113] server

[2114] The server retrieves price information from the approval database using SQL queries.

[2115] Input: Query to search for price information for "Product B"

[2116] Output: Price information (e.g., ¥100,000)

[2117] Step 7:

[2118] server

[2119] The server automatically generates a quotation in PDF format using the ReportLab library.

[2120] Input: Price information, adjustment data

[2121] Output: Quotation PDF (Example: quotate_2023.pdf)

[2122] Step 8:

[2123] server

[2124] The generated quotation is saved to the server, and a download link is generated.

[2125] Input: Quotation PDF

[2126] Output: Download link (Example: https: / / example.com / downloads / quote_2023.pdf)

[2127] Step 9:

[2128] server

[2129] The server sends the generated download link to the user's device.

[2130] Input: Download link

[2131] Output: Notification to user terminal

[2132] Step 10:

[2133] User

[2134] The user clicks the received download link to download the quote.

[2135] Input: Download link

[2136] Output: Download quotation

[2137] Follow-up schedule management

[2138] Step 1:

[2139] User

[2140] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[2141] Input: "Please send a sample."

[2142] Output: Follow-up request

[2143] Step 2:

[2144] terminal

[2145] The device saves requests that require follow-up and sends them to the server.

[2146] Input: Follow-up request

[2147] Output: Request to send to server

[2148] Step 3:

[2149] server

[2150] The server analyzes the received request and determines whether follow-up is necessary.

[2151] Input: Follow-up request

[2152] Output: Analysis results (e.g., "Sample submission required")

[2153] Step 4:

[2154] server

[2155] The server uses IBM Watson's emotion engine to recognize the user's emotions.

[2156] Input: Analysis results

[2157] Output: Sentimental data (e.g., urgency)

[2158] Step 5:

[2159] server

[2160] The server adjusts the priority and method of follow-up based on the recognized sentiment data.

[2161] Input: Sentiment data

[2162] Output: Adjustment data

[2163] Step 6:

[2164] server

[2165] The server uses the Google Calendar API to set the date and time that requires follow-up and registers it in the schedule management database.

[2166] Input: Adjustment data

[2167] Output: Follow-up schedule (Example: 3 days from now, 3 PM)

[2168] Step 7:

[2169] server

[2170] As the scheduled date and time approach, Firebase Cloud Messaging is used to generate a push notification as a reminder, which is then sent to the user and the responsible party.

[2171] Input: Follow-up schedule

[2172] Output: Push notification

[2173] Step 8:

[2174] terminal

[2175] Users and staff will receive push notifications and take appropriate action.

[2176] Input: Push notification

[2177] Output: Follow-up available

[2178] (Application Example 2)

[2179] 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."

[2180] In modern society, companies are required to improve the efficiency and quality of their customer service operations. In particular, electronic payment services demand prompt and appropriate customer service, while simultaneously making it difficult to consider customer emotions. Conventional systems could only provide mechanical responses to inquiries, making it difficult to increase customer satisfaction. Furthermore, tasks such as generating quotations and managing follow-ups required manual handling, resulting in a significant workload. This invention aims to solve these problems by providing a system that recognizes customer emotions and responds accordingly.

[2181] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries and user sentiment data, means for automatically generating and sending a response that reflects the sentiment based on the search results and sentiment data to the customer, and means for automatically generating an electronic receipt and providing the customer with a downloadable link. This makes it possible to provide a quick and appropriate response while taking customer sentiment into consideration.

[2182] "Internal company data" refers to all information generated and stored within a company, including customer data, product data, inventory data, and sales history.

[2183] "Customer inquiries" refer to requests for information, questions, and feedback that customers send to a company, and include those sent via telephone, email, web forms, etc.

[2184] "Analysis" refers to the process of deciphering and understanding received data, and performing operations to grasp its intent and meaning.

[2185] A "related database" refers to a collection of data that stores specific information and is structured in a way that allows for later searching and referencing.

[2186] "Emotional data" refers to data that identifies a user's emotional state (e.g., joy, anger, surprise, etc.) using natural language processing and emotion recognition technologies, and expresses it as numerical values ​​or tags.

[2187] "Automatically generating responses" refers to a system generating reply text without human intervention, based on pre-configured algorithms and rules.

[2188] An "electronic receipt" refers to a receipt issued in digital format, usually generated in PDF format, and provided via email or a download link.

[2189] A "quotation" is a document that shows the price and conditions of the products or services to be offered in advance, and serves as a basis for customers to consider purchasing.

[2190] A "downloadable link" refers to a URL (web address) from which a specific file or information can be obtained via the internet.

[2191] "Follow-up" refers to additional support and verification work carried out continuously after the initial response, and is a process aimed at maintaining and improving customer satisfaction.

[2192] "Push notifications as reminders" refers to displaying short messages on a user's smartphone or other device to alert them based on a specified date, time, or conditions.

[2193] This invention relates to a system for automating and streamlining customer service in electronic payment services. In particular, it aims to improve customer satisfaction by recognizing customer emotions and providing appropriate responses and follow-ups.

[2194] Hardware configuration

[2195] This invention is carried out using the following hardware:

[2196] Smartphone (iOS or Android)

[2197] Cloud servers (e.g., AWS or Google Cloud)

[2198] Software Configuration

[2199] This invention is carried out using the following software:

[2200] Smartphone applications (iOS: Swift, Android: Kotlin)

[2201] Server-side: Node.js, Python

[2202] Database: MySQL or PostgreSQL

[2203] Emotion recognition engine: IBM Watson Tone Analyzer or Google Cloud Natural Language API

[2204] Push notification service: Firebase Cloud Messaging (FCM)

[2205] Overview of Data Processing and Data Calculation

[2206] 1. User submits inquiry:

[2207] The user enters an inquiry about electronic payments via a smartphone app and clicks the submit button. The inquiry data is sent from the device to the server.

[2208] 2. Reception and analysis on the server:

[2209] The server receives the query data and extracts keywords using natural language processing techniques. Furthermore, it analyzes the user's sentiment data using an emotion recognition engine (IBM Watson Tone Analyzer or Google Cloud Natural Language API).

[2210] 3. Generating and sending response messages:

[2211] The server searches the electronic payment database for corresponding information based on the analysis results and sentiment data. It then automatically generates a response message that reflects the sentiment and sends it to the user's smartphone.

[2212] 4. Automatic generation of electronic receipts:

[2213] When a user completes a purchase, the server automatically generates an electronic receipt in PDF format. It then generates a download link for the electronic receipt and provides it to the user.

[2214] 5. Follow-up schedule management:

[2215] The server prioritizes follow-ups based on user sentiment data. The cloud server manages the dates and times when follow-ups are needed, and push notifications are sent as reminders to the user's and the assigned staff member's devices as the scheduled time approaches.

[2216] Specific example

[2217] For example, if a user sends an inquiry asking, "Please tell me my recent transaction history," the server uses its emotion recognition engine to recognize the user's emotion as "high stress." Next, the server retrieves the corresponding transaction history from the database and generates and sends a response message in polite language to alleviate the stress, such as, "Here is your transaction history. Please feel free to contact us anytime if you have any further questions."

[2218] Example of a prompt

[2219] "Please tell me your most recent credit card transaction history."

[2220] "Please resend the most recent receipt via email."

[2221] "I'd like to confirm the payment details."

[2222] This invention makes it possible to recognize user emotions and respond accordingly. This system not only increases customer satisfaction but also streamlines a company's customer service operations.

[2223] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[2224] Step 1:

[2225] The user enters their inquiry and clicks the submit button. The "inquiry details" are retrieved from the user's device as input data. This inquiry is then sent to the server via the smartphone.

[2226] Step 2:

[2227] The system analyzes query data received by the server. The input is the query data, and keywords are extracted using a natural language processing tool (e.g., Python's NLTK). The output is the extracted keywords.

[2228] Step 3:

[2229] The server uses an emotion recognition engine to analyze the user's emotional data. The input is "data from the inquiry," and the emotions are analyzed using an emotion recognition tool (e.g., IBM Watson Tone Analyzer). The output is "recognized emotional data."

[2230] Step 4:

[2231] The server performs a search of relevant databases based on the analyzed keywords and sentiment data. The inputs are "extracted keywords" and "recognized sentiment data," and the server queries the databases. The output is the "search results."

[2232] Step 5:

[2233] The server generates a response message that reflects emotions based on search results and sentiment data. The inputs are "search results" and "recognized sentiment data," and the server generates the response message using natural language generation technology (e.g., the GPT model). The output is the "response message."

[2234] Step 6:

[2235] The server sends the generated response message to the user's terminal. The input is the "response message," and the output is the "response message sent to the user's terminal."

[2236] Step 7:

[2237] When a user completes a purchase, the server automatically generates an electronic receipt. The input is "purchase data," and the electronic receipt is generated using a PDF generation tool. The output is a "PDF file of the electronic receipt."

[2238] Step 8:

[2239] The server generates a download link for the electronic receipt and provides it to the user. The input is a "PDF file of the electronic receipt," and the output is a "download link for the electronic receipt."

[2240] Step 9:

[2241] The server manages the schedule for follow-ups when necessary. Inputs are "follow-up request data" and "recognized sentiment data," and the server sets the follow-up priority. The output is the "configured follow-up schedule."

[2242] Step 10:

[2243] The server generates a push notification as a reminder as the scheduled date and time approaches. The input is the "follow-up schedule data," and the output is the "generated push notification." The notification is sent to the user's and the assigned person's devices.

[2244] 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.

[2245] 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.

[2246] 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.

[2247] [Fourth Embodiment]

[2248] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[2249] 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.

[2250] 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).

[2251] 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.

[2252] 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.

[2253] 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).

[2254] 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.

[2255] 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.

[2256] 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.

[2257] 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.

[2258] 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.

[2259] 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.

[2260] 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".

[2261] This invention is a system that utilizes internal company data to centrally manage automated responses to customer inquiries, automated generation of quotations, and follow-up schedules.

[2262] 1. Automated response to customer inquiries

[2263] User

[2264] 1. The user enters their inquiry details on their device and clicks the send button.

[2265] 2. For example, a user enters "I would like to know the stock status of product A."

[2266] terminal

[2267] 1. The terminal saves the user's input and sends the inquiry data to the server.

[2268] server

[2269] 1. The server analyzes the received query data.

[2270] 2. Based on the analysis results, search for relevant databases (e.g., inventory databases).

[2271] 3. Automatically generate a response message based on the search results (e.g., availability of product A).

[2272] 4. Send the generated response message to the user's terminal.

[2273] 2. Automatic generation of quotations

[2274] User

[2275] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[2276] 2. For example, you could request, "Please issue a quotation for product B."

[2277] terminal

[2278] 1. The terminal saves the request data and sends it to the server.

[2279] server

[2280] 1. The server parses the received request.

[2281] 2. Search the approval database for price data of the target company and product.

[2282] 3. Automatically generate a quotation based on the search results. (Example: PDF format)

[2283] 4. Save the generated quotation and create a download link.

[2284] 5. Send the download link to the user's device.

[2285] User

[2286] 1. The user clicks the received download link to download the quotation.

[2287] 3. Schedule management for follow-up

[2288] User

[2289] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[2290] terminal

[2291] 1. The device sends requests requiring follow-up to the server.

[2292] server

[2293] 1. The server analyzes the received request and determines whether follow-up is necessary.

[2294] 2. Set the date and time for follow-up and register it in the schedule management database.

[2295] 3. Generate a push notification as a reminder based on the set date and time.

[2296] 4. Send push notifications to the user's and the person in charge's devices.

[2297] terminal

[2298] 1. Users and staff will receive push notifications and take appropriate action.

[2299] Specific example

[2300] For example, consider a case where a user submits an inquiry requesting "Please send me a sample of product C." In this case, the server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a schedule for follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and, based on the instructions, confirm the sample shipment and follow up.

[2301] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and schedule management to automate and streamline customer service operations, thereby achieving fast and accurate customer service.

[2302] The following describes the processing flow.

[2303] Automated response to customer inquiries

[2304] Step 1:

[2305] The user enters their inquiry details on their device and clicks the send button.

[2306] Step 2:

[2307] The terminal saves the user's input and sends the inquiry data to the server.

[2308] Step 3:

[2309] The server analyzes the received query data and extracts keywords and related information.

[2310] Step 4:

[2311] Based on the analysis results, the server searches relevant databases (e.g., inventory databases).

[2312] Step 5:

[2313] The server generates an appropriate response message based on the search results.

[2314] Step 6:

[2315] The server sends the generated response message to the terminal.

[2316] Step 7:

[2317] The terminal receives a response from the server and displays it to the user.

[2318] Automatic generation of quotations

[2319] Step 1:

[2320] The user enters a request to request a quote and clicks the submit button.

[2321] Step 2:

[2322] The terminal saves the request data and sends it to the server.

[2323] Step 3:

[2324] The server analyzes the received request and extracts price data for the target company and product.

[2325] Step 4:

[2326] The server retrieves price information from the approval database.

[2327] Step 5:

[2328] The server automatically generates a quote based on the price information (e.g., in PDF format).

[2329] Step 6:

[2330] The server saves the generated quote and creates a download link.

[2331] Step 7:

[2332] The server sends a download link to the device.

[2333] Step 8:

[2334] The device receives the download link and displays it to the user.

[2335] Step 9:

[2336] The user clicks the download link to download the quote.

[2337] Follow-up schedule management

[2338] Step 1:

[2339] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[2340] Step 2:

[2341] The device sends requests to the server that require follow-up.

[2342] Step 3:

[2343] The server analyzes the received request and determines whether follow-up is necessary.

[2344] Step 4:

[2345] The server sets the date and time for follow-up and registers it in the schedule management database.

[2346] Step 5:

[2347] The server generates a push notification as a reminder when the scheduled date and time approach.

[2348] Step 6:

[2349] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[2350] Step 7:

[2351] The device receives push notifications and displays them to the user and the person in charge.

[2352] Step 8:

[2353] Users and their representatives will take appropriate follow-up actions based on push notifications.

[2354] (Example 1)

[2355] 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".

[2356] In corporate customer service operations, traditional manual responses are time-consuming and labor-intensive, leading to decreased customer satisfaction. In particular, efficiently managing inquiries, issuing quotations, and follow-ups is difficult, and there is a demand for quick and accurate responses. Furthermore, the lack of a unified management system for these tasks necessitates information integration and automation.

[2357] 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.

[2358] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for generating response messages using a generation AI model. This enables rapid and accurate automated responses to customer inquiries.

[2359] In this invention, the server includes means for searching a database of prices offered by each company and automatically generating a quotation in a format accessible to the customer; means for generating a downloadable link for the generated quotation and providing it to the customer; means for downloading the quotation via the provided link; and means for adjusting the content of the quotation using a generation AI model. This enables the rapid and automatic creation and provision of quotations.

[2360] In this invention, the server includes means for tracking the processing status of customer inquiries and setting a date and time when follow-up is required; means for sending push notifications as reminders to customers and personnel based on the set date and time; means for receiving push notifications; and means for generating the content of the reminder using a prompt statement. This enables efficient scheduling of follow-ups and timely responses to be taken.

[2361] These measures enable companies to automate and streamline customer service operations, contributing to improved customer satisfaction.

[2362] "Internal company data" refers to information and documents generated and stored within a company.

[2363] "Customer inquiries" refer to information about questions and requests that customers make to a company.

[2364] "Means of receiving and analyzing data" refers to the technology that allows a system to take in data from an external source and understand its contents.

[2365] A "related database" refers to a database that stores information related to a specific query.

[2366] "Searching methods" refer to techniques for finding information within a database based on specific criteria.

[2367] "Means of generating responses" refers to technologies that automatically create appropriate answers to inquiries.

[2368] A "generative AI model" refers to an algorithm that uses artificial intelligence to generate new information from data.

[2369] A "downloadable link" refers to a web address that allows a user to download a specific file by clicking on it.

[2370] A "quote" refers to a document that lists the price, details, and terms and conditions of a product or service.

[2371] "Means of tracking" refers to technologies that track and record the progress of a specific process or event.

[2372] "Follow-up" refers to additional actions or confirmations taken after the initial response or action.

[2373] "Push notifications" refer to a technology that sends information from a server to a device in real time.

[2374] A "prompt statement" refers to an input statement used to instruct a generative AI model to generate information.

[2375] This invention is a system that automates and streamlines customer service operations by utilizing internal company data. The program processing of this system is described in detail below.

[2376] Automated response to customer inquiries

[2377] 1. User

[2378] The user enters their inquiry details on their device and clicks the "Send" button.

[2379] For example, you would enter an inquiry such as, "Please tell me the stock status of product A."

[2380] 2. Terminal

[2381] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[2382] 3. Server

[2383] The server parses the received query data and uses natural language processing libraries (e.g., spaCy or NLTK) to analyze the information the user is seeking.

[2384] Based on the analysis results, the system searches relevant databases (e.g., inventory databases) and issues SQL queries.

[2385] Based on the search results, a prompt is sent to a generative AI model (e.g., OpenAI's GPT-4) to generate a response message. Example prompt: "How much stock do you currently have of product A?"

[2386] The generated response message is sent to the user's terminal in JSON format.

[2387] 4. Terminal

[2388] The terminal displays the received response message on the user interface.

[2389] Automatic generation of quotations

[2390] 1. User

[2391] The user enters the request for a quote and clicks the "Submit" button.

[2392] For example, you might enter, "Please issue a quotation for product B."

[2393] 2. Terminal

[2394] The terminal saves the request data to a local database and then sends it to the server.

[2395] 3. Server

[2396] The server analyzes the request data and extracts the necessary information using a natural language processing library.

[2397] Search the price database and issue an SQL query. Example SQL query: "SELECT price FROM products WHERE product_name='product B'"

[2398] Based on the search results, the system automatically generates a quotation in PDF format using libraries such as Apache PDFBox.

[2399] The generated quote is saved to cloud storage (e.g., AWS S3), and a downloadable link is generated and sent to the user's device.

[2400] 4. User

[2401] The user clicks the provided link and downloads the quote.

[2402] Follow-up schedule management

[2403] 1. User

[2404] The user enters a request that requires follow-up and clicks the "Send" button.

[2405] For example, you might enter, "Please send me a sample of product C."

[2406] 2. Terminal

[2407] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[2408] 3. Server

[2409] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[2410] Set the necessary follow-up dates and register them in the schedule management database (e.g., Microsoft SQL Server).

[2411] Using Firebase Cloud Messaging, push notifications are generated at a set date and time and sent to the user's and the assigned person's devices.

[2412] 4. Terminal

[2413] Users and staff members receive push notifications and follow up accordingly.

[2414] Hardware and software used

[2415] Server: Cloud server (e.g., AWS EC2, Microsoft Azure)

[2416] Database management systems: MySQL, Oracle, Microsoft SQL Server

[2417] Natural language processing libraries: spaCy, NLTK

[2418] Generative AI models: OpenAI GPT-4, etc.

[2419] Cloud storage: AWS S3

[2420] PDF generation library: Apache PDFBox

[2421] Push notification service: Firebase Cloud Messaging

[2422] Specific example

[2423] For example, if a user submits an inquiry requesting "Please send me a sample of product C," the following process takes place: The server analyzes the inquiry and determines that follow-up is necessary for sending the sample. The server sets a follow-up appointment in the schedule management database for three days later and sends a push notification as a reminder three days later. The user and the person in charge receive the push notification and confirm and carry out the sample shipment based on the instructions.

[2424] As described above, the system of the present invention integrates the use of internal data, automated responses, automated quotation generation, and follow-up schedule management to automate and streamline customer service operations, thereby achieving prompt and accurate customer service.

[2425] The flow of the specific processing in Example 1 will be explained using Figure 11.

[2426] Processing steps for automated responses to customer inquiries

[2427] Step 1:

[2428] User

[2429] The user enters their inquiry details on their device and clicks the "Send" button.

[2430] Input: User inquiry (e.g., "Please tell me the stock status of product A")

[2431] Output: A request is generated to send query data to the terminal.

[2432] Step 2:

[2433] terminal

[2434] The terminal temporarily stores the entered query data in a local database and sends it to the server in JSON format.

[2435] Input: User inquiry data

[2436] Output: Send query data to the server in JSON format.

[2437] Step 3:

[2438] server

[2439] The server analyzes the received query data and uses a natural language processing library to understand the intent of the query.

[2440] Software used: spaCy, NLTK

[2441] Input: Query data in JSON format

[2442] Data processing: Text analysis using natural language processing

[2443] Output: Analysis results of the inquiry

[2444] Step 4:

[2445] server

[2446] Based on the analysis results, SQL queries are issued to relevant databases (e.g., inventory databases) to perform searches.

[2447] Database management system used: MySQL

[2448] Input: Analysis results of the inquiry content

[2449] Data Calculation: Searching Related Databases

[2450] Output: Database search results (e.g., "Inventory quantity of product A")

[2451] Step 5:

[2452] server

[2453] A response message is generated using an AI model based on the database search results.

[2454] Generative AI model used: OpenAI GPT-4

[2455] Input: Database search results

[2456] Data processing: Generating response messages using a generative AI model.

[2457] Output: Response message (Example: "Product A is in stock")

[2458] Step 6:

[2459] server

[2460] The server sends the generated response message to the user's terminal in JSON format.

[2461] Input: Response message

[2462] Output: Send a JSON-formatted response message to the user's terminal.

[2463] Step 7:

[2464] terminal

[2465] The terminal displays the received response message on the user interface.

[2466] Input: Response message in JSON format

[2467] Output: Display in the user interface

[2468] Process steps for automatic quotation generation

[2469] Step 1:

[2470] User

[2471] The user enters the request for a quote and clicks the "Submit" button.

[2472] Input: Product name and quotation request (Example: "Please issue a quotation for product B")

[2473] Output: Data is input to the terminal, and a transmission command is generated.

[2474] Step 2:

[2475] terminal

[2476] The terminal saves the request data to a local database and then sends it to the server.

[2477] Input: User's request data

[2478] Output: Send request data to the server in JSON format.

[2479] Step 3:

[2480] server

[2481] The server parses the request data and searches the database to retrieve relevant pricing information.

[2482] Software used: Natural language processing libraries (spaCy, NLTK)

[2483] Input: Request data

[2484] Data processing: Text analysis using natural language processing

[2485] Output: Analyzed request content

[2486] Step 4:

[2487] server

[2488] Based on the parsed request, an SQL query is issued to the price database to retrieve the necessary price information.

[2489] Database management system used: Oracle

[2490] Input: Parsed request content

[2491] Data Calculation: Searching Price Databases

[2492] Output: Price data (Example: "Price information for product B")

[2493] Step 5:

[2494] server

[2495] Based on the acquired price data, an AI model is used to automatically generate a quotation.

[2496] Generative AI model used: OpenAI GPT-4

[2497] PDF generation library used: Apache PDFBox

[2498] Input: Price data

[2499] Data calculation: Generating a PDF quotation

[2500] Output: Generated quotation (PDF format)

[2501] Step 6:

[2502] server

[2503] The generated quote is saved to cloud storage, and a downloadable link is created and sent to the user's device.

[2504] Cloud storage to use: AWS S3

[2505] Input: Generated quotation (PDF format)

[2506] Output: Download link

[2507] Step 7:

[2508] User

[2509] The user clicks the provided link and downloads the quote.

[2510] Input: Download link

[2511] Output: Downloaded quotation (PDF format)

[2512] Follow-up scheduling process steps

[2513] Step 1:

[2514] User

[2515] The user enters a request that requires follow-up and clicks the "Send" button.

[2516] Input: Follow-up request (Example: "Please send a sample of product C")

[2517] Output: Data is input to the terminal, and a transmission command is generated.

[2518] Step 2:

[2519] terminal

[2520] The terminal saves requests requiring follow-up to a local database and sends them to the server.

[2521] Input: Request data requiring follow-up

[2522] Output: Send request data to the server in JSON format.

[2523] Step 3:

[2524] server

[2525] The server analyzes the follow-up request and uses a natural language processing library to determine whether follow-up is necessary.

[2526] Software used: Natural language processing libraries (spaCy, NLTK)

[2527] Input: Follow-up request data

[2528] Data processing: Text analysis using natural language processing

[2529] Output: Result of the assessment of the need for follow-up.

[2530] Step 4:

[2531] server

[2532] Set the date and time for follow-up and register it in the schedule management database.

[2533] Database management system used: Microsoft SQL Server

[2534] Input: Result of the assessment of the need for follow-up.

[2535] Data processing: Setting and registering follow-up dates and times.

[2536] Output: Configured follow-up schedule

[2537] Step 5:

[2538] server

[2539] Based on the configured date and time, push notifications are generated using Firebase Cloud Messaging and sent to the user's and assigned personnel's devices.

[2540] Push notification service used: Firebase Cloud Messaging

[2541] Input: Configured follow-up schedule

[2542] Data processing: Generating and sending push notifications

[2543] Output: Push notifications to user and staff terminals

[2544] Step 6:

[2545] terminal

[2546] Users and staff members receive push notifications and follow up accordingly.

[2547] Input: Push notification

[2548] Output: Follow-up will be performed.

[2549] Through the specific processing steps outlined above, the system can automate and streamline customer service operations.

[2550] (Application Example 1)

[2551] 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".

[2552] Traditional e-commerce sites struggled to respond quickly and accurately to customer inquiries, particularly with manual processes like checking inventory and issuing quotes, resulting in inefficiencies. Furthermore, managing follow-up schedules was cumbersome, often leading to delays in notifications to both customers and staff. This resulted in decreased customer satisfaction and missed sales opportunities. There is a need to solve these problems and automate and streamline customer service.

[2553] 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.

[2554] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data; means for searching relevant databases based on the analyzed inquiries; means for automatically generating and sending responses to customers based on the search results; means for analyzing customer inquiries regarding specific product information or inventory status using natural language processing; means for performing natural language processing using a generative AI model to automatically generate appropriate responses; means for searching each company's pricing database and automatically generating quotations in a format accessible to customers; means for generating and providing downloadable links for the generated quotations to customers; means for receiving quotation requests, analyzing product information and pricing data using a generative AI model, and automatically generating quotations; means for tracking the processing status of customer inquiries and setting dates and times for follow-up; means for sending push notifications to customers and personnel as reminders based on the set dates and times; and means for using a schedule management database to automatically generate reminders when follow-up is required. This enables not only a quick and accurate response to customer inquiries but also efficient automatic quotation generation and follow-up schedule management.

[2555] "Internal company data" refers to the collection of all information and data managed within a company.

[2556] An "inquiry" refers to a question or request sent by a customer to confirm product information, stock availability, etc.

[2557] "Analysis" is the process of breaking down the content of an received inquiry and interpreting its meaning and intent.

[2558] A "database" is a system that stores and manages related information in an organization.

[2559] "Searching" is the process of finding specific information within a database.

[2560] "Response" refers to the answers or information provided in response to customer inquiries.

[2561] "Natural language processing" is the technology that enables computers to understand and process human language.

[2562] A "generative AI model" refers to an algorithm or framework that uses artificial intelligence to generate new text or data.

[2563] A "price database" is a database where product price information is stored in an organ.

[2564] A "quotation" is a document that shows the price and conditions of a specific product or service.

[2565] "Follow-up" refers to additional confirmations or follow-ups conducted after the initial response.

[2566] A "schedule management database" is a system that organizes and stores follow-up dates and appointments.

[2567] A "reminder" is a message sent to notify you of an action that needs to be taken at a specific date and time.

[2568] A "push notification" is a notification message that is sent to a user's device in real time.

[2569] The specific system for implementing the present invention is capable of efficiently and quickly processing customer inquiries and easily managing the automatic generation of quotations and follow-up schedules. This system is configured as follows.

[2570] Automated response to customer inquiries

[2571] The server automatically receives and analyzes customer inquiries using internal company data. It receives the inquiry content entered by the customer on their device and sends it to the server. The server uses Google Cloud AI to perform natural language processing and analyze the inquiry content. It then searches relevant databases (e.g., product inventory database) and automatically generates an appropriate response. The generated response is sent to the customer's device.

[2572] Specific example

[2573] For example, if a customer asks, "What is the stock status of product A?", the server analyzes the inquiry, retrieves the stock status of product A from the product inventory database, and automatically generates a response message saying, "Product A is currently in stock."

[2574] Example of a prompt:

[2575] Inquiry: "Please tell me the stock status of product A."

[2576] Generated AI prompt: "Check the inventory status of product A and respond to the customer whether it is in stock or not."

[2577] Automatic generation of quotations

[2578] When a customer requests a quote for a specific product, the request data is sent from the terminal to the server. The server receives and analyzes the request. Next, it retrieves the price data for the product from the price database and automatically generates a quote using a generation AI model. The generated quote is saved in PDF format, and a download link is sent to the customer.

[2579] Specific example

[2580] When a customer requests "Please issue a quote for product B," the server analyzes the request, retrieves the price information for product B from the price database, and automatically generates a quote. A download link for the generated quote is then sent to the customer's device.

[2581] Example of a prompt:

[2582] Quote Request: "Please issue a quote for product B."

[2583] Generated AI prompt: "Retrieve pricing data for product B, generate a quote, and provide it to the customer."

[2584] Follow-up schedule management

[2585] If a customer requires follow-up on an inquiry, they send a follow-up request from their device to the server. The server parses this request and sets a date and time for the follow-up. This information is registered in the schedule management database, and a push notification is generated as a reminder based on the set date and time. This push notification is sent to the customer's and the agent's devices.

[2586] Specific example

[2587] For example, if a customer requests a sample of product C, the server analyzes the request and determines that follow-up is necessary. It sets a schedule for a follow-up three days later and registers it in the schedule management database. Three days later, a push notification is automatically sent as a reminder.

[2588] Example of a prompt:

[2589] Follow-up request: "Please send me a sample of product C."

[2590] Generated AI prompt: "Please schedule a follow-up for sending a sample of Product C in 3 days and send a reminder."

[2591] As described above, the present invention is a system that enables rapid and accurate responses to customer inquiries, automatic generation of quotations, and efficient scheduling of follow-ups. This improves customer satisfaction and streamlines operations.

[2592] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[2593] Processing steps for automated responses to customer inquiries

[2594] Step 1:

[2595] The user enters their inquiry details on their device and clicks the send button. An example of input data is, "Please tell me the stock status of product A." Input: Inquiry details. Output: Sending of inquiry data.

[2596] Step 2:

[2597] The terminal saves user input and sends query data to the server. Network communication is performed for data transfer. Input: Query data. Output: Data sent to the server.

[2598] Step 3:

[2599] The server receives query data, which is then processed using natural language processing and analysis with Google Cloud AI. Input: Query data. Output: Analysis results.

[2600] Step 4:

[2601] The server searches relevant databases (e.g., inventory databases) based on the analysis results and retrieves the necessary information. Input: Analysis results. Output: Search results.

[2602] Step 5:

[2603] The server automatically generates response messages based on search results, using a generation AI model. Input: Search results. Output: Response message.

[2604] Step 6:

[2605] The server generates a response message and sends it to the user's terminal. Input: Response message. Output: Message sent to the user's terminal.

[2606] Process steps for automatic quotation generation

[2607] Step 1:

[2608] The user requests a quotation and clicks the send button on their terminal. Example input: "Please issue a quotation for product B." Input: Quotation request. Output: Quotation request data sent.

[2609] Step 2:

[2610] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Estimate request data. Output: Data sent to the server.

[2611] Step 3:

[2612] The server analyzes the request and uses a generated AI model to retrieve product information and pricing data. Input: Quotation request data. Output: Analysis results and pricing information.

[2613] Step 4:

[2614] The server automatically generates a quotation based on the analysis results and saves it in PDF format. Input: Analysis results and price information. Output: Quotation in PDF format.

[2615] Step 5:

[2616] The server generates a download link for the generated quotation and sends it to the user's device. Input: Quotation in PDF format. Output: Sending of download link.

[2617] Follow-up scheduling process steps

[2618] Step 1:

[2619] The user enters a request requiring follow-up and clicks the send button on their device. Example input: "Please send a sample of product C." Input: Follow-up request. Output: Sending of follow-up request data.

[2620] Step 2:

[2621] The terminal saves the request data and sends it to the server. Network communication is performed for data transfer. Input: Follow-up request data. Output: Data sent to the server.

[2622] Step 3:

[2623] The server analyzes the received request data and sets the date and time for follow-up. Input: Follow-up request data. Output: Follow-up schedule.

[2624] Step 4:

[2625] The server registers the follow-up date and time in the schedule management database. Input: Follow-up schedule. Output: Schedule registration information.

[2626] Step 5:

[2627] The server generates a push notification as a reminder based on the configured date and time. Input: Follow-up schedule. Output: Push notification.

[2628] Step 6:

[2629] The server generates push notifications and sends them to the user's and the assigned personnel's devices. Input: Push notification. Output: Notification sent to the user's and the assigned personnel's devices.

[2630] 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.

[2631] This invention is a system that utilizes internal company data to automatically respond to customer inquiries, automatically generate quotations, manage follow-up schedules, and combines this with an emotion engine that recognizes user emotions.

[2632] 1. Automated response to customer inquiries

[2633] User

[2634] 1. The user enters their inquiry details on their device and clicks the send button.

[2635] 2. For example, a user enters "I would like to know the stock status of product A."

[2636] terminal

[2637] 1. The terminal saves the user's input and sends the inquiry data to the server.

[2638] server

[2639] 1. The server analyzes the received query data and extracts keywords and related information.

[2640] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[2641] 3. Based on the recognized emotions, generate data to adjust the response message.

[2642] 4. Based on the analysis results and sentiment data, search relevant databases (e.g., inventory databases).

[2643] 5. Based on the search results, generate appropriate response messages that reflect sentiment data.

[2644] 6. Send the generated response message to the user's terminal.

[2645] 2. Automatic generation of quotations

[2646] User

[2647] 1. The user enters a request to request the issuance of a quotation and clicks the submit button.

[2648] 2. For example, you could request, "Please issue a quotation for product B."

[2649] terminal

[2650] 1. The terminal saves the request data and sends it to the server.

[2651] server

[2652] 1. The server analyzes the received request and extracts price data for the target company and product.

[2653] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[2654] 3. Based on the recognized emotion data, generate data to adjust the contents of the quotation.

[2655] 4. Retrieve price information from the approval database.

[2656] 5. Automatically generate quotes based on price information and sentiment data (e.g., in PDF format).

[2657] 6. Save the generated quotation and create a download link.

[2658] 7. Send the download link to the user's device.

[2659] User

[2660] 1. The user clicks the received download link to download the quotation.

[2661] 3. Schedule management for follow-up

[2662] User

[2663] 1. If follow-up is needed when a user submits an inquiry, for example, request "Please send a sample."

[2664] terminal

[2665] 1. The device sends requests requiring follow-up to the server.

[2666] server

[2667] 1. The server analyzes the received request and determines whether follow-up is necessary.

[2668] 2. Furthermore, an emotion engine is used to recognize the user's emotions.

[2669] 3. Based on the recognized emotional data, adjust the priorities and methods of follow-up.

[2670] 4. Set the date and time for follow-up and register it in the schedule management database.

[2671] 5. When the set date and time approach, generate a push notification as a reminder.

[2672] 6. Send the generated push notification to the user's and the person in charge's devices.

[2673] terminal

[2674] 1. Users and staff will receive push notifications and take appropriate action.

[2675] Specific example

[2676] For example, consider a case where a user submits an inquiry requesting a sample of product C. In this case, the server analyzes the inquiry and determines that follow-up is necessary to send the sample. It also uses an emotion engine to recognize if the user feels a sense of urgency. The server sets a schedule for a follow-up in three days and sends a push notification as a reminder three days later. The user and the person in charge then confirm the sample delivery and follow up based on these instructions.

[2677] As described above, the system of the present invention can automate and streamline customer service operations, and provide a better customer experience through responses and methods that take into account the user's emotions.

[2678] The following describes the processing flow.

[2679] Automated response to customer inquiries

[2680] Step 1:

[2681] The user enters their inquiry details on their device and clicks the send button.

[2682] Step 2:

[2683] The terminal saves the user's input and sends the inquiry data to the server.

[2684] Step 3:

[2685] The server analyzes the received query data and extracts keywords and related information.

[2686] Step 4:

[2687] The server uses an emotion engine to recognize the user's emotions.

[2688] Step 5:

[2689] The server searches relevant databases (e.g., inventory databases) based on the analysis results and sentiment data.

[2690] Step 6:

[2691] The server generates a response message based on the search results.

[2692] Step 7:

[2693] The server incorporates sentiment data into the generated response message and makes appropriate adjustments.

[2694] Step 8:

[2695] The server sends the generated response message to the terminal.

[2696] Step 9:

[2697] The terminal receives a response from the server and displays it to the user.

[2698] Automatic generation of quotations

[2699] Step 1:

[2700] The user enters a request to request a quote and clicks the submit button.

[2701] Step 2:

[2702] The terminal saves the request data and sends it to the server.

[2703] Step 3:

[2704] The server analyzes the received request and extracts price data for the target company and product.

[2705] Step 4:

[2706] The server uses an emotion engine to recognize the user's emotions.

[2707] Step 5:

[2708] The server retrieves price information from the approval database.

[2709] Step 6:

[2710] The server automatically generates quotes (e.g., in PDF format) based on price information and sentiment data.

[2711] Step 7:

[2712] The server saves the generated quote and creates a download link.

[2713] Step 8:

[2714] The server sends a download link to the device.

[2715] Step 9:

[2716] The device receives the download link and displays it to the user.

[2717] Step 10:

[2718] The user clicks the download link to download the quote.

[2719] Follow-up schedule management

[2720] Step 1:

[2721] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[2722] Step 2:

[2723] The device sends requests to the server that require follow-up.

[2724] Step 3:

[2725] The server analyzes the received request and determines whether follow-up is necessary.

[2726] Step 4:

[2727] The server uses an emotion engine to recognize the user's emotions.

[2728] Step 5:

[2729] The server sets the date and time for follow-up and registers it in the schedule management database.

[2730] Step 6:

[2731] The server generates a push notification as a reminder when the scheduled date and time approach.

[2732] Step 7:

[2733] The server sends the generated push notifications to the user's and the assigned personnel's devices.

[2734] Step 8:

[2735] The device receives push notifications and displays them to the user and the person in charge.

[2736] Step 9:

[2737] Users and their representatives will take appropriate follow-up actions based on push notifications.

[2738] Specifically, the emotion engine analyzes the emotional data of user inquiries and requests, and the server operates to quickly and accurately adjust the response content if, for example, the user is in a hurry, thereby increasing the priority of follow-up. This enables more personalized customer service.

[2739] (Example 2)

[2740] 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".

[2741] Traditional in-house customer service systems have suffered from low accuracy in automated responses to customer inquiries, and have been unable to consider customer emotions when generating quotations or managing follow-up schedules. This can lead to decreased customer satisfaction and delays in response times, potentially negatively impacting a company's credibility and performance. Therefore, there is a need for highly accurate automated responses, quotation generation, and follow-up management that take customer emotions into consideration.

[2742] 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.

[2743] In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries, means for automatically generating and sending responses to customers based on the search results, and means for adjusting the analysis results using an emotion engine that recognizes the user's emotions. This enables a quick and appropriate response that takes customer emotions into consideration.

[2744] "Internal company data" refers to all data collected and stored within a company, including customer information, product information, inquiry history, inventory information, etc.

[2745] An "inquiry" refers to a question or request made by a customer to a company, and includes inquiries about product information, pricing, and services.

[2746] "Analysis" is the process of processing received inquiries and data using analytical tools and algorithms to extract necessary information, keywords, sentiments, and so on.

[2747] A "related database" is a database that stores data searchable based on specific criteria, and may include information on products and services, inventory information, and so on.

[2748] A "response" refers to an answer or message generated based on the analysis results and search results from related databases, and is sent to the customer in the form of email, chat message, push notification, etc.

[2749] An "emotion engine" refers to a technology that recognizes and analyzes emotions from user input text, utilizing natural language processing to determine the user's mood and tone.

[2750] A "quotation" is a document that contains price information for products or services offered to a customer, and is usually automatically generated in PDF format.

[2751] "Follow-up" refers to additional responses to the initial inquiry or request, and includes sending product samples and responding to further inquiries.

[2752] "Push notifications" are notification messages sent to a user's device in real time and are used to provide reminders and important updates.

[2753] This invention is a system that utilizes internal company data to automate responses to customer inquiries, automate the generation of quotations, manage follow-up schedules, and combines an emotion engine that recognizes user emotions. The embodiments of this system are described in detail below.

[2754] Automated response to customer inquiries

[2755] User

[2756] The user enters their inquiry from their device and clicks the send button. For example, the user might enter, "I would like to know the stock status of product A."

[2757] terminal

[2758] The terminal saves user input and sends query data to the server. This communication uses an API endpoint via an internet connection.

[2759] server

[2760] The server analyzes the received query data. This analysis uses text analysis tools such as AWS Textract. Furthermore, it utilizes IBM Watson's natural language processing (NLP) service to recognize the user's sentiment. Based on the recognized sentiment data, the content of the response message is adjusted. Next, the server searches a database such as MySQL to retrieve the necessary information. For example, it searches the inventory status of product A in the inventory database. Based on the search results, an appropriate response message is generated. The generated response message, reflecting the sentiment data, is sent to the user's terminal.

[2761] Automatic generation of quotations

[2762] User

[2763] The user enters a request to issue a quote and clicks the submit button. For example, they might request, "Please issue a quote for product B."

[2764] terminal

[2765] The terminal saves the request data and sends it to the server.

[2766] server

[2767] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product. The server uses IBM Watson's NLP service to recognize sentiment and adjusts the content of the quote based on the sentiment data. The server retrieves price information from the approval database using SQL queries and generates a quote in PDF format using the ReportLab library. The generated quote is saved on the server, and a download link is created. The download link is sent to the user's terminal.

[2768] User

[2769] The user clicks the received download link to download the quote.

[2770] Follow-up schedule management

[2771] User

[2772] When a user submits an inquiry, if follow-up is needed, for example, they might request, "Please send me a sample."

[2773] terminal

[2774] The device saves requests that require follow-up and sends them to the server.

[2775] server

[2776] The server analyzes the received request and determines the need for follow-up. Furthermore, it uses IBM Watson's sentiment engine to recognize the user's emotions. Based on the recognized emotion data, it adjusts the priority and method of follow-up. The server uses the Google Calendar API to set the date and time for follow-up and registers it in the schedule management database. As the set date and time approach, it generates a push notification using Firebase Cloud Messaging and sends it to the user's and assigned personnel's devices.

[2777] terminal

[2778] Users and staff will receive push notifications and take appropriate action.

[2779] Specific example

[2780] For example, if a user sends an inquiry saying, "Please send me a sample of product C," the server analyzes the inquiry and determines that follow-up is necessary to send the sample. Furthermore, if the emotion engine detects that the user is feeling urgent, the server will set an emergency follow-up for 3 days later at 3 PM and send a push notification as a reminder. Based on these instructions, the user and the person in charge will confirm the sample delivery and follow up.

[2781] Example of a prompt

[2782] The following are examples of prompts to input into the generating AI model.

[2783] You are an engineer developing a system to automate customer service within a company. This system will include automated responses to customer inquiries, automated quotation generation, and follow-up scheduling. Please describe the appropriate processing flow and the technologies to be used to meet the following requirements.

[2784] Request: A user has submitted an inquiry about the stock status of product A. Please describe the system flow for handling this inquiry.

[2785] As described above, the present invention automates and streamlines customer service operations, enabling responses that take user emotions into consideration. As a result, higher customer satisfaction can be achieved, and the company's credibility can be improved.

[2786] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2787] Automated response to customer inquiries

[2788] Step 1:

[2789] User

[2790] The user enters their inquiry details on their device and clicks the send button.

[2791] Input: "Please tell me the stock status of product A."

[2792] Output: Query data

[2793] Step 2:

[2794] terminal

[2795] The terminal saves the user's input and sends the inquiry data to the server.

[2796] Input: Inquiry data

[2797] Output: Request to send to server

[2798] Step 3:

[2799] server

[2800] The server analyzes the received query data. AWS Textract is used for this analysis, extracting keywords from the text.

[2801] Input: Submitted inquiry data

[2802] Output: Analysis results (e.g., "Product A", "Inventory Status")

[2803] Step 4:

[2804] server

[2805] The server uses IBM Watson's NLP service to recognize the emotions contained in the query.

[2806] Input: Analysis results

[2807] Output: Sentiment data (e.g., confusion, interest)

[2808] Step 5:

[2809] server

[2810] The server generates data to adjust response messages based on recognized emotion data.

[2811] Input: Sentiment data

[2812] Output: Adjustment data

[2813] Step 6:

[2814] server

[2815] The server searches databases such as MySQL and retrieves relevant information (such as inventory information). It then executes SQL queries to extract the necessary data.

[2816] Input: Query to search for "Inventory information for product A"

[2817] Output: Inventory information (e.g., In stock)

[2818] Step 7:

[2819] server

[2820] The server generates an appropriate response message based on inventory information and sentiment data.

[2821] Input: Inventory information, adjustment data

[2822] Output: Response message (Example: "Thank you. Product A is currently in stock.")

[2823] Step 8:

[2824] server

[2825] The server sends the generated response message to the user's terminal.

[2826] Input: Response message

[2827] Output: Notification to user terminal

[2828] Automatic generation of quotations

[2829] Step 1:

[2830] User

[2831] The user enters a request to request a quote and clicks the submit button.

[2832] Input: "Please issue a quotation for product B."

[2833] Output: Request data

[2834] Step 2:

[2835] terminal

[2836] The terminal saves the request data and sends it to the server.

[2837] Input: Request data

[2838] Output: Request to send to server

[2839] Step 3:

[2840] server

[2841] The server analyzes the received request using the Python Pandas library and extracts price data for the target company and product.

[2842] Input: Request data

[2843] Output: Analysis results (e.g., target company, product information)

[2844] Step 4:

[2845] server

[2846] The server uses IBM Watson's NLP service to recognize the user's emotions.

[2847] Input: Analysis results

[2848] Output: Sentiment data (e.g., interest, expectation)

[2849] Step 5:

[2850] server

[2851] The server generates data to adjust the contents of the estimate based on the recognized emotion data.

[2852] Input: Sentiment data

[2853] Output: Adjustment data

[2854] Step 6:

[2855] server

[2856] The server retrieves price information from the approval database using SQL queries.

[2857] Input: Query to search for price information for "Product B"

[2858] Output: Price information (e.g., ¥100,000)

[2859] Step 7:

[2860] server

[2861] The server automatically generates a quotation in PDF format using the ReportLab library.

[2862] Input: Price information, adjustment data

[2863] Output: Quotation PDF (Example: quotate_2023.pdf)

[2864] Step 8:

[2865] server

[2866] The generated quotation is saved to the server, and a download link is generated.

[2867] Input: Quotation PDF

[2868] Output: Download link (Example: https: / / example.com / downloads / quote_2023.pdf)

[2869] Step 9:

[2870] server

[2871] The server sends the generated download link to the user's device.

[2872] Input: Download link

[2873] Output: Notification to user terminal

[2874] Step 10:

[2875] User

[2876] The user clicks the received download link to download the quote.

[2877] Input: Download link

[2878] Output: Download quotation

[2879] Follow-up schedule management

[2880] Step 1:

[2881] User

[2882] If a follow-up is needed when a user submits an inquiry, for example, you might ask them to "please send a sample."

[2883] Input: "Please send a sample."

[2884] Output: Follow-up request

[2885] Step 2:

[2886] terminal

[2887] The device saves requests that require follow-up and sends them to the server.

[2888] Input: Follow-up request

[2889] Output: Request to send to server

[2890] Step 3:

[2891] server

[2892] The server analyzes the received request and determines whether follow-up is necessary.

[2893] Input: Follow-up request

[2894] Output: Analysis results (e.g., "Sample submission required")

[2895] Step 4:

[2896] server

[2897] The server uses IBM Watson's emotion engine to recognize the user's emotions.

[2898] Input: Analysis results

[2899] Output: Sentimental data (e.g., urgency)

[2900] Step 5:

[2901] server

[2902] The server adjusts the priority and method of follow-up based on the recognized sentiment data.

[2903] Input: Sentiment data

[2904] Output: Adjustment data

[2905] Step 6:

[2906] server

[2907] The server uses the Google Calendar API to set the date and time that requires follow-up and registers it in the schedule management database.

[2908] Input: Adjustment data

[2909] Output: Follow-up schedule (Example: 3 days from now, 3 PM)

[2910] Step 7:

[2911] server

[2912] As the scheduled date and time approach, Firebase Cloud Messaging is used to generate a push notification as a reminder, which is then sent to the user and the responsible party.

[2913] Input: Follow-up schedule

[2914] Output: Push notification

[2915] Step 8:

[2916] terminal

[2917] Users and staff will receive push notifications and take appropriate action.

[2918] Input: Push notification

[2919] Output: Follow-up available

[2920] (Application Example 2)

[2921] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2922] In modern society, companies are required to improve the efficiency and quality of their customer service operations. In particular, electronic payment services demand prompt and appropriate customer service, while simultaneously making it difficult to consider customer emotions. Conventional systems could only provide mechanical responses to inquiries, making it difficult to increase customer satisfaction. Furthermore, tasks such as generating quotations and managing follow-ups required manual handling, resulting in a significant workload. This invention aims to solve these problems by providing a system that recognizes customer emotions and responds accordingly.

[2923] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automatically receiving and analyzing customer inquiries using internal company data, means for searching relevant databases based on the analyzed inquiries and user sentiment data, means for automatically generating and sending a response that reflects the sentiment based on the search results and sentiment data to the customer, and means for automatically generating an electronic receipt and providing the customer with a downloadable link. This makes it possible to provide a quick and appropriate response while taking customer sentiment into consideration.

[2924] "Internal company data" refers to all information generated and stored within a company, including customer data, product data, inventory data, and sales history.

[2925] "Customer inquiries" refer to requests for information, questions, and feedback that customers send to a company, and include those sent via telephone, email, web forms, etc.

[2926] "Analysis" refers to the process of deciphering and understanding received data, and performing operations to grasp its intent and meaning.

[2927] A "related database" refers to a collection of data that stores specific information and is structured in a way that allows for later searching and referencing.

[2928] "Emotional data" refers to data that identifies a user's emotional state (e.g., joy, anger, surprise, etc.) using natural language processing and emotion recognition technologies, and expresses it as numerical values ​​or tags.

[2929] "Automatically generating responses" refers to a system generating reply text without human intervention, based on pre-configured algorithms and rules.

[2930] An "electronic receipt" refers to a receipt issued in digital format, usually generated in PDF format, and provided via email or a download link.

[2931] A "quotation" is a document that shows the price and conditions of the products or services to be offered in advance, and serves as a basis for customers to consider purchasing.

[2932] A "downloadable link" refers to a URL (web address) from which a specific file or information can be obtained via the internet.

[2933] "Follow-up" refers to additional support and verification work carried out continuously after the initial response, and is a process aimed at maintaining and improving customer satisfaction.

[2934] "Push notifications as reminders" refers to displaying short messages on a user's smartphone or other device to alert them based on a specified date, time, or conditions.

[2935] This invention relates to a system for automating and streamlining customer service in electronic payment services. In particular, it aims to improve customer satisfaction by recognizing customer emotions and providing appropriate responses and follow-ups.

[2936] Hardware configuration

[2937] This invention is carried out using the following hardware:

[2938] Smartphone (iOS or Android)

[2939] Cloud servers (e.g., AWS or Google Cloud)

[2940] Software Configuration

[2941] This invention is carried out using the following software:

[2942] Smartphone applications (iOS: Swift, Android: Kotlin)

[2943] Server-side: Node.js, Python

[2944] Database: MySQL or PostgreSQL

[2945] Emotion recognition engine: IBM Watson Tone Analyzer or Google Cloud Natural Language API

[2946] Push notification service: Firebase Cloud Messaging (FCM)

[2947] Overview of Data Processing and Data Calculation

[2948] 1. User submits inquiry:

[2949] The user enters an inquiry about electronic payments via a smartphone app and clicks the submit button. The inquiry data is sent from the device to the server.

[2950] 2. Reception and analysis on the server:

[2951] The server receives the query data and extracts keywords using natural language processing techniques. Furthermore, it analyzes the user's sentiment data using an emotion recognition engine (IBM Watson Tone Analyzer or Google Cloud Natural Language API).

[2952] 3. Generating and sending response messages:

[2953] The server searches the electronic payment database for corresponding information based on the analysis results and sentiment data. It then automatically generates a response message that reflects the sentiment and sends it to the user's smartphone.

[2954] 4. Automatic generation of electronic receipts:

[2955] When a user completes a purchase, the server automatically generates an electronic receipt in PDF format. It then generates a download link for the electronic receipt and provides it to the user.

[2956] 5. Follow-up schedule management:

[2957] The server prioritizes follow-ups based on user sentiment data. The cloud server manages the dates and times when follow-ups are needed, and push notifications are sent as reminders to the user's and the assigned staff member's devices as the scheduled time approaches.

[2958] Specific example

[2959] For example, if a user sends an inquiry asking, "Please tell me my recent transaction history," the server uses its emotion recognition engine to recognize the user's emotion as "high stress." Next, the server retrieves the corresponding transaction history from the database and generates and sends a response message in polite language to alleviate the stress, such as, "Here is your transaction history. Please feel free to contact us anytime if you have any further questions."

[2960] Example of a prompt

[2961] "Please tell me your most recent credit card transaction history."

[2962] "Please resend the most recent receipt via email."

[2963] "I'd like to confirm the payment details."

[2964] This invention makes it possible to recognize user emotions and respond accordingly. This system not only increases customer satisfaction but also streamlines a company's customer service operations.

[2965] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[2966] Step 1:

[2967] The user enters their inquiry and clicks the submit button. The "inquiry details" are retrieved from the user's device as input data. This inquiry is then sent to the server via the smartphone.

[2968] Step 2:

[2969] The system analyzes query data received by the server. The input is the query data, and keywords are extracted using a natural language p...

Claims

1. A method for automatically receiving and analyzing customer inquiries using internal company data, A means of searching related databases based on the analyzed query, A means of automatically generating and sending a response to the customer based on the search results, A system that includes this.

2. A means of searching the price databases of various companies and automatically generating quotations in a format accessible to customers, A means of generating a downloadable link for the generated quotation and providing it to the customer, The system according to claim 1, including the following:

3. A means to track the processing status of customer inquiries and set dates and times when follow-up is necessary, A means of sending push notifications to customers and representatives as reminders based on a set date and time, The system according to claim 1, including the following:

Images