System

Abstract

A system is provided.SOLUTION: The system includes a means for receiving an inquiry from a customer, a means for analyzing the contents of the received inquiry, a means for generating an automatic reply based on the analysis result, a means for transmitting the generated automatic reply to the customer, a means for recording the contents of the inquiry and the automatic reply, and a means for reporting the recorded information to a person in charge.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Companies, especially small and medium-sized businesses, often experience delays in responding to customer inquiries. This delay can lead to a decline in customer satisfaction (CS), which can have a negative impact on the company's reputation and sales. Furthermore, when inquiries are handled manually, it can result in a significant drain on human resources and make efficient business operations difficult. Furthermore, it can be difficult to determine the importance and urgency of the inquiry, making it difficult to follow up appropriately. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means. First, a means for receiving inquiries from customers is provided, and the received inquiry content is analyzed using technology such as natural language processing. Next, a means for generating an appropriate automatic reply based on the analysis results is provided, and a means for sending the generated automatic reply to the customer is also provided. Furthermore, a means for recording the inquiry content and the automatic reply content is provided, and a means for reporting the recorded information to a person in charge is also included. The content of the automatic reply can be selected from pre-prepared templates and can be customized as needed. Furthermore, by providing a means for a person in charge to receive reports via a chat application, rapid follow-up and advanced customer service can be achieved.

[0006] "Means for receiving inquiries from customers" refers to the system for receiving inquiries sent by customers via telephone, email, web forms, etc.

[0007] "Means for analyzing the content of received inquiries" refers to a mechanism for analyzing the content of received inquiries using technologies such as natural language processing (NLP) and identifying keywords and intent.

[0008] The "means for generating an automatic reply based on the analysis results" is a mechanism for creating an appropriate automatic reply message in accordance with the analyzed inquiry content.

[0009] The "means for generating automatic replies" is a mechanism for creating automatic replies to received inquiries according to predefined templates and rules.

[0010] The "means for sending the generated automatic reply to the customer" is a mechanism for sending the generated automatic reply to the customer's contact information (email address, phone number, etc.).

[0011] "Means for recording inquiry content and automatic reply content" refers to a mechanism for saving the content of received inquiries and the content of automatic replies in a database or log.

[0012] "Means for reporting recorded information to the person in charge" is a mechanism for notifying the person in charge of saved inquiries and automatic reply information and sharing the response status. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0014] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0017] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0018] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0019] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0026] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0033] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0034] The present invention is a system for responding to customer inquiries, and an embodiment thereof will be described below. This system includes a means for receiving customer inquiries, a means for analyzing the content of the received inquiry, a means for generating an automatic reply based on the analysis results, a means for sending the generated automatic reply to the customer, a means for recording the content of the inquiry and the automatic reply, and a means for reporting the recorded information to a person in charge.

[0035] System Overview

[0036] This system improves customer satisfaction and streamlines responses by having the server receive customer inquiries and then appropriately analyze and respond to the content. Furthermore, the system also has the function of recording the content of the automatic response and reporting it to the person in charge, allowing the person in charge to follow up promptly and appropriately.

[0037] Program processing explanation

[0038] 1. Receiving Inquiries

[0039] The server receives customer inquiries from the mail server or web form.

[0040] Example: The server receives an email from a customer ("I want to know the delivery status of my item").

[0041] 2. Analysis of inquiry content

[0042] The server uses natural language processing techniques to analyze the received query.

[0043] Example: Detect the keyword "delivery status" in the inquiry text and determine that it is a question about delivery.

[0044] 3. Generate Auto-Reply

[0045] The server generates an appropriate automatic reply based on the analysis results.

[0046] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[0047] 4. Sending Auto-Reply

[0048] The server sends the generated auto-reply to the customer.

[0049] Example: Send the generated text to the customer's email address (customer@example.com).

[0050] 5. Recording of inquiries and automated responses

[0051] The server records the inquiry and the automatic reply in a database.

[0052] Example: Save the inquiry and auto-reply message along with the customer ID.

[0053] 6. Report to the person in charge

[0054] The server sends a notification to the agent's chat application to report the recorded information to the agent.

[0055] For example, a message will be sent to the agent's chat window saying, "Customer A has inquired about the delivery status. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details.'"

[0056] Specific examples

[0057] 1. Specific examples of receiving inquiries

[0058] A customer sends an email asking, "What's the delivery status of my item?"

[0059] The server receives this email and begins analysis according to the analysis plan.

[0060] 2. Example of generating and sending an auto-reply

[0061] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[0062] The server immediately sends this automatic reply email to the customer.

[0063] 3. Examples of Recording and Reporting

[0064] The server records the inquiry and the automatic reply in a database along with information such as "Customer A, delivery status, reply content, time."

[0065] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[0066] Specific examples of follow-up processes

[0067] Once the report is received, the responsible person will review the situation and obtain further information if necessary (e.g., contact the delivery company).

[0068] The person in charge will then give the customer a specific reply, saying, "We apologize for the wait. Your order is scheduled to be delivered on the next business day."

[0069] In this way, by implementing this system, it becomes possible to respond quickly and appropriately to customer inquiries, thereby improving customer satisfaction (CS) and operational efficiency.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The server receives inquiries from customers.

[0073] When a customer submits an inquiry via email or web form, the server receives the content.

[0074] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[0075] Step 2:

[0076] The server analyzes the received inquiry.

[0077] Natural language processing (NLP) technology is used to analyze keywords and intent in inquiries.

[0078] Example: Extract important keywords such as "delivery status" and "order number".

[0079] Step 3:

[0080] The server generates an automatic reply based on the analysis results.

[0081] Choose the appropriate response template for your inquiry and customize it as needed.

[0082] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[0083] Step 4:

[0084] The server sends the generated auto-reply to the customer.

[0085] The generated reply is sent to the customer's email address.

[0086] Example: An auto-reply email is sent to customer@example.com.

[0087] Step 5:

[0088] The server records the inquiry and the automatic reply.

[0089] Information such as inquiry details, auto-reply messages, and timestamps is stored in a database.

[0090] Example: Customer ID, inquiry, reply, and timestamp are recorded in the database.

[0091] Step 6:

[0092] The server reports the recorded information to the person in charge.

[0093] Send a report message to the agent's chat application about the inquiry and an automated reply.

[0094] For example, "Customer A has inquired about the status of their delivery. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[0095] Step 7:

[0096] The person in charge will review the report and take additional action if necessary.

[0097] We will review reports via the chat application and take further action if necessary.

[0098] Example: Contact the delivery company to confirm the specific delivery date and time, and then contact the customer again to say, "Sorry for the wait. Your order is expected to arrive the next business day."

[0099] Example 1

[0100] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0101] Conventional inquiry response systems required manual responses to customer inquiries, which was time-consuming and labor-intensive. Furthermore, the automated responses were formulaic and inflexible, which posed a risk of lowering customer satisfaction. This made it difficult to balance the efficiency and quality of customer responses.

[0102] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0103] In this invention, the server includes means for receiving inquiries from customers, means for analyzing the content of the received inquiries using natural language processing technology, means for generating an automatic reply using a generative AI model based on the analysis results, means for sending the generated automatic reply to the customer, means for recording the content of the inquiry and the content of the automatic reply in a database, and means for reporting the recorded information to a person in charge using a chat application. This enables quick and appropriate responses to customer inquiries, thereby improving the efficiency and quality of customer service.

[0104] "Means of receiving inquiries from customers" refers to the function by which the server receives the content of inquiries from customers via a mail server, web form, etc.

[0105] "Means of analyzing the content of received inquiries using natural language processing technology" refers to the function of extracting and analyzing the intent and keywords of the inquiry using text analysis tools and libraries (e.g., NLTK, spaCy) from the content of received inquiries.

[0106] "Means for generating an automatic reply using a generative AI model based on the analysis results" refers to a function that generates an optimal automatic reply using a generative AI model (e.g., GPT-3 (registered trademark)) based on the analyzed inquiry content.

[0107] "Means for sending generated auto-replies to customers" refers to the function of sending generated auto-replies to customers' email addresses using the SMTP protocol.

[0108] "Means for recording the inquiry content and the automatic reply content in a database" refers to the function of saving the received inquiry content and the generated automatic reply content in a database (e.g., MySQL (registered trademark), MongoDB) along with a unique ID and timestamp.

[0109] "Means of reporting recorded information to the person in charge using a chat application" refers to the function of notifying and reporting the recorded information to the chat application used by the person in charge (e.g., Slack, MICROSOFT (registered trademark) TEAMS (registered trademark)).

[0110] "Means for inputting a prompt sentence into a generative AI model" refers to the operation of inputting a customized prompt sentence into a generative AI model and generating an automatic reply sentence.

[0111] "Means for a person in charge to receive reports through a chat application" refers to a function for a person in charge to receive reports regarding automatic replies through a chat application used by the person in charge.

[0112] The present invention is a system for responding promptly and appropriately to customer inquiries, and a specific embodiment thereof is described below. This system includes functions for receiving customer inquiries, analyzing them, generating automatic replies using a generative AI model, recording the replies, and reporting them to a responsible person.

[0113] System Configuration

[0114] Hardware Configuration

[0115] The main hardware configuration of this system consists of a server, a mail server, end-user terminals, and terminals for personnel. The server is a machine equipped with a high-performance processor and large memory, and links the mail server with a database to process various inquiries.

[0116] Software Configuration

[0117] The software configuration of this system is as follows:

[0118] 1. Mail server software: Mail server software such as Postfix or Sendmail is used to receive inquiry emails from customers.

[0119] 2. Web Forms (Backend): The front end is built using HTML and JavaScript (registered trademark), and the back end is implemented using Python languages ​​such as Flask and Django.

[0120] 3. Natural Language Processing Libraries: Use natural language processing libraries such as NLTK or spaCy.

[0121] 4. Generative AI models: Generative AI models, such as GPT-3, are used to generate automated replies.

[0122] 5. Database: Use MySQL or MongoDB to record inquiries and auto-reply content.

[0123] 6. Chat applications: Use APIs such as Slack or Microsoft Teams to send reports to the appropriate person.

[0124] Specific examples

[0125] 1. Specific examples of receiving inquiries

[0126] A customer sends an email asking, "What is the status of my delivery?" The server receives this email through the Postfix mail server and starts processing it as a new inquiry.

[0127] 2. Analysis of inquiry content

[0128] The server analyzes the received email using a natural language processing library (e.g., NLTK). This analysis involves tokenizing the sentence, tagging parts of speech, and extracting keywords. For example, the keyword "delivery status" is extracted, and it is determined that this is a delivery inquiry.

[0129] 3. Generate Auto-Reply

[0130] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. An example of a prompt sentence is given as input: "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." As a result, the reply sentence generated is "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with details shortly."

[0131] 4. Sending Auto-Reply

[0132] The server sends the generated auto-reply to the customer's email address using the SMTP protocol.

[0133] 5. Recording of inquiries and automated responses

[0134] The server records the inquiry and automatic reply in a MySQL database along with a unique ID and timestamp.

[0135] 6. Report to the person in charge

[0136] The server sends the recorded information to the chat application (e.g., Slack) used by the person in charge. The message "A new inquiry has been received" is displayed along with the inquiry details and an automatic reply.

[0137] As described above, the system of the present invention is designed to support prompt and appropriate responses to customer inquiries, and by using this system, customer satisfaction and response efficiency can be significantly improved.

[0138] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0139] Step 1:

[0140] The server receives customer inquiries from a mail server or web form. The input is the customer's inquiry, which is received via a mail server (Postfix or Sendmail). The server detects new emails and saves them in storage. Specifically, the server monitors ports 25 and 587 and saves new emails in storage.

[0141] Step 2:

[0142] The server analyzes the received inquiry using natural language processing technology. Specifically, it takes the received text data (inquiry content) as input and uses natural language processing libraries (NLTK or spaCy) to tokenize, tag parts of speech, and extract keywords. This process outputs the intent of the inquiry and key keywords. Specifically, the server splits the received text, extracts keywords such as "delivery status," and saves them as the analysis results.

[0143] Step 3:

[0144] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. The input is the analysis results and a prompt sentence, which is fed into the generative AI model to output the automatic reply. A specific example of a prompt sentence is, "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." In concrete terms, the server inputs the prompt sentence into the generative AI model and generates an appropriate reply sentence (e.g., "Thank you for contacting us. Your order is currently being prepared for shipping.").

[0145] Step 4:

[0146] The server sends the generated auto-reply to the customer using the SMTP protocol. The input is the generated auto-reply message and the customer's email address, and this is sent via the SMTP server. Specifically, the server inserts the generated reply message into the body of the email and sends it to the customer's email address.

[0147] Step 5:

[0148] The server records the inquiry and auto-reply content in a database. The input is the inquiry, auto-reply content, and related metadata (e.g., unique ID and timestamp), which are then saved in a database (MySQL or MongoDB). Specifically, the server executes an INSERT query to save the inquiry and auto-reply content in the database.

[0149] Step 6:

[0150] The server reports the recorded information to the person in charge using a chat application. The input is the recorded inquiry content and the automatic reply content, and this is notified to the person in charge's chat application (Slack or Microsoft Teams). Specifically, the server uses the chat API (Webhook URL) to send a message saying "A new inquiry has been received."

[0151] (Application example 1)

[0152] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0153] Traditional customer support systems require a lot of time and effort to respond to inquiries, which can lead to poor customer satisfaction. Furthermore, the lack of real-time feedback makes it difficult for agents to follow up appropriately. This can result in poor service quality and lead to customer attrition.

[0154] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0155] In this invention, the server includes a means for receiving inquiries from customers, a means for analyzing the content of the received inquiries using natural language processing technology, and a means for generating an automatic reply using a generative AI model based on the analysis results, thereby enabling quick and accurate responses to customer inquiries and enabling personnel to receive reports in real time via a chat notification system.

[0156] "Means for receiving inquiries" refers to methods for receiving inquiries from customers, such as email, a web form, or a smartphone application.

[0157] The "means for analyzing the content of the inquiry" is a method for understanding and analyzing the content of the received inquiry using natural language processing technology.

[0158] "Generative AI model" means an artificial intelligence model used to generate appropriate automated replies based on the analysis results.

[0159] "Means for generating automatic replies" refers to methods for automatically creating replies based on the analyzed inquiry content using pre-prepared templates or generative AI models.

[0160] The "means for sending automatic replies" refers to the method for sending the generated automatic replies to customers, which may be email, a smartphone app, or other means of communication.

[0161] The "means for recording the contents of inquiries and automatic replies" is a method for storing the contents of received inquiries and the automatic replies thereto in a database.

[0162] The "means for reporting the recorded information to the person in charge" is a method for notifying the person in charge of the recorded information, and is carried out via a chat notification system.

[0163] "Means of interacting with customers through smartphone applications" refers to a method by which customers can make inquiries and receive replies via their smartphones.

[0164] The present invention can be applied to a customer support system for a mail-order site. A specific embodiment of this system will be described below.

[0165] Hardware and software used

[0166] Server: Receives and analyzes inquiries, and generates and sends automatic replies.

[0167] Node.js, Express: Server Application Frameworks

[0168] MySQL: Database Management System

[0169] NLTK, SpaCy: Natural Language Processing Libraries

[0170] Firebase: Notification System

[0171] Smartphone: Customers submit inquiries and receive automated responses.

[0172] Customer applications: contact form, inbox

[0173] System processing flow

[0174] 1. Receiving Inquiries

[0175] A user submits an inquiry using the inquiry form of the smartphone application.

[0176] The server receives this query and analyzes the content using natural language processing techniques.

[0177] Example: A customer sends an inquiry asking, "What is the delivery status of my item?"

[0178] 2. Analysis of inquiry content

[0179] The server parses the query using an NLP library (NLTK or SpaCy).

[0180] Identify the content of the inquiry based on keywords and context.

[0181] Example: Extract the keyword "delivery status" to determine that this is an inquiry about delivery.

[0182] 3. Generate Auto-Reply

[0183] Based on the analysis results, the server generates an automatic reply using a generative AI model.

[0184] Customize your reply based on templates as needed.

[0185] For example, generate an automated reply that says, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[0186] 4. Sending Auto-Reply

[0187] The server sends an automated reply generated using Firebase to the customer's smartphone application.

[0188] Example: A generated auto-reply is displayed in the customer's app.

[0189] 5. Recording of inquiries and automated responses

[0190] The server records the inquiry and the automatic reply in a MySQL database.

[0191] This includes data such as customer ID, inquiry content, auto-reply content, and time of sending.

[0192] Example: Customer A's inquiries and automatic replies are stored in a database.

[0193] 6. Report to the person in charge

[0194] The server notifies the responsible party through Firebase and reports the recorded information.

[0195] Based on this, the person in charge will carry out the necessary follow-up.

[0196] Example: "Customer A contacted us regarding the status of their delivery. We sent an automated reply with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[0197] Examples of concrete examples and prompts

[0198] Specific examples

[0199] 1. Specific examples of receiving inquiries

[0200] A customer sends an email asking, "What's the delivery status of my item?"

[0201] The server receives this email and begins analyzing it.

[0202] 2. Example of generating and sending an auto-reply

[0203] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[0204] The server immediately sends this automatic reply email to the customer.

[0205] 3. Examples of Recording and Reporting

[0206] The server records the inquiry and the automatic reply in a database along with information such as customer ID, delivery status, reply content, and time.

[0207] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[0208] Prompt Sentence Examples

[0209] A customer asks, "What's the status of my delivery?" You want to generate an automated reply with the following information:

[0210] Delivery status

[0211] Thank you for contacting me

[0212] Expression to contact you later

[0213] This allows the system to respond to customer inquiries quickly and accurately, and enables staff to grasp information in real time.

[0214] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0215] Step 1:

[0216] Receiving inquiries

[0217] A user submits a question using the inquiry form of the smartphone application, and the server receives the inquiry. The input is the inquiry information submitted by the user, and the output is the raw inquiry data stored in the server.

[0218] Step 2:

[0219] Analysis of inquiry content

[0220] The server analyzes the received query using natural language processing techniques. Specifically, it uses libraries such as NLTK and SpaCy to tokenize the text and extract keywords and context. The input is the raw query data, and the output is the analyzed query data (e.g., including the keyword "delivery status").

[0221] Step 3:

[0222] Generate auto-replies

[0223] The server generates an auto-reply message using a generative AI model based on the analysis results, including customizing pre-defined templates. The input is the analyzed inquiry data and template information, and the output is the generated auto-reply message.

[0224] Step 4:

[0225] Sending automatic replies

[0226] The server sends the generated auto-reply to the customer's smartphone app. Notifications are sent using Firebase. The input is the generated auto-reply message, and the output is the reply message displayed on the user's device.

[0227] Step 5:

[0228] Record of inquiries and automatic replies

[0229] The server logs queries and generated auto-replies in a MySQL database. The input is the raw query data and the generated auto-reply messages, and the output is the logged data stored in the database.

[0230] Step 6:

[0231] Report to the person in charge

[0232] The server sends notifications to the person in charge via Firebase and reports the recorded information. The input is the recorded data and the content of the notification message, and the output is the report message displayed on the person's terminal.

[0233] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0234] The present invention is a system for responding to customer inquiries efficiently and quickly, thereby improving customer satisfaction. By combining an emotion engine, the present invention can recognize customer emotions and adjust the content of automatic replies accordingly.

[0235] System Overview

[0236] The system includes the following means:

[0237] 1. A means of receiving customer inquiries.

[0238] 2. A means of analyzing the inquiries received.

[0239] 3. A means of generating automated responses based on the analysis results.

[0240] 4. A means of sending generated auto-replies to customers.

[0241] 5. A means of recording inquiries and automated responses.

[0242] 6. A means of reporting recorded information to personnel.

[0243] 7. An emotion engine that recognizes customer emotions from the inquiries received.

[0244] 8. A means to tailor the content of auto-replies based on perceived customer sentiment.

[0245] 9. A means of reporting customer emotions recognized by the emotion engine to the agent.

[0246] Program processing explanation

[0247] 1. Receiving Inquiries

[0248] The server receives customer inquiries from the mail server or web form.

[0249] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[0250] 2. Analysis of inquiry content

[0251] The server uses natural language processing techniques to analyze the received query.

[0252] Example: Extract important keywords such as "delivery status" and "order number".

[0253] 3. Customer Emotion Recognition

[0254] The emotion engine analyzes customer emotions from the inquiry, for example recognizing whether the customer is angry, distressed, or happy.

[0255] The results of the emotion recognition are recorded as information such as "confused" or "angry."

[0256] 4. Generate Auto-Reply

[0257] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[0258] For example: Generate a basic reply like "Thank you for contacting us. Your order is currently being prepared for shipping." But if the customer is upset, add an apology like "We're sorry for the inconvenience. We're working to resolve the issue soon."

[0259] 5. Sending Auto-Reply

[0260] The server sends the generated auto-reply to the customer.

[0261] Example: Send the generated text to the customer's email address (customer@example.com).

[0262] 6. Recording of inquiries and automated responses

[0263] The server records the inquiry and the automatic reply in a database.

[0264] Example: Save the inquiry and auto-reply message along with the customer ID.

[0265] 7. Report to the person in charge

[0266] The server reports the recorded information and the customer's emotions recognized by the emotion engine to the agent.

[0267] For example: "Customer A inquired about the status of their delivery. An automated reply was sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'"

[0268] The agent receives the report through a chat application and takes further appropriate action based on it.

[0269] Specific examples

[0270] 1. Specific examples of receiving inquiries

[0271] A customer sends an email asking, "Please let me know the delivery status of my item."

[0272] The server receives this email and begins analyzing it.

[0273] 2. Specific examples of emotion recognition

[0274] The emotion engine analyzes the content of the email and recognizes the emotion "confusion."

[0275] The result of the emotion recognition is recorded as "The customer is confused."

[0276] 3. Example of generating and sending an auto-reply

[0277] Based on the keyword "delivery status," the server generates an automatic reply message saying, "Thank you for contacting us. Your order is currently being prepared for shipping."

[0278] Because the customer's emotion is recognized as "confused," an additional sentence is also generated: "We apologize for the inconvenience."

[0279] The generated reply is sent to the customer's email address.

[0280] 4. Examples of Recording and Reporting

[0281] The server records the inquiry content, the content of the automatic reply, and the results of emotion recognition in a database.

[0282] The server reports to the person in charge, "Customer A made an inquiry and an automatic reply was sent with the following content. The customer's emotion was also recognized as 'confused.'"

[0283] 5. Specific examples of follow-up processes

[0284] The person in charge receives the report and checks the situation.

[0285] The person in charge will contact the delivery company to confirm the specific delivery date and time.

[0286] The person in charge will contact the customer again and say, "Sorry for the wait. Your order is scheduled to be delivered on the next business day."

[0287] According to the present invention, customer satisfaction can be further improved by recognizing the emotions of customers and taking appropriate measures accordingly.

[0288] The processing flow will be explained below.

[0289] Step 1:

[0290] The server receives customer inquiries from the mail server or web form.

[0291] When a customer sends an inquiry email, the server receives the content.

[0292] Example: A customer sends an email asking, "Please let me know the status of my delivery."

[0293] Step 2:

[0294] The server analyzes the received inquiry.

[0295] Natural language processing (NLP) technology is used to analyze the keywords and intent of inquiries.

[0296] Example: Identifying the "Delivery Status" and "Order Number" to identify that this is a delivery inquiry.

[0297] Step 3:

[0298] The emotion engine recognizes customer emotions from the content of the inquiry.

[0299] Extract emotions from customer text and determine whether they are angry, confused, happy, etc.

[0300] Example: Recognizing the emotion of "confusion" from a customer's writing.

[0301] Step 4:

[0302] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[0303] In addition to basic reply templates, add sentiment-based phrases.

[0304] Example: "Thank you for contacting us. Your order is currently being prepared for shipping." Add an apology such as "We apologize for any inconvenience caused."

[0305] Step 5:

[0306] The server sends the generated auto-reply to the customer.

[0307] The generated reply is sent to the customer's email address.

[0308] Example: An auto-reply email is sent to customer@example.com.

[0309] Step 6:

[0310] The server records the inquiry content, the automatic reply content, and the emotion recognition results.

[0311] The inquiry content, automatic reply message, timestamp, and emotion recognition results are stored in a database.

[0312] For example, customer ID, inquiry content, reply content, emotion recognition results, and timestamp are recorded.

[0313] Step 7:

[0314] The server reports the recorded information to the person in charge.

[0315] A report message containing the inquiry, automatic reply content, and emotion recognition results is sent to the agent's chat application.

[0316] For example, "Customer A has inquired about the delivery status. An automated reply was sent with the following content: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'" This message is sent to the agent.

[0317] Step 8:

[0318] The person in charge will review the report and take additional action if necessary.

[0319] We will review the report via the chat application and conduct any further investigation or action required.

[0320] Example: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then contacts the customer again to say, "Sorry for the wait. Your order is expected to be delivered on the next business day."

[0321] By implementing this series of steps, it is possible to respond to customer inquiries quickly and appropriately, thereby improving customer satisfaction.

[0322] Example 2

[0323] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0324] Modern customer support systems are required to respond to customer inquiries quickly and accurately. However, with a wide range of customer inquiries, it is not easy to generate appropriate responses while understanding the customer's feelings, which can lead to delayed responses or inappropriate replies. In such cases, there is a risk of customer satisfaction decreasing, so an effective system to resolve this issue is needed.

[0325] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving an inquiry from a customer, a means for analyzing the received inquiry using natural language processing technology, and a means for generating an automatic reply based on the analysis results and a means for analyzing the customer's emotions. This enables accurate analysis of the customer's inquiry and generation of an appropriate automatic reply based on the analysis results and the recognized emotions. Furthermore, by combining a means for recording the inquiry content and the automatic reply content in a database and reporting the recorded information and the customer's emotion recognition results to the staff, reporting to the staff and follow-up can be smoothly performed. Furthermore, by using a generative AI model when generating the automatic reply and including a means for creating prompt sentences, more advanced and flexible reply content can be provided. This system can improve customer satisfaction.

[0326] "Means of receiving customer inquiries" refers to the function of receiving customer inquiries via email or web forms.

[0327] "Means for analyzing the content of received inquiries using natural language processing technology" refers to the function of analyzing the content of received inquiries using natural language processing technology and extracting important keywords and intent.

[0328] "Means for analyzing analysis results and customer emotions" refers to the function of recognizing customer emotions using the analyzed inquiry content and an emotion analysis engine.

[0329] "Means for generating an automatic reply" refers to a function that generates an automatic reply based on the analysis results and recognized sentiment.

[0330] A "generative AI model" is an artificial intelligence model used for text generation, such as GPT (Generative Pre-trained Transformer).

[0331] A "prompt" is a text sentence that provides specific context or instructions for input to a generative AI model.

[0332] "Means for sending automated responses to customers" refers to the ability to send generated automated responses to customers' email addresses or other means of communication.

[0333] "Means for recording the inquiry content and the automatic reply content in a database" refers to a function for saving the inquiry content and the generated automatic reply in a database.

[0334] "Means for reporting the recorded information and the emotion recognition results of the customer to the person in charge" refers to a function for reporting the information recorded in the database and the emotion recognition results to the person in charge.

[0335] "Communication Applications" refers to chat applications and messaging services used by personnel to receive reports.

[0336] This invention is a system for efficiently and quickly responding to customer inquiries and improving customer satisfaction. By combining natural language processing technology and emotion recognition, this invention can analyze the content of customer inquiries, recognize the customer's emotions, and adjust the content of automatic replies accordingly. It also uses a generative AI model to generate high-quality automatic replies.

[0337] The system uses the following hardware and software:

[0338] Hardware: Servers, mail servers, database servers

[0339] Software: Natural language processing APIs (Google® Cloud Natural Language API, Amazon Comprehend), emotion recognition engines (IBM Watson® Tone Analyzer, Microsoft Azure® Text Analytics), generative AI models (OpenAI® GPT-3, GPT-4®), email sending services (SendGrid, Amazon SES), communication applications (Slack, Microsoft Teams)

[0340] 1. Receiving Inquiries

[0341] The server receives customer inquiries from the mail server or web form. For example, the mail server detects new emails and saves them in a specified directory on the system.

[0342] 2. Analysis of inquiry content

[0343] The server reads the query information from the specified directory and analyzes it using natural language processing techniques, such as Google Cloud Natural Language API and Amazon Comprehend, to extract important keywords and intent.

[0344] 3. Customer Emotion Recognition

[0345] The server sends the inquiry to an emotion engine (such as IBM Watson Tone Analyzer or Microsoft Azure Text Analytics), which analyzes the inquiry and recognizes the customer's emotions. The results are recorded as tags such as "confused" or "angry."

[0346] 4. Generate Auto-Reply

[0347] The server generates an automated reply based on the analysis and emotion recognition results. It uses a generative AI model (such as OpenAI's GPT-3) to create a prompt and input it into the model. Examples of prompts include:

[0348] The customer inquired, "Please let me know the status of my delivery," and it was recognized that the customer was confused.

[0349] This is then input into a generative AI model to generate high-quality automated replies.

[0350] 5. Sending Auto-Reply

[0351] The server generates an automatic reply and sends it to the customer. It uses an email sending service such as SendGrid or Amazon SES to send the automatic reply to the customer's email address.

[0352] 6. Recording of inquiries and automated responses

[0353] The server records the inquiry and the automatic reply in a database. Database software such as MySQL or MongoDB is used to store the inquiry, reply, and emotion tag.

[0354] 7. Report to the person in charge

[0355] The server reports the recorded information and emotion recognition results to the person in charge, sending the information to the person in charge's chat room using a communication application such as Slack or Microsoft Teams, and the person in charge can then take necessary follow-up action based on the information.

[0356] As described above, the present invention makes it possible to respond to customer inquiries quickly and accurately, and is expected to improve customer satisfaction.

[0357] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0358] Step 1: Receiving an inquiry

[0359] The server receives inquiries from customers, specifically from a mail server or a web form, and temporarily stores the inquiry information in a specified directory.

[0360] Input: Customer inquiry emails and web form data

[0361] Output: Query data saved in temporary storage directory

[0362] Specific operation: When a new email is detected on the mail server (for example, general mail server software), the system saves the email in the 'inbox' directory.

[0363] Step 2: Analyzing the inquiry

[0364] The server reads the query information from the specified directory, analyzes the data using natural language processing technology, and extracts important keywords and intent.

[0365] Input: Query data read from temporary storage directory

[0366] Output: Extracted keywords and intents (e.g., "Delivery status," "Order number," etc.)

[0367] What it does: It uses a Python script to call the Google Cloud Natural Language API to extract key keywords and intent from the query.

[0368] Step 3: Recognize customer emotions

[0369] The server sends the inquiry to the emotion engine, which analyzes the inquiry and recognizes the customer's emotion tag.

[0370] Input: Inquiry details

[0371] Output: Recognized emotion tag (e.g., "confused," "angry," etc.)

[0372] Specific behavior: Uses IBM Watson Tone Analyzer to analyze inquiries and identify "confusion" and "anger."

[0373] Step 4: Generate an auto-reply

[0374] The server generates an automatic reply based on the analysis results and emotion recognition results, and uses a generative AI model to create prompts and generate reply content.

[0375] Input: Extracted keywords, intent, and sentiment tags

[0376] Output: Generated autoresponder

[0377] Specific operation: The server generates a prompt sentence such as: "A customer inquired, 'Please tell me the delivery status of my item,' and it was recognized that the customer was confused." This is input into the generative AI model to obtain a high-quality automatic reply sentence.

[0378] Step 5: Sending Auto-Reply

[0379] The server uses an email sending service to send the generated auto-reply to the customer.

[0380] Input: Generated autoresponder and customer email address

[0381] Output: Email sent to customer

[0382] Specific behavior: Calls SendGrid's API and sends the generated reply to the customer's email address (e.g., customer@example.com).

[0383] Step 6: Record your inquiries and automated responses

[0384] The server records the inquiry content, the automatic reply content, and the emotion recognition results in a database.

[0385] Input: Inquiry content, auto-reply content, emotion tag

[0386] Output: Records stored in the database

[0387] Specific behavior: Connects to a MySQL database and saves the inquiry content, response content, and emotion tag as records.

[0388] Step 7: Report to the person in charge

[0389] The server reports the recorded information and emotion recognition results to the person in charge, who receives the report through a communication application.

[0390] Input: Information stored in the database and emotion recognition results

[0391] Output: Report message sent to the agent

[0392] Specific behavior: Using the Slack API, a message is sent to the agent's chat room stating, "Customer A has made an inquiry and an automated reply has been sent with the following content. The customer's sentiment was recognized as 'confused'."

[0393] Step 8: Handle follow-up

[0394] The responsible person receives the report and manually follows up as needed.

[0395] Input: Report message received by the agent

[0396] Output: Result of additional customer actions

[0397] Specific actions: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then uses that information to contact the customer again, saying, "Sorry for the wait. Your order is expected to be delivered the next business day."

[0398] (Application example 2)

[0399] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0400] Conventional food delivery services have issues with delayed responses to customer inquiries and low customer satisfaction due to uniform replies that do not consider customer feelings. In addition, it takes time and effort for staff to manually analyze the content of inquiries, understand the feelings of customers, and provide appropriate replies, which is inefficient.

[0401] The specific processing 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 receiving inquiries from customers, means for analyzing the content of the received inquiries, and means for generating an automatic reply based on the analysis results. This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's emotions, thereby improving customer satisfaction. In addition, by using an emotion engine to recognize the customer's emotions and adjusting the content of the automatic reply according to the recognized emotions, more appropriate and personalized responses can be provided.

[0402] "Customer" means any person who purchases or uses a product or service.

[0403] An "inquiry" is when a customer raises a question or concern about a product or service.

[0404] "Means for receiving" refers to the method or device by which the system collects customer inquiries.

[0405] "Means for analysis" refers to the method or device by which the system understands and analyzes the content of the query.

[0406] "Means for generating an automatic reply" refers to a method or device by which the system automatically generates a reply based on the analysis results.

[0407] "Means for sending" refers to a method or device for delivering the generated reply to the customer.

[0408] "Recording means" refers to a method or device for saving the contents of inquiries and automatic replies.

[0409] "Reporting means" means the method or device by which recorded information is communicated to responsible personnel.

[0410] The "emotion engine" is a system that recognizes customer emotions from the content of inquiries.

[0411] "Adjusting means" is a method or device for varying the content of the automated response based on perceived customer sentiment.

[0412] The present invention relates to a customer support system for food delivery services. This system uses an emotion engine to recognize customer emotions and generate automatic replies accordingly, in order to respond quickly and accurately to customer inquiries.

[0413] System Configuration

[0414] The system includes the following major components:

[0415] 1. Server

[0416] Receiving inquiries, analyzing them, and generating and sending automatic replies

[0417] 2. Database

[0418] Record of inquiries and automatic replies

[0419] 3. Emotion Engine

[0420] Customer Emotion Recognition

[0421] 4. Natural Language Processing (NLP) Libraries

[0422] Analysis of inquiry content

[0423] 5. Smartphone App

[0424] An interface for customers to make inquiries

[0425] System Operation Details

[0426] Hardware and Software Usage

[0427] The server (e.g., AWS EC2, Google Cloud Compute Engine) handles all processes from receiving inquiries to sending automated replies. The server receives inquiries via a mail server or web form and analyzes them using a natural language processing (NLP) library (e.g., spaCy, NLTK).

[0428] An emotion engine (e.g., Google Cloud Natural Language API, AWS Comprehend) recognizes customer emotions from the content of inquiries and identifies emotions such as "confused" or "angry."

[0429] The database (e.g., MySQL, PostgreSQL) records the received inquiries, the generated auto-replies, and the emotion recognition results.

[0430] The smartphone app is developed using frameworks such as React Native and provides a user interface for customers to submit inquiries.

[0431] Details of data processing and calculation

[0432] 1. Receiving Inquiries

[0433] The server receives customer inquiries via email or web form, and the inquiries are stored in a database in text format.

[0434] 2. Analysis of inquiry content

[0435] The server uses an NLP library to analyze the inquiry and extract important keywords such as delivery delays and order cancellations.

[0436] 3. Emotion recognition

[0437] The server uses an emotion engine to recognize the customer's emotions from the analyzed inquiry, for example, identifying emotions such as "angry" or "confused."

[0438] 4. Generate Auto-Reply

[0439] The server generates an automatic reply based on the analysis results and emotion recognition results. For example, if a delivery is delayed and the customer is angry, the server generates a reply that includes an apology, such as "We apologize for the wait. Your delivery is currently on its way. We will deliver it to you soon."

[0440] 5. Sending Auto-Reply

[0441] The server notifies the customer's app of the generated auto-reply.

[0442] 6. Records

[0443] The server records the inquiry and the automatic reply in a database and shares it with the customer support team.

[0444] 7. Report to the person in charge

[0445] The server reports the automatically generated reply and customer sentiment to the agent, and if necessary, further follow-up is carried out.

[0446] Examples of concrete examples and prompts

[0447] Specific examples

[0448] Examples of receiving inquiries:

[0449] A customer sends an inquiry saying, "My pizza order hasn't arrived yet, what's going on?"

[0450] Examples of emotion recognition:

[0451] The emotion engine analyzes the content of the email and recognizes the emotion "anger."

[0452] Examples of generating and sending auto-replies:

[0453] Based on the keyword "delayed delivery" and the emotion "angry," the server generates an automatic reply saying, "We apologize for the wait. Your order is currently being delivered. We will deliver it to you soon." and sends it to the customer.

[0454] Prompt Sentence Examples

[0455] Prompts to recognize customer sentiment:

[0456] "Customer asks: 'My pizza order hasn't arrived yet, what's going on?' Recognize customer sentiment and generate appropriate automated responses."

[0457] This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[0458] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0459] Step 1:

[0460] Receiving inquiries

[0461] Input: A customer submits an inquiry through a smartphone app.

[0462] Specific behavior: A customer uses the smartphone app's inquiry form or chat function to enter and send an inquiry in text format, such as, "The item I ordered hasn't arrived yet. What's going on?"

[0463] Data processing: The smartphone app generates an API request to send the inquiry to the server.

[0464] Output: The server receives the API request and saves the query in text format.

[0465] Step 2:

[0466] Analysis of inquiry content

[0467] Input: Text data of the query received by the server.

[0468] What it does: The server parses the query using a natural language processing (NLP) library (e.g., spaCy or NLTK) and extracts important keywords (e.g., "delivery" or "delay").

[0469] Data processing: Tokenize the text data, tag each token with a part of speech, and extract important keywords.

[0470] Output: A list of keywords extracted as a result of the analysis.

[0471] Step 3:

[0472] emotion recognition

[0473] Input: Text data of the inquiry and extracted keyword list.

[0474] Specific operation: Using an emotion engine (e.g., Google Cloud Natural Language API or AWS Comprehend), the server recognizes the customer's emotion from the inquiry content.

[0475] Data processing: The emotion engine analyzes the text data and calculates an emotion score. For example, it makes a judgment such as "The customer is angry" based on the emotion score.

[0476] Output: Analysis results including recognized emotion information (e.g., "angry").

[0477] Step 4:

[0478] Generate auto-replies

[0479] Input: Analysis results (keyword list and sentiment information).

[0480] What it does: The server generates an appropriate auto-reply based on the analysis results, choosing a basic reply from a template and adjusting the content based on the sentiment.

[0481] Data processing: Generate an automatic reply by combining template sentences and analysis results. For example, based on "Delivery delayed" and "Angry," generate a reply such as "Sorry for the wait. Delivery is currently in progress. We will deliver it to you soon."

[0482] Output: The generated auto-reply.

[0483] Step 5:

[0484] Sending automatic replies

[0485] Input: The generated auto-response.

[0486] Specific operation: Generates an API request to notify the customer's smartphone app of the automatic reply message generated by the server.

[0487] Data processing: Auto-reply messages are embedded into API requests and sent to the customer's app.

[0488] Output: The automated reply that will be displayed on the customer's smartphone app.

[0489] Step 6:

[0490] Record of inquiries and automatic replies

[0491] Input: Enquiry and generated auto-reply.

[0492] Specific operation: The server records the inquiry and the automatic reply in a database.

[0493] Data processing: Convert the text data into database format and save it along with the inquiry ID.

[0494] Output: A record of the inquiry and auto-reply stored in the database.

[0495] Step 7:

[0496] Report to the person in charge

[0497] Input: Inquiry details, auto-reply details and emotional information recorded in the database.

[0498] Specific operation: The server generates a notification to report to the agent and sends it to the agent's chat application.

[0499] Data processing: Compile inquiry details, automatic response details, and emotional information and convert them into a report format.

[0500] Output: Report notification sent to the responsible party.

[0501] This series of steps makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[0502] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0503] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0504] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0505] [Second embodiment]

[0506] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0507] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0508] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0509] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0510] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0511] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0512] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0513] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0514] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0515] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0516] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0517] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0518] The present invention is a system for responding to customer inquiries, and an embodiment thereof will be described below. This system includes a means for receiving customer inquiries, a means for analyzing the content of the received inquiry, a means for generating an automatic reply based on the analysis results, a means for sending the generated automatic reply to the customer, a means for recording the content of the inquiry and the automatic reply, and a means for reporting the recorded information to a person in charge.

[0519] System Overview

[0520] This system improves customer satisfaction and streamlines responses by having the server receive customer inquiries and then appropriately analyze and respond to the content. Furthermore, the system also has the function of recording the content of the automatic response and reporting it to the person in charge, allowing the person in charge to follow up promptly and appropriately.

[0521] Program processing explanation

[0522] 1. Receiving Inquiries

[0523] The server receives customer inquiries from the mail server or web form.

[0524] Example: The server receives an email from a customer ("I want to know the delivery status of my item").

[0525] 2. Analysis of inquiry content

[0526] The server uses natural language processing techniques to analyze the received query.

[0527] Example: Detect the keyword "delivery status" in the inquiry text and determine that it is a question about delivery.

[0528] 3. Generate Auto-Reply

[0529] The server generates an appropriate automatic reply based on the analysis results.

[0530] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[0531] 4. Sending Auto-Reply

[0532] The server sends the generated auto-reply to the customer.

[0533] Example: Send the generated text to the customer's email address (customer@example.com).

[0534] 5. Recording of inquiries and automated responses

[0535] The server records the inquiry and the automatic reply in a database.

[0536] Example: Save the inquiry and auto-reply message along with the customer ID.

[0537] 6. Report to the person in charge

[0538] The server sends a notification to the agent's chat application to report the recorded information to the agent.

[0539] For example, a message will be sent to the agent's chat window saying, "Customer A has inquired about the delivery status. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details.'"

[0540] Specific examples

[0541] 1. Specific examples of receiving inquiries

[0542] A customer sends an email asking, "What's the delivery status of my item?"

[0543] The server receives this email and begins analysis according to the analysis plan.

[0544] 2. Example of generating and sending an auto-reply

[0545] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[0546] The server immediately sends this automatic reply email to the customer.

[0547] 3. Examples of Recording and Reporting

[0548] The server records the inquiry and the automatic reply in a database along with information such as "Customer A, delivery status, reply content, time."

[0549] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[0550] Specific examples of follow-up processes

[0551] Once the report is received, the responsible person will review the situation and obtain further information if necessary (e.g., contact the delivery company).

[0552] The person in charge will then give the customer a specific reply, saying, "We apologize for the wait. Your order is scheduled to be delivered on the next business day."

[0553] In this way, by implementing this system, it becomes possible to respond quickly and appropriately to customer inquiries, thereby improving customer satisfaction (CS) and operational efficiency.

[0554] The processing flow will be explained below.

[0555] Step 1:

[0556] The server receives inquiries from customers.

[0557] When a customer submits an inquiry via email or web form, the server receives the content.

[0558] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[0559] Step 2:

[0560] The server analyzes the received inquiry.

[0561] Natural language processing (NLP) technology is used to analyze keywords and intent in inquiries.

[0562] Example: Extract important keywords such as "delivery status" and "order number".

[0563] Step 3:

[0564] The server generates an automatic reply based on the analysis results.

[0565] Choose the appropriate response template for your inquiry and customize it as needed.

[0566] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[0567] Step 4:

[0568] The server sends the generated auto-reply to the customer.

[0569] The generated reply is sent to the customer's email address.

[0570] Example: An auto-reply email is sent to customer@example.com.

[0571] Step 5:

[0572] The server records the inquiry and the automatic reply.

[0573] Information such as inquiry details, auto-reply messages, and timestamps is stored in a database.

[0574] Example: Customer ID, inquiry, reply, and timestamp are recorded in the database.

[0575] Step 6:

[0576] The server reports the recorded information to the person in charge.

[0577] Send a report message to the agent's chat application about the inquiry and an automated reply.

[0578] For example, "Customer A has inquired about the status of their delivery. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[0579] Step 7:

[0580] The person in charge will review the report and take additional action if necessary.

[0581] We will review reports via the chat application and take further action if necessary.

[0582] Example: Contact the delivery company to confirm the specific delivery date and time, and then contact the customer again to say, "Sorry for the wait. Your order is expected to arrive the next business day."

[0583] Example 1

[0584] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0585] Conventional inquiry response systems required manual responses to customer inquiries, which was time-consuming and labor-intensive. Furthermore, the automated responses were formulaic and inflexible, which posed a risk of lowering customer satisfaction. This made it difficult to balance the efficiency and quality of customer responses.

[0586] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0587] In this invention, the server includes means for receiving inquiries from customers, means for analyzing the content of the received inquiries using natural language processing technology, means for generating an automatic reply using a generative AI model based on the analysis results, means for sending the generated automatic reply to the customer, means for recording the content of the inquiry and the content of the automatic reply in a database, and means for reporting the recorded information to a person in charge using a chat application. This enables quick and appropriate responses to customer inquiries, thereby improving the efficiency and quality of customer service.

[0588] "Means of receiving inquiries from customers" refers to the function by which the server receives the content of inquiries from customers via a mail server, web form, etc.

[0589] "Means of analyzing the content of received inquiries using natural language processing technology" refers to the function of extracting and analyzing the intent and keywords of the inquiry using text analysis tools and libraries (e.g., NLTK, spaCy) from the content of received inquiries.

[0590] "Means for generating an automatic reply using a generative AI model based on the analysis results" refers to a function that uses a generative AI model (e.g., GPT-3) to generate an optimal automatic reply based on the analyzed inquiry content.

[0591] "Means for sending generated auto-replies to customers" refers to the function of sending generated auto-replies to customers' email addresses using the SMTP protocol.

[0592] "Means for recording the inquiry content and the automatic reply content in a database" refers to the function of saving the received inquiry content and the generated automatic reply content in a database (e.g., MySQL, MongoDB) along with a unique ID and timestamp.

[0593] "Means of reporting recorded information to the person in charge using a chat application" refers to the function of notifying and reporting recorded information to the chat application used by the person in charge (e.g., Slack, Microsoft Teams).

[0594] "Means for inputting a prompt sentence into a generative AI model" refers to the operation of inputting a customized prompt sentence into a generative AI model and generating an automatic reply sentence.

[0595] "Means for a person in charge to receive reports through a chat application" refers to a function for a person in charge to receive reports regarding automatic replies through a chat application used by the person in charge.

[0596] The present invention is a system for responding promptly and appropriately to customer inquiries, and a specific embodiment thereof is described below. This system includes functions for receiving customer inquiries, analyzing them, generating automatic replies using a generative AI model, recording the replies, and reporting them to a responsible person.

[0597] System Configuration

[0598] Hardware Configuration

[0599] The main hardware configuration of this system consists of a server, a mail server, end-user terminals, and terminals for personnel. The server is a machine equipped with a high-performance processor and large memory, and links the mail server with a database to process various inquiries.

[0600] Software Configuration

[0601] The software configuration of this system is as follows:

[0602] 1. Mail server software: Mail server software such as Postfix or Sendmail is used to receive inquiry emails from customers.

[0603] 2. Web Forms (Backend): The front end is built with HTML and JavaScript, and the back end is implemented using Python languages ​​such as Flask and Django.

[0604] 3. Natural Language Processing Libraries: Use natural language processing libraries such as NLTK or spaCy.

[0605] 4. Generative AI models: Generative AI models, such as GPT-3, are used to generate automated replies.

[0606] 5. Database: Use MySQL or MongoDB to record inquiries and auto-reply content.

[0607] 6. Chat applications: Use APIs such as Slack or Microsoft Teams to send reports to the appropriate person.

[0608] Specific examples

[0609] 1. Specific examples of receiving inquiries

[0610] A customer sends an email asking, "What is the status of my delivery?" The server receives this email through the Postfix mail server and starts processing it as a new inquiry.

[0611] 2. Analysis of inquiry content

[0612] The server analyzes the received email using a natural language processing library (e.g., NLTK). This analysis involves tokenizing the sentence, tagging parts of speech, and extracting keywords. For example, the keyword "delivery status" is extracted, and it is determined that this is a delivery inquiry.

[0613] 3. Generate Auto-Reply

[0614] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. An example of a prompt sentence is given as input: "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." As a result, the reply sentence generated is "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with details shortly."

[0615] 4. Sending Auto-Reply

[0616] The server sends the generated auto-reply to the customer's email address using the SMTP protocol.

[0617] 5. Recording of inquiries and automated responses

[0618] The server records the inquiry and automatic reply in a MySQL database along with a unique ID and timestamp.

[0619] 6. Report to the person in charge

[0620] The server sends the recorded information to the chat application (e.g., Slack) used by the person in charge. The message "A new inquiry has been received" is displayed along with the inquiry details and an automatic reply.

[0621] As described above, the system of the present invention is designed to support prompt and appropriate responses to customer inquiries, and by using this system, customer satisfaction and response efficiency can be significantly improved.

[0622] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0623] Step 1:

[0624] The server receives customer inquiries from a mail server or web form. The input is the customer's inquiry, which is received via a mail server (Postfix or Sendmail). The server detects new emails and saves them in storage. Specifically, the server monitors ports 25 and 587 and saves new emails in storage.

[0625] Step 2:

[0626] The server analyzes the received inquiry using natural language processing technology. Specifically, it takes the received text data (inquiry content) as input and uses natural language processing libraries (NLTK or spaCy) to tokenize, tag parts of speech, and extract keywords. This process outputs the intent of the inquiry and key keywords. Specifically, the server splits the received text, extracts keywords such as "delivery status," and saves them as the analysis results.

[0627] Step 3:

[0628] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. The input is the analysis results and a prompt sentence, which is fed into the generative AI model to output the automatic reply. A specific example of a prompt sentence is, "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." In concrete terms, the server inputs the prompt sentence into the generative AI model and generates an appropriate reply sentence (e.g., "Thank you for contacting us. Your order is currently being prepared for shipping.").

[0629] Step 4:

[0630] The server sends the generated auto-reply to the customer using the SMTP protocol. The input is the generated auto-reply message and the customer's email address, and this is sent via the SMTP server. Specifically, the server inserts the generated reply message into the body of the email and sends it to the customer's email address.

[0631] Step 5:

[0632] The server records the inquiry and auto-reply content in a database. The input is the inquiry, auto-reply content, and related metadata (e.g., unique ID and timestamp), which are then saved in a database (MySQL or MongoDB). Specifically, the server executes an INSERT query to save the inquiry and auto-reply content in the database.

[0633] Step 6:

[0634] The server reports the recorded information to the person in charge using a chat application. The input is the recorded inquiry content and the automatic reply content, and this is notified to the person in charge's chat application (Slack or Microsoft Teams). Specifically, the server uses the chat API (Webhook URL) to send a message saying "A new inquiry has been received."

[0635] (Application example 1)

[0636] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0637] Traditional customer support systems require a lot of time and effort to respond to inquiries, which can lead to poor customer satisfaction. Furthermore, the lack of real-time feedback makes it difficult for agents to follow up appropriately. This can result in poor service quality and lead to customer attrition.

[0638] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0639] In this invention, the server includes a means for receiving inquiries from customers, a means for analyzing the content of the received inquiries using natural language processing technology, and a means for generating an automatic reply using a generative AI model based on the analysis results, thereby enabling quick and accurate responses to customer inquiries and enabling personnel to receive reports in real time via a chat notification system.

[0640] "Means for receiving inquiries" refers to methods for receiving inquiries from customers, such as email, a web form, or a smartphone application.

[0641] The "means for analyzing the content of the inquiry" is a method for understanding and analyzing the content of the received inquiry using natural language processing technology.

[0642] "Generative AI model" means an artificial intelligence model used to generate appropriate automated replies based on the analysis results.

[0643] "Means for generating automatic replies" refers to methods for automatically creating replies based on the analyzed inquiry content using pre-prepared templates or generative AI models.

[0644] The "means for sending automatic replies" refers to the method for sending the generated automatic replies to customers, which may be email, a smartphone app, or other means of communication.

[0645] The "means for recording the contents of inquiries and automatic replies" is a method for storing the contents of received inquiries and the automatic replies thereto in a database.

[0646] The "means for reporting the recorded information to the person in charge" is a method for notifying the person in charge of the recorded information, and is carried out via a chat notification system.

[0647] "Means of interacting with customers through smartphone applications" refers to a method by which customers can make inquiries and receive replies via their smartphones.

[0648] The present invention can be applied to a customer support system for a mail-order site. A specific embodiment of this system will be described below.

[0649] Hardware and software used

[0650] Server: Receives and analyzes inquiries, and generates and sends automatic replies.

[0651] Node.js, Express: Server Application Frameworks

[0652] MySQL: Database Management System

[0653] NLTK, SpaCy: Natural Language Processing Libraries

[0654] Firebase: Notification System

[0655] Smartphone: Customers submit inquiries and receive automated responses.

[0656] Customer applications: contact form, inbox

[0657] System processing flow

[0658] 1. Receiving Inquiries

[0659] A user submits an inquiry using the inquiry form of the smartphone application.

[0660] The server receives this query and analyzes the content using natural language processing techniques.

[0661] Example: A customer sends an inquiry asking, "What is the delivery status of my item?"

[0662] 2. Analysis of inquiry content

[0663] The server parses the query using an NLP library (NLTK or SpaCy).

[0664] Identify the content of the inquiry based on keywords and context.

[0665] Example: Extract the keyword "delivery status" to determine that this is an inquiry about delivery.

[0666] 3. Generate Auto-Reply

[0667] Based on the analysis results, the server generates an automatic reply using a generative AI model.

[0668] Customize your reply based on templates as needed.

[0669] For example, generate an automated reply that says, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[0670] 4. Sending Auto-Reply

[0671] The server sends an automated reply generated using Firebase to the customer's smartphone application.

[0672] Example: A generated auto-reply is displayed in the customer's app.

[0673] 5. Recording of inquiries and automated responses

[0674] The server records the inquiry and the automatic reply in a MySQL database.

[0675] This includes data such as customer ID, inquiry content, auto-reply content, and time of sending.

[0676] Example: Customer A's inquiries and automatic replies are stored in a database.

[0677] 6. Report to the person in charge

[0678] The server notifies the responsible party through Firebase and reports the recorded information.

[0679] Based on this, the person in charge will carry out the necessary follow-up.

[0680] Example: "Customer A contacted us regarding the status of their delivery. We sent an automated reply with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[0681] Examples of concrete examples and prompts

[0682] Specific examples

[0683] 1. Specific examples of receiving inquiries

[0684] A customer sends an email asking, "What's the delivery status of my item?"

[0685] The server receives this email and begins analyzing it.

[0686] 2. Example of generating and sending an auto-reply

[0687] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[0688] The server immediately sends this automatic reply email to the customer.

[0689] 3. Examples of Recording and Reporting

[0690] The server records the inquiry and the automatic reply in a database along with information such as customer ID, delivery status, reply content, and time.

[0691] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[0692] Prompt Sentence Examples

[0693] A customer asks, "What's the status of my delivery?" You want to generate an automated reply with the following information:

[0694] Delivery status

[0695] Thank you for contacting me

[0696] Expression to contact you later

[0697] This allows the system to respond to customer inquiries quickly and accurately, and enables staff to grasp information in real time.

[0698] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0699] Step 1:

[0700] Receiving inquiries

[0701] A user submits a question using the inquiry form of the smartphone application, and the server receives the inquiry. The input is the inquiry information submitted by the user, and the output is the raw inquiry data stored in the server.

[0702] Step 2:

[0703] Analysis of inquiry content

[0704] The server analyzes the received query using natural language processing techniques. Specifically, it uses libraries such as NLTK and SpaCy to tokenize the text and extract keywords and context. The input is the raw query data, and the output is the analyzed query data (e.g., including the keyword "delivery status").

[0705] Step 3:

[0706] Generate auto-replies

[0707] The server generates an auto-reply message using a generative AI model based on the analysis results, including customizing pre-defined templates. The input is the analyzed inquiry data and template information, and the output is the generated auto-reply message.

[0708] Step 4:

[0709] Sending automatic replies

[0710] The server sends the generated auto-reply to the customer's smartphone app. Notifications are sent using Firebase. The input is the generated auto-reply message, and the output is the reply message displayed on the user's device.

[0711] Step 5:

[0712] Record of inquiries and automatic replies

[0713] The server logs queries and generated auto-replies in a MySQL database. The input is the raw query data and the generated auto-reply messages, and the output is the logged data stored in the database.

[0714] Step 6:

[0715] Report to the person in charge

[0716] The server sends notifications to the person in charge via Firebase and reports the recorded information. The input is the recorded data and the content of the notification message, and the output is the report message displayed on the person's terminal.

[0717] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0718] The present invention is a system for responding to customer inquiries efficiently and quickly, thereby improving customer satisfaction. By combining an emotion engine, the present invention can recognize customer emotions and adjust the content of automatic replies accordingly.

[0719] System Overview

[0720] The system includes the following means:

[0721] 1. A means of receiving customer inquiries.

[0722] 2. A means of analyzing the inquiries received.

[0723] 3. A means of generating automated responses based on the analysis results.

[0724] 4. A means of sending generated auto-replies to customers.

[0725] 5. A means of recording inquiries and automated responses.

[0726] 6. A means of reporting recorded information to personnel.

[0727] 7. An emotion engine that recognizes customer emotions from the inquiries received.

[0728] 8. A means to tailor the content of auto-replies based on perceived customer sentiment.

[0729] 9. A means of reporting customer emotions recognized by the emotion engine to the agent.

[0730] Program processing explanation

[0731] 1. Receiving Inquiries

[0732] The server receives customer inquiries from the mail server or web form.

[0733] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[0734] 2. Analysis of inquiry content

[0735] The server uses natural language processing techniques to analyze the received query.

[0736] Example: Extract important keywords such as "delivery status" and "order number".

[0737] 3. Customer Emotion Recognition

[0738] The emotion engine analyzes customer emotions from the inquiry, for example recognizing whether the customer is angry, distressed, or happy.

[0739] The results of the emotion recognition are recorded as information such as "confused" or "angry."

[0740] 4. Generate Auto-Reply

[0741] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[0742] For example: Generate a basic reply like "Thank you for contacting us. Your order is currently being prepared for shipping." But if the customer is upset, add an apology like "We're sorry for the inconvenience. We're working to resolve the issue soon."

[0743] 5. Sending Auto-Reply

[0744] The server sends the generated auto-reply to the customer.

[0745] Example: Send the generated text to the customer's email address (customer@example.com).

[0746] 6. Recording of inquiries and automated responses

[0747] The server records the inquiry and the automatic reply in a database.

[0748] Example: Save the inquiry and auto-reply message along with the customer ID.

[0749] 7. Report to the person in charge

[0750] The server reports the recorded information and the customer's emotions recognized by the emotion engine to the agent.

[0751] For example: "Customer A inquired about the status of their delivery. An automated reply was sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'"

[0752] The agent receives the report through a chat application and takes further appropriate action based on it.

[0753] Specific examples

[0754] 1. Specific examples of receiving inquiries

[0755] A customer sends an email asking, "Please let me know the delivery status of my item."

[0756] The server receives this email and begins analyzing it.

[0757] 2. Specific examples of emotion recognition

[0758] The emotion engine analyzes the content of the email and recognizes the emotion "confusion."

[0759] The result of the emotion recognition is recorded as "The customer is confused."

[0760] 3. Example of generating and sending an auto-reply

[0761] Based on the keyword "delivery status," the server generates an automatic reply message saying, "Thank you for contacting us. Your order is currently being prepared for shipping."

[0762] Because the customer's emotion is recognized as "confused," an additional sentence is also generated: "We apologize for the inconvenience."

[0763] The generated reply is sent to the customer's email address.

[0764] 4. Examples of Recording and Reporting

[0765] The server records the inquiry content, the content of the automatic reply, and the results of emotion recognition in a database.

[0766] The server reports to the person in charge, "Customer A made an inquiry and an automatic reply was sent with the following content. The customer's emotion was also recognized as 'confused.'"

[0767] 5. Specific examples of follow-up processes

[0768] The person in charge receives the report and checks the situation.

[0769] The person in charge will contact the delivery company to confirm the specific delivery date and time.

[0770] The person in charge will contact the customer again and say, "Sorry for the wait. Your order is scheduled to be delivered on the next business day."

[0771] According to the present invention, customer satisfaction can be further improved by recognizing the emotions of customers and taking appropriate measures accordingly.

[0772] The processing flow will be explained below.

[0773] Step 1:

[0774] The server receives customer inquiries from the mail server or web form.

[0775] When a customer sends an inquiry email, the server receives the content.

[0776] Example: A customer sends an email asking, "Please let me know the status of my delivery."

[0777] Step 2:

[0778] The server analyzes the received inquiry.

[0779] Natural language processing (NLP) technology is used to analyze the keywords and intent of inquiries.

[0780] Example: Identifying the "Delivery Status" and "Order Number" to identify that this is a delivery inquiry.

[0781] Step 3:

[0782] The emotion engine recognizes customer emotions from the content of the inquiry.

[0783] Extract emotions from customer text and determine whether they are angry, confused, happy, etc.

[0784] Example: Recognizing the emotion of "confusion" from a customer's writing.

[0785] Step 4:

[0786] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[0787] In addition to basic reply templates, add sentiment-based phrases.

[0788] Example: "Thank you for contacting us. Your order is currently being prepared for shipping." Add an apology such as "We apologize for any inconvenience caused."

[0789] Step 5:

[0790] The server sends the generated auto-reply to the customer.

[0791] The generated reply is sent to the customer's email address.

[0792] Example: An auto-reply email is sent to customer@example.com.

[0793] Step 6:

[0794] The server records the inquiry content, the automatic reply content, and the emotion recognition results.

[0795] The inquiry content, automatic reply message, timestamp, and emotion recognition results are stored in a database.

[0796] For example, customer ID, inquiry content, reply content, emotion recognition results, and timestamp are recorded.

[0797] Step 7:

[0798] The server reports the recorded information to the person in charge.

[0799] A report message containing the inquiry, automatic reply content, and emotion recognition results is sent to the agent's chat application.

[0800] For example, "Customer A has inquired about the delivery status. An automated reply was sent with the following content: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'" This message is sent to the agent.

[0801] Step 8:

[0802] The person in charge will review the report and take additional action if necessary.

[0803] We will review the report via the chat application and conduct any further investigation or action required.

[0804] Example: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then contacts the customer again to say, "Sorry for the wait. Your order is expected to be delivered on the next business day."

[0805] By implementing this series of steps, it is possible to respond to customer inquiries quickly and appropriately, thereby improving customer satisfaction.

[0806] Example 2

[0807] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0808] Modern customer support systems are required to respond to customer inquiries quickly and accurately. However, with a wide range of customer inquiries, it is not easy to generate appropriate responses while understanding the customer's feelings, which can lead to delayed responses or inappropriate replies. In such cases, there is a risk of customer satisfaction decreasing, so an effective system to resolve this issue is needed.

[0809] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving an inquiry from a customer, a means for analyzing the received inquiry using natural language processing technology, and a means for generating an automatic reply based on the analysis results and a means for analyzing the customer's emotions. This enables accurate analysis of the customer's inquiry and generation of an appropriate automatic reply based on the analysis results and the recognized emotions. Furthermore, by combining a means for recording the inquiry content and the automatic reply content in a database and reporting the recorded information and the customer's emotion recognition results to the staff, reporting to the staff and follow-up can be smoothly performed. Furthermore, by using a generative AI model when generating the automatic reply and including a means for creating prompt sentences, more advanced and flexible reply content can be provided. This system can improve customer satisfaction.

[0810] "Means of receiving customer inquiries" refers to the function of receiving customer inquiries via email or web forms.

[0811] "Means for analyzing the content of received inquiries using natural language processing technology" refers to the function of analyzing the content of received inquiries using natural language processing technology and extracting important keywords and intent.

[0812] "Means for analyzing analysis results and customer emotions" refers to the function of recognizing customer emotions using the analyzed inquiry content and an emotion analysis engine.

[0813] "Means for generating an automatic reply" refers to a function that generates an automatic reply based on the analysis results and recognized sentiment.

[0814] A "generative AI model" is an artificial intelligence model used for text generation, such as GPT (Generative Pre-trained Transformer).

[0815] A "prompt" is a text sentence that provides specific context or instructions for input to a generative AI model.

[0816] "Means for sending automated responses to customers" refers to the ability to send generated automated responses to customers' email addresses or other means of communication.

[0817] "Means for recording the inquiry content and the automatic reply content in a database" refers to a function for saving the inquiry content and the generated automatic reply in a database.

[0818] "Means for reporting the recorded information and the emotion recognition results of the customer to the person in charge" refers to a function for reporting the information recorded in the database and the emotion recognition results to the person in charge.

[0819] "Communication Applications" refers to chat applications and messaging services used by personnel to receive reports.

[0820] This invention is a system for efficiently and quickly responding to customer inquiries and improving customer satisfaction. By combining natural language processing technology and emotion recognition, this invention can analyze the content of customer inquiries, recognize the customer's emotions, and adjust the content of automatic replies accordingly. It also uses a generative AI model to generate high-quality automatic replies.

[0821] The system uses the following hardware and software:

[0822] Hardware: Servers, mail servers, database servers

[0823] Software: Natural language processing APIs (Google Cloud Natural Language API, Amazon Comprehend), emotion recognition engines (IBM Watson Tone Analyzer, Microsoft Azure Text Analytics), generative AI models (OpenAI GPT-3, GPT-4), email sending services (SendGrid, Amazon SES), communication applications (Slack, Microsoft Teams)

[0824] 1. Receiving Inquiries

[0825] The server receives customer inquiries from the mail server or web form. For example, the mail server detects new emails and saves them in a specified directory on the system.

[0826] 2. Analysis of inquiry content

[0827] The server reads the query information from the specified directory and analyzes it using natural language processing techniques, such as Google Cloud Natural Language API and Amazon Comprehend, to extract important keywords and intent.

[0828] 3. Customer Emotion Recognition

[0829] The server sends the inquiry to an emotion engine (such as IBM Watson Tone Analyzer or Microsoft Azure Text Analytics), which analyzes the inquiry and recognizes the customer's emotions. The results are recorded as tags such as "confused" or "angry."

[0830] 4. Generate Auto-Reply

[0831] The server generates an automated reply based on the analysis and emotion recognition results. It uses a generative AI model (such as OpenAI's GPT-3) to create a prompt and input it into the model. Examples of prompts include:

[0832] The customer inquired, "Please let me know the status of my delivery," and it was recognized that the customer was confused.

[0833] This is then input into a generative AI model to generate high-quality automated replies.

[0834] 5. Sending Auto-Reply

[0835] The server generates an automatic reply and sends it to the customer. It uses an email sending service such as SendGrid or Amazon SES to send the automatic reply to the customer's email address.

[0836] 6. Recording of inquiries and automated responses

[0837] The server records the inquiry and the automatic reply in a database. Database software such as MySQL or MongoDB is used to store the inquiry, reply, and emotion tag.

[0838] 7. Report to the person in charge

[0839] The server reports the recorded information and emotion recognition results to the person in charge, sending the information to the person in charge's chat room using a communication application such as Slack or Microsoft Teams, and the person in charge can then take necessary follow-up action based on the information.

[0840] As described above, the present invention makes it possible to respond to customer inquiries quickly and accurately, and is expected to improve customer satisfaction.

[0841] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0842] Step 1: Receiving an inquiry

[0843] The server receives inquiries from customers, specifically from a mail server or a web form, and temporarily stores the inquiry information in a specified directory.

[0844] Input: Customer inquiry emails and web form data

[0845] Output: Query data saved in temporary storage directory

[0846] Specific operation: When a new email is detected on the mail server (for example, general mail server software), the system saves the email in the 'inbox' directory.

[0847] Step 2: Analyzing the inquiry

[0848] The server reads the query information from the specified directory, analyzes the data using natural language processing technology, and extracts important keywords and intent.

[0849] Input: Query data read from temporary storage directory

[0850] Output: Extracted keywords and intents (e.g., "Delivery status," "Order number," etc.)

[0851] What it does: It uses a Python script to call the Google Cloud Natural Language API to extract key keywords and intent from the query.

[0852] Step 3: Recognize customer emotions

[0853] The server sends the inquiry to the emotion engine, which analyzes the inquiry and recognizes the customer's emotion tag.

[0854] Input: Inquiry details

[0855] Output: Recognized emotion tag (e.g., "confused," "angry," etc.)

[0856] Specific behavior: Uses IBM Watson Tone Analyzer to analyze inquiries and identify "confusion" and "anger."

[0857] Step 4: Generate an auto-reply

[0858] The server generates an automatic reply based on the analysis results and emotion recognition results, and uses a generative AI model to create prompts and generate reply content.

[0859] Input: Extracted keywords, intent, and sentiment tags

[0860] Output: Generated autoresponder

[0861] Specific operation: The server generates a prompt sentence such as: "A customer inquired, 'Please tell me the delivery status of my item,' and it was recognized that the customer was confused." This is input into the generative AI model to obtain a high-quality automatic reply sentence.

[0862] Step 5: Sending Auto-Reply

[0863] The server uses an email sending service to send the generated auto-reply to the customer.

[0864] Input: Generated autoresponder and customer email address

[0865] Output: Email sent to customer

[0866] Specific behavior: Calls SendGrid's API and sends the generated reply to the customer's email address (e.g., customer@example.com).

[0867] Step 6: Record your inquiries and automated responses

[0868] The server records the inquiry content, the automatic reply content, and the emotion recognition results in a database.

[0869] Input: Inquiry content, auto-reply content, emotion tag

[0870] Output: Records stored in the database

[0871] Specific behavior: Connects to a MySQL database and saves the inquiry content, response content, and emotion tag as records.

[0872] Step 7: Report to the person in charge

[0873] The server reports the recorded information and emotion recognition results to the person in charge, who receives the report through a communication application.

[0874] Input: Information stored in the database and emotion recognition results

[0875] Output: Report message sent to the agent

[0876] Specific behavior: Using the Slack API, a message is sent to the agent's chat room stating, "Customer A has made an inquiry and an automated reply has been sent with the following content. The customer's sentiment was recognized as 'confused'."

[0877] Step 8: Handle follow-up

[0878] The responsible person receives the report and manually follows up as needed.

[0879] Input: Report message received by the agent

[0880] Output: Result of additional customer actions

[0881] Specific actions: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then uses that information to contact the customer again, saying, "Sorry for the wait. Your order is expected to be delivered the next business day."

[0882] (Application example 2)

[0883] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0884] Conventional food delivery services have issues with delayed responses to customer inquiries and low customer satisfaction due to uniform replies that do not consider customer feelings. In addition, it takes time and effort for staff to manually analyze the content of inquiries, understand the feelings of customers, and provide appropriate replies, which is inefficient.

[0885] The specific processing 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 receiving inquiries from customers, means for analyzing the content of the received inquiries, and means for generating an automatic reply based on the analysis results. This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's emotions, thereby improving customer satisfaction. In addition, by using an emotion engine to recognize the customer's emotions and adjusting the content of the automatic reply according to the recognized emotions, more appropriate and personalized responses can be provided.

[0886] "Customer" means any person who purchases or uses a product or service.

[0887] An "inquiry" is when a customer raises a question or concern about a product or service.

[0888] "Means for receiving" refers to the method or device by which the system collects customer inquiries.

[0889] "Means for analysis" refers to the method or device by which the system understands and analyzes the content of the query.

[0890] "Means for generating an automatic reply" refers to a method or device by which the system automatically generates a reply based on the analysis results.

[0891] "Means for sending" refers to a method or device for delivering the generated reply to the customer.

[0892] "Recording means" refers to a method or device for saving the contents of inquiries and automatic replies.

[0893] "Reporting means" means the method or device by which recorded information is communicated to responsible personnel.

[0894] The "emotion engine" is a system that recognizes customer emotions from the content of inquiries.

[0895] "Adjusting means" is a method or device for varying the content of the automated response based on perceived customer sentiment.

[0896] The present invention relates to a customer support system for food delivery services. This system uses an emotion engine to recognize customer emotions and generate automatic replies accordingly, in order to respond quickly and accurately to customer inquiries.

[0897] System Configuration

[0898] The system includes the following major components:

[0899] 1. Server

[0900] Receiving inquiries, analyzing them, and generating and sending automatic replies

[0901] 2. Database

[0902] Record of inquiries and automatic replies

[0903] 3. Emotion Engine

[0904] Customer Emotion Recognition

[0905] 4. Natural Language Processing (NLP) Libraries

[0906] Analysis of inquiry content

[0907] 5. Smartphone App

[0908] An interface for customers to make inquiries

[0909] System Operation Details

[0910] Hardware and Software Usage

[0911] The server (e.g., AWS EC2, Google Cloud Compute Engine) handles all the processes from receiving inquiries to sending automated replies. The server receives inquiries via a mail server or web form and analyzes them using a natural language processing (NLP) library (e.g., spaCy, NLTK).

[0912] An emotion engine (e.g., Google Cloud Natural Language API, AWS Comprehend) recognizes customer emotions from the content of inquiries and identifies emotions such as "confused" or "angry."

[0913] The database (e.g., MySQL, PostgreSQL) records the received inquiries, the generated auto-replies, and the emotion recognition results.

[0914] The smartphone app is developed using frameworks such as React Native and provides a user interface for customers to submit inquiries.

[0915] Details of data processing and calculation

[0916] 1. Receiving Inquiries

[0917] The server receives customer inquiries via email or web form, and the inquiries are stored in a database in text format.

[0918] 2. Analysis of inquiry content

[0919] The server uses an NLP library to analyze the inquiry and extract important keywords such as delivery delays and order cancellations.

[0920] 3. Emotion recognition

[0921] The server uses an emotion engine to recognize the customer's emotions from the analyzed inquiry, for example, identifying emotions such as "angry" or "confused."

[0922] 4. Generate Auto-Reply

[0923] The server generates an automatic reply based on the analysis results and emotion recognition results. For example, if a delivery is delayed and the customer is angry, the server generates a reply that includes an apology, such as "We apologize for the wait. Your delivery is currently on its way. We will deliver it to you soon."

[0924] 5. Sending Auto-Reply

[0925] The server notifies the customer's app of the generated auto-reply.

[0926] 6. Records

[0927] The server records the inquiry and the automatic reply in a database and shares it with the customer support team.

[0928] 7. Report to the person in charge

[0929] The server reports the automatically generated reply and customer sentiment to the agent, and if necessary, further follow-up is carried out.

[0930] Examples of concrete examples and prompts

[0931] Specific examples

[0932] Examples of receiving inquiries:

[0933] A customer sends an inquiry saying, "My pizza order hasn't arrived yet, what's going on?"

[0934] Examples of emotion recognition:

[0935] The emotion engine analyzes the content of the email and recognizes the emotion "anger."

[0936] Examples of generating and sending auto-replies:

[0937] Based on the keyword "delayed delivery" and the emotion "angry," the server generates an automatic reply saying, "We apologize for the wait. Your order is currently being delivered. We will deliver it to you soon." and sends it to the customer.

[0938] Prompt Sentence Examples

[0939] Prompts to recognize customer sentiment:

[0940] "Customer asks: 'My pizza order hasn't arrived yet, what's going on?' Recognize customer sentiment and generate appropriate automated responses."

[0941] This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[0942] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0943] Step 1:

[0944] Receiving inquiries

[0945] Input: A customer submits an inquiry through a smartphone app.

[0946] Specific behavior: A customer uses the smartphone app's inquiry form or chat function to enter and send an inquiry in text format, such as, "The item I ordered hasn't arrived yet. What's going on?"

[0947] Data processing: The smartphone app generates an API request to send the inquiry to the server.

[0948] Output: The server receives the API request and saves the query in text format.

[0949] Step 2:

[0950] Analysis of inquiry content

[0951] Input: Text data of the query received by the server.

[0952] What it does: The server parses the query using a natural language processing (NLP) library (e.g., spaCy or NLTK) and extracts important keywords (e.g., "delivery" or "delay").

[0953] Data processing: Tokenize the text data, tag each token with a part of speech, and extract important keywords.

[0954] Output: A list of keywords extracted as a result of the analysis.

[0955] Step 3:

[0956] emotion recognition

[0957] Input: Text data of the inquiry and extracted keyword list.

[0958] Specific operation: Using an emotion engine (e.g., Google Cloud Natural Language API or AWS Comprehend), the server recognizes the customer's emotion from the inquiry content.

[0959] Data processing: The emotion engine analyzes the text data and calculates an emotion score. For example, it makes a judgment such as "The customer is angry" based on the emotion score.

[0960] Output: Analysis results including recognized emotion information (e.g., "angry").

[0961] Step 4:

[0962] Generate auto-replies

[0963] Input: Analysis results (keyword list and sentiment information).

[0964] What it does: The server generates an appropriate auto-reply based on the analysis results, choosing a basic reply from a template and adjusting the content based on the sentiment.

[0965] Data processing: Generate an automatic reply by combining template sentences and analysis results. For example, based on "Delivery delayed" and "Angry," generate a reply such as "Sorry for the wait. Delivery is currently in progress. We will deliver it to you soon."

[0966] Output: The generated auto-reply.

[0967] Step 5:

[0968] Sending automatic replies

[0969] Input: The generated auto-response.

[0970] Specific operation: Generates an API request to notify the customer's smartphone app of the automatic reply message generated by the server.

[0971] Data processing: Auto-reply messages are embedded into API requests and sent to the customer's app.

[0972] Output: The automated reply that will be displayed on the customer's smartphone app.

[0973] Step 6:

[0974] Record of inquiries and automatic replies

[0975] Input: Enquiry and generated auto-reply.

[0976] Specific operation: The server records the inquiry and the automatic reply in a database.

[0977] Data processing: Convert the text data into database format and save it along with the inquiry ID.

[0978] Output: A record of the inquiry and auto-reply stored in the database.

[0979] Step 7:

[0980] Report to the person in charge

[0981] Input: Inquiry details, auto-reply details and emotional information recorded in the database.

[0982] Specific operation: The server generates a notification to report to the agent and sends it to the agent's chat application.

[0983] Data processing: Compile inquiry details, automatic response details, and emotional information and convert them into a report format.

[0984] Output: Report notification sent to the responsible party.

[0985] This series of steps makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[0986] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0987] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0988] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0989] [Third embodiment]

[0990] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0991] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0992] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0993] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0994] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0995] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0996] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0997] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0998] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0999] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1000] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1001] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1002] The present invention is a system for responding to customer inquiries, and an embodiment thereof will be described below. This system includes a means for receiving customer inquiries, a means for analyzing the content of the received inquiry, a means for generating an automatic reply based on the analysis results, a means for sending the generated automatic reply to the customer, a means for recording the content of the inquiry and the automatic reply, and a means for reporting the recorded information to a person in charge.

[1003] System Overview

[1004] This system improves customer satisfaction and streamlines responses by having the server receive customer inquiries and then appropriately analyze and respond to the content. Furthermore, the system also has the function of recording the content of the automatic response and reporting it to the person in charge, allowing the person in charge to follow up promptly and appropriately.

[1005] Program processing explanation

[1006] 1. Receiving Inquiries

[1007] The server receives customer inquiries from the mail server or web form.

[1008] Example: The server receives an email from a customer ("I want to know the delivery status of my item").

[1009] 2. Analysis of inquiry content

[1010] The server uses natural language processing techniques to analyze the received query.

[1011] Example: Detect the keyword "delivery status" in the inquiry text and determine that it is a question about delivery.

[1012] 3. Generate Auto-Reply

[1013] The server generates an appropriate automatic reply based on the analysis results.

[1014] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[1015] 4. Sending Auto-Reply

[1016] The server sends the generated auto-reply to the customer.

[1017] Example: Send the generated text to the customer's email address (customer@example.com).

[1018] 5. Recording of inquiries and automated responses

[1019] The server records the inquiry and the automatic reply in a database.

[1020] Example: Save the inquiry and auto-reply message along with the customer ID.

[1021] 6. Report to the person in charge

[1022] The server sends a notification to the agent's chat application to report the recorded information to the agent.

[1023] For example, a message will be sent to the agent's chat window saying, "Customer A has inquired about the delivery status. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details.'"

[1024] Specific examples

[1025] 1. Specific examples of receiving inquiries

[1026] A customer sends an email asking, "What's the delivery status of my item?"

[1027] The server receives this email and begins analysis according to the analysis plan.

[1028] 2. Example of generating and sending an auto-reply

[1029] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[1030] The server immediately sends this automatic reply email to the customer.

[1031] 3. Examples of Recording and Reporting

[1032] The server records the inquiry and the automatic reply in a database along with information such as "Customer A, delivery status, reply content, time."

[1033] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[1034] Specific examples of follow-up processes

[1035] Once the report is received, the responsible person will review the situation and obtain further information if necessary (e.g., contact the delivery company).

[1036] The person in charge will then give the customer a specific reply, saying, "We apologize for the wait. Your order is scheduled to be delivered on the next business day."

[1037] In this way, by implementing this system, it becomes possible to respond quickly and appropriately to customer inquiries, thereby improving customer satisfaction (CS) and operational efficiency.

[1038] The processing flow will be explained below.

[1039] Step 1:

[1040] The server receives inquiries from customers.

[1041] When a customer submits an inquiry via email or web form, the server receives the content.

[1042] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[1043] Step 2:

[1044] The server analyzes the received inquiry.

[1045] Natural language processing (NLP) technology is used to analyze keywords and intent in inquiries.

[1046] Example: Extract important keywords such as "delivery status" and "order number".

[1047] Step 3:

[1048] The server generates an automatic reply based on the analysis results.

[1049] Choose the appropriate response template for your inquiry and customize it as needed.

[1050] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[1051] Step 4:

[1052] The server sends the generated auto-reply to the customer.

[1053] The generated reply is sent to the customer's email address.

[1054] Example: An auto-reply email is sent to customer@example.com.

[1055] Step 5:

[1056] The server records the inquiry and the automatic reply.

[1057] Information such as inquiry details, auto-reply messages, and timestamps is stored in a database.

[1058] Example: Customer ID, inquiry, reply, and timestamp are recorded in the database.

[1059] Step 6:

[1060] The server reports the recorded information to the person in charge.

[1061] Send a report message to the agent's chat application about the inquiry and an automated reply.

[1062] For example, "Customer A has inquired about the status of their delivery. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[1063] Step 7:

[1064] The person in charge will review the report and take additional action if necessary.

[1065] We will review reports via the chat application and take further action if necessary.

[1066] Example: Contact the delivery company to confirm the specific delivery date and time, and then contact the customer again to say, "Sorry for the wait. Your order is expected to arrive the next business day."

[1067] Example 1

[1068] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1069] Conventional inquiry response systems required manual responses to customer inquiries, which was time-consuming and labor-intensive. Furthermore, the automated responses were formulaic and inflexible, which posed a risk of lowering customer satisfaction. This made it difficult to balance the efficiency and quality of customer responses.

[1070] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1071] In this invention, the server includes means for receiving inquiries from customers, means for analyzing the content of the received inquiries using natural language processing technology, means for generating an automatic reply using a generative AI model based on the analysis results, means for sending the generated automatic reply to the customer, means for recording the content of the inquiry and the content of the automatic reply in a database, and means for reporting the recorded information to a person in charge using a chat application. This enables quick and appropriate responses to customer inquiries, thereby improving the efficiency and quality of customer service.

[1072] "Means of receiving inquiries from customers" refers to the function by which the server receives the content of inquiries from customers via a mail server, web form, etc.

[1073] "Means of analyzing the content of received inquiries using natural language processing technology" refers to the function of extracting and analyzing the intent and keywords of the inquiry using text analysis tools and libraries (e.g., NLTK, spaCy) from the content of received inquiries.

[1074] "Means for generating an automatic reply using a generative AI model based on the analysis results" refers to a function that uses a generative AI model (e.g., GPT-3) to generate an optimal automatic reply based on the analyzed inquiry content.

[1075] "Means for sending generated auto-replies to customers" refers to the function of sending generated auto-replies to customers' email addresses using the SMTP protocol.

[1076] "Means for recording the inquiry content and the automatic reply content in a database" refers to the function of saving the received inquiry content and the generated automatic reply content in a database (e.g., MySQL, MongoDB) along with a unique ID and timestamp.

[1077] "Means of reporting recorded information to the person in charge using a chat application" refers to the function of notifying and reporting recorded information to the chat application used by the person in charge (e.g., Slack, Microsoft Teams).

[1078] "Means for inputting a prompt sentence into a generative AI model" refers to the operation of inputting a customized prompt sentence into a generative AI model and generating an automatic reply sentence.

[1079] "Means for a person in charge to receive reports through a chat application" refers to a function for a person in charge to receive reports regarding automatic replies through a chat application used by the person in charge.

[1080] The present invention is a system for responding promptly and appropriately to customer inquiries, and a specific embodiment thereof is described below. This system includes functions for receiving customer inquiries, analyzing them, generating automatic replies using a generative AI model, recording the replies, and reporting them to a responsible person.

[1081] System Configuration

[1082] Hardware Configuration

[1083] The main hardware configuration of this system consists of a server, a mail server, end-user terminals, and terminals for personnel. The server is a machine equipped with a high-performance processor and large memory, and links the mail server with a database to process various inquiries.

[1084] Software Configuration

[1085] The software configuration of this system is as follows:

[1086] 1. Mail server software: Mail server software such as Postfix or Sendmail is used to receive inquiry emails from customers.

[1087] 2. Web Forms (Backend): The front end is built with HTML and JavaScript, and the back end is implemented using Python languages ​​such as Flask and Django.

[1088] 3. Natural Language Processing Libraries: Use natural language processing libraries such as NLTK or spaCy.

[1089] 4. Generative AI models: Generative AI models, such as GPT-3, are used to generate automated replies.

[1090] 5. Database: Use MySQL or MongoDB to record inquiries and auto-reply content.

[1091] 6. Chat applications: Use APIs such as Slack or Microsoft Teams to send reports to the appropriate person.

[1092] Specific examples

[1093] 1. Specific examples of receiving inquiries

[1094] A customer sends an email asking, "What is the status of my delivery?" The server receives this email through the Postfix mail server and starts processing it as a new inquiry.

[1095] 2. Analysis of inquiry content

[1096] The server analyzes the received email using a natural language processing library (e.g., NLTK). This analysis involves tokenizing the sentence, tagging parts of speech, and extracting keywords. For example, the keyword "delivery status" is extracted, and it is determined that this is a delivery inquiry.

[1097] 3. Generate Auto-Reply

[1098] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. An example of a prompt sentence is given as input: "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." As a result, the reply sentence generated is "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with details shortly."

[1099] 4. Sending Auto-Reply

[1100] The server sends the generated auto-reply to the customer's email address using the SMTP protocol.

[1101] 5. Recording of inquiries and automated responses

[1102] The server records the inquiry and automatic reply in a MySQL database along with a unique ID and timestamp.

[1103] 6. Report to the person in charge

[1104] The server sends the recorded information to the chat application (e.g., Slack) used by the person in charge. The message "A new inquiry has been received" is displayed along with the inquiry details and an automatic reply.

[1105] As described above, the system of the present invention is designed to support prompt and appropriate responses to customer inquiries, and by using this system, customer satisfaction and response efficiency can be significantly improved.

[1106] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1107] Step 1:

[1108] The server receives customer inquiries from a mail server or web form. The input is the customer's inquiry, which is received via a mail server (Postfix or Sendmail). The server detects new emails and saves them in storage. Specifically, the server monitors ports 25 and 587 and saves new emails in storage.

[1109] Step 2:

[1110] The server analyzes the received inquiry using natural language processing technology. Specifically, it takes the received text data (inquiry content) as input and uses natural language processing libraries (NLTK or spaCy) to tokenize, tag parts of speech, and extract keywords. This process outputs the intent of the inquiry and key keywords. Specifically, the server splits the received text, extracts keywords such as "delivery status," and saves them as the analysis results.

[1111] Step 3:

[1112] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. The input is the analysis results and a prompt sentence, which is fed into the generative AI model to output the automatic reply. A specific example of a prompt sentence is, "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." In concrete terms, the server inputs the prompt sentence into the generative AI model and generates an appropriate reply sentence (e.g., "Thank you for contacting us. Your order is currently being prepared for shipping.").

[1113] Step 4:

[1114] The server sends the generated auto-reply to the customer using the SMTP protocol. The input is the generated auto-reply message and the customer's email address, and this is sent via the SMTP server. Specifically, the server inserts the generated reply message into the body of the email and sends it to the customer's email address.

[1115] Step 5:

[1116] The server records the inquiry and auto-reply content in a database. The input is the inquiry, auto-reply content, and related metadata (e.g., unique ID and timestamp), which are then saved in a database (MySQL or MongoDB). Specifically, the server executes an INSERT query to save the inquiry and auto-reply content in the database.

[1117] Step 6:

[1118] The server reports the recorded information to the person in charge using a chat application. The input is the recorded inquiry content and the automatic reply content, and this is notified to the person in charge's chat application (Slack or Microsoft Teams). Specifically, the server uses the chat API (Webhook URL) to send a message saying "A new inquiry has been received."

[1119] (Application example 1)

[1120] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1121] Traditional customer support systems require a lot of time and effort to respond to inquiries, which can lead to poor customer satisfaction. Furthermore, the lack of real-time feedback makes it difficult for agents to follow up appropriately. This can result in poor service quality and lead to customer attrition.

[1122] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1123] In this invention, the server includes a means for receiving inquiries from customers, a means for analyzing the content of the received inquiries using natural language processing technology, and a means for generating an automatic reply using a generative AI model based on the analysis results, thereby enabling quick and accurate responses to customer inquiries and enabling personnel to receive reports in real time via a chat notification system.

[1124] "Means for receiving inquiries" refers to methods for receiving inquiries from customers, such as email, a web form, or a smartphone application.

[1125] The "means for analyzing the content of the inquiry" is a method for understanding and analyzing the content of the received inquiry using natural language processing technology.

[1126] "Generative AI model" means an artificial intelligence model used to generate appropriate automated replies based on the analysis results.

[1127] "Means for generating automatic replies" refers to methods for automatically creating replies based on the analyzed inquiry content using pre-prepared templates or generative AI models.

[1128] The "means for sending automatic replies" refers to the method for sending the generated automatic replies to customers, which may be email, a smartphone app, or other means of communication.

[1129] The "means for recording the contents of inquiries and automatic replies" is a method for storing the contents of received inquiries and the automatic replies thereto in a database.

[1130] The "means for reporting the recorded information to the person in charge" is a method for notifying the person in charge of the recorded information, and is carried out via a chat notification system.

[1131] "Means of interacting with customers through smartphone applications" refers to a method by which customers can make inquiries and receive replies via their smartphones.

[1132] The present invention can be applied to a customer support system for a mail-order site. A specific embodiment of this system will be described below.

[1133] Hardware and software used

[1134] Server: Receives and analyzes inquiries, and generates and sends automatic replies.

[1135] Node.js, Express: Server Application Frameworks

[1136] MySQL: Database Management System

[1137] NLTK, SpaCy: Natural Language Processing Libraries

[1138] Firebase: Notification System

[1139] Smartphone: Customers submit inquiries and receive automated responses.

[1140] Customer applications: contact form, inbox

[1141] System processing flow

[1142] 1. Receiving Inquiries

[1143] A user submits an inquiry using the inquiry form of the smartphone application.

[1144] The server receives this query and analyzes the content using natural language processing techniques.

[1145] Example: A customer sends an inquiry asking, "What is the delivery status of my item?"

[1146] 2. Analysis of inquiry content

[1147] The server parses the query using an NLP library (NLTK or SpaCy).

[1148] Identify the content of the inquiry based on keywords and context.

[1149] Example: Extract the keyword "delivery status" to determine that this is an inquiry about delivery.

[1150] 3. Generate Auto-Reply

[1151] Based on the analysis results, the server generates an automatic reply using a generative AI model.

[1152] Customize your reply based on templates as needed.

[1153] For example, generate an automated reply that says, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[1154] 4. Sending Auto-Reply

[1155] The server sends an automated reply generated using Firebase to the customer's smartphone application.

[1156] Example: A generated auto-reply is displayed in the customer's app.

[1157] 5. Recording of inquiries and automated responses

[1158] The server records the inquiry and the automatic reply in a MySQL database.

[1159] This includes data such as customer ID, inquiry content, auto-reply content, and time of sending.

[1160] Example: Customer A's inquiries and automatic replies are stored in a database.

[1161] 6. Report to the person in charge

[1162] The server notifies the responsible party through Firebase and reports the recorded information.

[1163] Based on this, the person in charge will carry out the necessary follow-up.

[1164] Example: "Customer A contacted us regarding the status of their delivery. We sent an automated reply with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[1165] Examples of concrete examples and prompts

[1166] Specific examples

[1167] 1. Specific examples of receiving inquiries

[1168] A customer sends an email asking, "What's the delivery status of my item?"

[1169] The server receives this email and begins analyzing it.

[1170] 2. Example of generating and sending an auto-reply

[1171] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[1172] The server immediately sends this automatic reply email to the customer.

[1173] 3. Examples of Recording and Reporting

[1174] The server records the inquiry and the automatic reply in a database along with information such as customer ID, delivery status, reply content, and time.

[1175] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[1176] Prompt Sentence Examples

[1177] A customer asks, "What's the status of my delivery?" You want to generate an automated reply with the following information:

[1178] Delivery status

[1179] Thank you for contacting me

[1180] Expression to contact you later

[1181] This allows the system to respond to customer inquiries quickly and accurately, and enables staff to grasp information in real time.

[1182] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1183] Step 1:

[1184] Receiving inquiries

[1185] A user submits a question using the inquiry form of the smartphone application, and the server receives the inquiry. The input is the inquiry information submitted by the user, and the output is the raw inquiry data stored in the server.

[1186] Step 2:

[1187] Analysis of inquiry content

[1188] The server analyzes the received query using natural language processing techniques. Specifically, it uses libraries such as NLTK and SpaCy to tokenize the text and extract keywords and context. The input is the raw query data, and the output is the analyzed query data (e.g., including the keyword "delivery status").

[1189] Step 3:

[1190] Generate auto-replies

[1191] The server generates an auto-reply message using a generative AI model based on the analysis results, including customizing pre-defined templates. The input is the analyzed inquiry data and template information, and the output is the generated auto-reply message.

[1192] Step 4:

[1193] Sending automatic replies

[1194] The server sends the generated auto-reply to the customer's smartphone app. Notifications are sent using Firebase. The input is the generated auto-reply message, and the output is the reply message displayed on the user's device.

[1195] Step 5:

[1196] Record of inquiries and automatic replies

[1197] The server logs queries and generated auto-replies in a MySQL database. The input is the raw query data and the generated auto-reply messages, and the output is the logged data stored in the database.

[1198] Step 6:

[1199] Report to the person in charge

[1200] The server sends notifications to the person in charge via Firebase and reports the recorded information. The input is the recorded data and the content of the notification message, and the output is the report message displayed on the person's terminal.

[1201] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1202] The present invention is a system for responding to customer inquiries efficiently and quickly, thereby improving customer satisfaction. By combining an emotion engine, the present invention can recognize customer emotions and adjust the content of automatic replies accordingly.

[1203] System Overview

[1204] The system includes the following means:

[1205] 1. A means of receiving customer inquiries.

[1206] 2. A means of analyzing the inquiries received.

[1207] 3. A means of generating automated responses based on the analysis results.

[1208] 4. A means of sending generated auto-replies to customers.

[1209] 5. A means of recording inquiries and automated responses.

[1210] 6. A means of reporting recorded information to personnel.

[1211] 7. An emotion engine that recognizes customer emotions from the inquiries received.

[1212] 8. A means to tailor the content of auto-replies based on perceived customer sentiment.

[1213] 9. A means of reporting customer emotions recognized by the emotion engine to the agent.

[1214] Program processing explanation

[1215] 1. Receiving Inquiries

[1216] The server receives customer inquiries from the mail server or web form.

[1217] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[1218] 2. Analysis of inquiry content

[1219] The server uses natural language processing techniques to analyze the received query.

[1220] Example: Extract important keywords such as "delivery status" and "order number".

[1221] 3. Customer Emotion Recognition

[1222] The emotion engine analyzes customer emotions from the inquiry, for example recognizing whether the customer is angry, distressed, or happy.

[1223] The results of the emotion recognition are recorded as information such as "confused" or "angry."

[1224] 4. Generate Auto-Reply

[1225] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[1226] For example: Generate a basic reply like "Thank you for contacting us. Your order is currently being prepared for shipping." But if the customer is upset, add an apology like "We're sorry for the inconvenience. We're working to resolve the issue soon."

[1227] 5. Sending Auto-Reply

[1228] The server sends the generated auto-reply to the customer.

[1229] Example: Send the generated text to the customer's email address (customer@example.com).

[1230] 6. Recording of inquiries and automated responses

[1231] The server records the inquiry and the automatic reply in a database.

[1232] Example: Save the inquiry and auto-reply message along with the customer ID.

[1233] 7. Report to the person in charge

[1234] The server reports the recorded information and the customer's emotions recognized by the emotion engine to the agent.

[1235] For example: "Customer A inquired about the status of their delivery. An automated reply was sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'"

[1236] The agent receives the report through a chat application and takes further appropriate action based on it.

[1237] Specific examples

[1238] 1. Specific examples of receiving inquiries

[1239] A customer sends an email asking, "Please let me know the delivery status of my item."

[1240] The server receives this email and begins analyzing it.

[1241] 2. Specific examples of emotion recognition

[1242] The emotion engine analyzes the content of the email and recognizes the emotion "confusion."

[1243] The result of the emotion recognition is recorded as "The customer is confused."

[1244] 3. Example of generating and sending an auto-reply

[1245] Based on the keyword "delivery status," the server generates an automatic reply message saying, "Thank you for contacting us. Your order is currently being prepared for shipping."

[1246] Because the customer's emotion is recognized as "confused," an additional sentence is also generated: "We apologize for the inconvenience."

[1247] The generated reply is sent to the customer's email address.

[1248] 4. Examples of Recording and Reporting

[1249] The server records the inquiry content, the content of the automatic reply, and the results of emotion recognition in a database.

[1250] The server reports to the person in charge, "Customer A made an inquiry and an automatic reply was sent with the following content. The customer's emotion was also recognized as 'confused.'"

[1251] 5. Specific examples of follow-up processes

[1252] The person in charge receives the report and checks the situation.

[1253] The person in charge will contact the delivery company to confirm the specific delivery date and time.

[1254] The person in charge will contact the customer again and say, "Sorry for the wait. Your order is scheduled to be delivered on the next business day."

[1255] According to the present invention, customer satisfaction can be further improved by recognizing the emotions of customers and taking appropriate measures accordingly.

[1256] The processing flow will be explained below.

[1257] Step 1:

[1258] The server receives customer inquiries from the mail server or web form.

[1259] When a customer sends an inquiry email, the server receives the content.

[1260] Example: A customer sends an email asking, "Please let me know the status of my delivery."

[1261] Step 2:

[1262] The server analyzes the received inquiry.

[1263] Natural language processing (NLP) technology is used to analyze the keywords and intent of inquiries.

[1264] Example: Identifying the "Delivery Status" and "Order Number" to identify that this is a delivery inquiry.

[1265] Step 3:

[1266] The emotion engine recognizes customer emotions from the content of the inquiry.

[1267] Extract emotions from customer text and determine whether they are angry, confused, happy, etc.

[1268] Example: Recognizing the emotion of "confusion" from a customer's writing.

[1269] Step 4:

[1270] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[1271] In addition to basic reply templates, add sentiment-based phrases.

[1272] Example: "Thank you for contacting us. Your order is currently being prepared for shipping." Add an apology such as "We apologize for any inconvenience caused."

[1273] Step 5:

[1274] The server sends the generated auto-reply to the customer.

[1275] The generated reply is sent to the customer's email address.

[1276] Example: An auto-reply email is sent to customer@example.com.

[1277] Step 6:

[1278] The server records the inquiry content, the automatic reply content, and the emotion recognition results.

[1279] The inquiry content, automatic reply message, timestamp, and emotion recognition results are stored in a database.

[1280] For example, customer ID, inquiry content, reply content, emotion recognition results, and timestamp are recorded.

[1281] Step 7:

[1282] The server reports the recorded information to the person in charge.

[1283] A report message containing the inquiry, automatic reply content, and emotion recognition results is sent to the agent's chat application.

[1284] For example, "Customer A has inquired about the delivery status. An automated reply was sent with the following content: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'" This message is sent to the agent.

[1285] Step 8:

[1286] The person in charge will review the report and take additional action if necessary.

[1287] We will review the report via the chat application and conduct any further investigation or action required.

[1288] Example: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then contacts the customer again to say, "Sorry for the wait. Your order is expected to be delivered on the next business day."

[1289] By implementing this series of steps, it is possible to respond to customer inquiries quickly and appropriately, thereby improving customer satisfaction.

[1290] Example 2

[1291] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1292] Modern customer support systems are required to respond to customer inquiries quickly and accurately. However, with a wide range of customer inquiries, it is not easy to generate appropriate responses while understanding the customer's feelings, which can lead to delayed responses or inappropriate replies. In such cases, there is a risk of customer satisfaction decreasing, so an effective system to resolve this issue is needed.

[1293] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving an inquiry from a customer, a means for analyzing the received inquiry using natural language processing technology, and a means for generating an automatic reply based on the analysis results and a means for analyzing the customer's emotions. This enables accurate analysis of the customer's inquiry and generation of an appropriate automatic reply based on the analysis results and the recognized emotions. Furthermore, by combining a means for recording the inquiry content and the automatic reply content in a database and reporting the recorded information and the customer's emotion recognition results to the staff, reporting to the staff and follow-up can be smoothly performed. Furthermore, by using a generative AI model when generating the automatic reply and including a means for creating prompt sentences, more advanced and flexible reply content can be provided. This system can improve customer satisfaction.

[1294] "Means of receiving customer inquiries" refers to the function of receiving customer inquiries via email or web forms.

[1295] "Means for analyzing the content of received inquiries using natural language processing technology" refers to the function of analyzing the content of received inquiries using natural language processing technology and extracting important keywords and intent.

[1296] "Means for analyzing analysis results and customer emotions" refers to the function of recognizing customer emotions using the analyzed inquiry content and an emotion analysis engine.

[1297] "Means for generating an automatic reply" refers to a function that generates an automatic reply based on the analysis results and recognized sentiment.

[1298] A "generative AI model" is an artificial intelligence model used for text generation, such as GPT (Generative Pre-trained Transformer).

[1299] A "prompt" is a text sentence that provides specific context or instructions for input to a generative AI model.

[1300] "Means for sending automated responses to customers" refers to the ability to send generated automated responses to customers' email addresses or other means of communication.

[1301] "Means for recording the inquiry content and the automatic reply content in a database" refers to a function for saving the inquiry content and the generated automatic reply in a database.

[1302] "Means for reporting the recorded information and the emotion recognition results of the customer to the person in charge" refers to a function for reporting the information recorded in the database and the emotion recognition results to the person in charge.

[1303] "Communication Applications" refers to chat applications and messaging services used by personnel to receive reports.

[1304] This invention is a system for efficiently and quickly responding to customer inquiries and improving customer satisfaction. By combining natural language processing technology and emotion recognition, this invention can analyze the content of customer inquiries, recognize the customer's emotions, and adjust the content of automatic replies accordingly. It also uses a generative AI model to generate high-quality automatic replies.

[1305] The system uses the following hardware and software:

[1306] Hardware: Servers, mail servers, database servers

[1307] Software: Natural language processing APIs (Google Cloud Natural Language API, Amazon Comprehend), emotion recognition engines (IBM Watson Tone Analyzer, Microsoft Azure Text Analytics), generative AI models (OpenAI GPT-3, GPT-4), email sending services (SendGrid, Amazon SES), communication applications (Slack, Microsoft Teams)

[1308] 1. Receiving Inquiries

[1309] The server receives customer inquiries from the mail server or web form. For example, the mail server detects new emails and saves them in a specified directory on the system.

[1310] 2. Analysis of inquiry content

[1311] The server reads the query information from the specified directory and analyzes it using natural language processing techniques, such as Google Cloud Natural Language API and Amazon Comprehend, to extract important keywords and intent.

[1312] 3. Customer Emotion Recognition

[1313] The server sends the inquiry to an emotion engine (such as IBM Watson Tone Analyzer or Microsoft Azure Text Analytics), which analyzes the inquiry and recognizes the customer's emotions. The results are recorded as tags such as "confused" or "angry."

[1314] 4. Generate Auto-Reply

[1315] The server generates an automated reply based on the analysis and emotion recognition results. It uses a generative AI model (such as OpenAI's GPT-3) to create a prompt and input it into the model. Examples of prompts include:

[1316] The customer inquired, "Please let me know the status of my delivery," and it was recognized that the customer was confused.

[1317] This is then input into a generative AI model to generate high-quality automated replies.

[1318] 5. Sending Auto-Reply

[1319] The server generates an automatic reply and sends it to the customer. It uses an email sending service such as SendGrid or Amazon SES to send the automatic reply to the customer's email address.

[1320] 6. Recording of inquiries and automated responses

[1321] The server records the inquiry and the automatic reply in a database. Database software such as MySQL or MongoDB is used to store the inquiry, reply, and emotion tag.

[1322] 7. Report to the person in charge

[1323] The server reports the recorded information and emotion recognition results to the person in charge, sending the information to the person in charge's chat room using a communication application such as Slack or Microsoft Teams, and the person in charge can then take necessary follow-up action based on the information.

[1324] As described above, the present invention makes it possible to respond to customer inquiries quickly and accurately, and is expected to improve customer satisfaction.

[1325] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1326] Step 1: Receiving an inquiry

[1327] The server receives inquiries from customers, specifically from a mail server or a web form, and temporarily stores the inquiry information in a specified directory.

[1328] Input: Customer inquiry emails and web form data

[1329] Output: Query data saved in temporary storage directory

[1330] Specific operation: When a new email is detected on the mail server (for example, general mail server software), the system saves the email in the 'inbox' directory.

[1331] Step 2: Analyzing the inquiry

[1332] The server reads the query information from the specified directory, analyzes the data using natural language processing technology, and extracts important keywords and intent.

[1333] Input: Query data read from temporary storage directory

[1334] Output: Extracted keywords and intents (e.g., "Delivery status," "Order number," etc.)

[1335] What it does: It uses a Python script to call the Google Cloud Natural Language API to extract key keywords and intent from the query.

[1336] Step 3: Recognize customer emotions

[1337] The server sends the inquiry to the emotion engine, which analyzes the inquiry and recognizes the customer's emotion tag.

[1338] Input: Inquiry details

[1339] Output: Recognized emotion tag (e.g., "confused," "angry," etc.)

[1340] Specific behavior: Uses IBM Watson Tone Analyzer to analyze inquiries and identify "confusion" and "anger."

[1341] Step 4: Generate an auto-reply

[1342] The server generates an automatic reply based on the analysis results and emotion recognition results, and uses a generative AI model to create prompts and generate reply content.

[1343] Input: Extracted keywords, intent, and sentiment tags

[1344] Output: Generated autoresponder

[1345] Specific operation: The server generates a prompt sentence such as: "A customer inquired, 'Please tell me the delivery status of my item,' and it was recognized that the customer was confused." This is input into the generative AI model to obtain a high-quality automatic reply sentence.

[1346] Step 5: Sending Auto-Reply

[1347] The server uses an email sending service to send the generated auto-reply to the customer.

[1348] Input: Generated autoresponder and customer email address

[1349] Output: Email sent to customer

[1350] Specific behavior: Calls SendGrid's API and sends the generated reply to the customer's email address (e.g., customer@example.com).

[1351] Step 6: Record your inquiries and automated responses

[1352] The server records the inquiry content, the automatic reply content, and the emotion recognition results in a database.

[1353] Input: Inquiry content, auto-reply content, emotion tag

[1354] Output: Records stored in the database

[1355] Specific behavior: Connects to a MySQL database and saves the inquiry content, response content, and emotion tag as records.

[1356] Step 7: Report to the person in charge

[1357] The server reports the recorded information and emotion recognition results to the person in charge, who receives the report through a communication application.

[1358] Input: Information stored in the database and emotion recognition results

[1359] Output: Report message sent to the agent

[1360] Specific behavior: Using the Slack API, a message is sent to the agent's chat room stating, "Customer A has made an inquiry and an automated reply has been sent with the following content. The customer's sentiment was recognized as 'confused'."

[1361] Step 8: Handle follow-up

[1362] The responsible person receives the report and manually follows up as needed.

[1363] Input: Report message received by the agent

[1364] Output: Result of additional customer actions

[1365] Specific actions: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then uses that information to contact the customer again, saying, "Sorry for the wait. Your order is expected to be delivered the next business day."

[1366] (Application example 2)

[1367] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1368] Conventional food delivery services have issues with delayed responses to customer inquiries and low customer satisfaction due to uniform replies that do not consider customer feelings. In addition, it takes time and effort for staff to manually analyze the content of inquiries, understand the feelings of customers, and provide appropriate replies, which is inefficient.

[1369] The specific processing 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 receiving inquiries from customers, means for analyzing the content of the received inquiries, and means for generating an automatic reply based on the analysis results. This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's emotions, thereby improving customer satisfaction. In addition, by using an emotion engine to recognize the customer's emotions and adjusting the content of the automatic reply according to the recognized emotions, more appropriate and personalized responses can be provided.

[1370] "Customer" means any person who purchases or uses a product or service.

[1371] An "inquiry" is when a customer raises a question or concern about a product or service.

[1372] "Means for receiving" refers to the method or device by which the system collects customer inquiries.

[1373] "Means for analysis" refers to the method or device by which the system understands and analyzes the content of the query.

[1374] "Means for generating an automatic reply" refers to a method or device by which the system automatically generates a reply based on the analysis results.

[1375] "Means for sending" refers to a method or device for delivering the generated reply to the customer.

[1376] "Recording means" refers to a method or device for saving the contents of inquiries and automatic replies.

[1377] "Reporting means" means the method or device by which recorded information is communicated to responsible personnel.

[1378] The "emotion engine" is a system that recognizes customer emotions from the content of inquiries.

[1379] "Adjusting means" is a method or device for varying the content of the automated response based on perceived customer sentiment.

[1380] The present invention relates to a customer support system for food delivery services. This system uses an emotion engine to recognize customer emotions and generate automatic replies accordingly, in order to respond quickly and accurately to customer inquiries.

[1381] System Configuration

[1382] The system includes the following major components:

[1383] 1. Server

[1384] Receiving inquiries, analyzing them, and generating and sending automatic replies

[1385] 2. Database

[1386] Record of inquiries and automatic replies

[1387] 3. Emotion Engine

[1388] Customer Emotion Recognition

[1389] 4. Natural Language Processing (NLP) Libraries

[1390] Analysis of inquiry content

[1391] 5. Smartphone App

[1392] An interface for customers to make inquiries

[1393] System Operation Details

[1394] Hardware and Software Usage

[1395] The server (e.g., AWS EC2, Google Cloud Compute Engine) handles all the processes from receiving inquiries to sending automated replies. The server receives inquiries via a mail server or web form and analyzes them using a natural language processing (NLP) library (e.g., spaCy, NLTK).

[1396] An emotion engine (e.g., Google Cloud Natural Language API, AWS Comprehend) recognizes customer emotions from the content of inquiries and identifies emotions such as "confused" or "angry."

[1397] The database (e.g., MySQL, PostgreSQL) records the received inquiries, the generated auto-replies, and the emotion recognition results.

[1398] The smartphone app is developed using frameworks such as React Native and provides a user interface for customers to submit inquiries.

[1399] Details of data processing and calculation

[1400] 1. Receiving Inquiries

[1401] The server receives customer inquiries via email or web form, and the inquiries are stored in a database in text format.

[1402] 2. Analysis of inquiry content

[1403] The server uses an NLP library to analyze the inquiry and extract important keywords such as delivery delays and order cancellations.

[1404] 3. Emotion recognition

[1405] The server uses an emotion engine to recognize the customer's emotions from the analyzed inquiry, for example, identifying emotions such as "angry" or "confused."

[1406] 4. Generate Auto-Reply

[1407] The server generates an automatic reply based on the analysis results and emotion recognition results. For example, if a delivery is delayed and the customer is angry, the server generates a reply that includes an apology, such as "We apologize for the wait. Your delivery is currently on its way. We will deliver it to you soon."

[1408] 5. Sending Auto-Reply

[1409] The server notifies the customer's app of the generated auto-reply.

[1410] 6. Records

[1411] The server records the inquiry and the automatic reply in a database and shares it with the customer support team.

[1412] 7. Report to the person in charge

[1413] The server reports the automatically generated reply and customer sentiment to the agent, and if necessary, further follow-up is carried out.

[1414] Examples of concrete examples and prompts

[1415] Specific examples

[1416] Examples of receiving inquiries:

[1417] A customer sends an inquiry saying, "My pizza order hasn't arrived yet, what's going on?"

[1418] Examples of emotion recognition:

[1419] The emotion engine analyzes the content of the email and recognizes the emotion "anger."

[1420] Examples of generating and sending auto-replies:

[1421] Based on the keyword "delayed delivery" and the emotion "angry," the server generates an automatic reply saying, "We apologize for the wait. Your order is currently being delivered. We will deliver it to you soon." and sends it to the customer.

[1422] Prompt Sentence Examples

[1423] Prompts to recognize customer sentiment:

[1424] "Customer asks: 'My pizza order hasn't arrived yet, what's going on?' Recognize customer sentiment and generate appropriate automated responses."

[1425] This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[1426] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1427] Step 1:

[1428] Receiving inquiries

[1429] Input: A customer submits an inquiry through a smartphone app.

[1430] Specific behavior: A customer uses the smartphone app's inquiry form or chat function to enter and send an inquiry in text format, such as, "The item I ordered hasn't arrived yet. What's going on?"

[1431] Data processing: The smartphone app generates an API request to send the inquiry to the server.

[1432] Output: The server receives the API request and saves the query in text format.

[1433] Step 2:

[1434] Analysis of inquiry content

[1435] Input: Text data of the query received by the server.

[1436] What it does: The server parses the query using a natural language processing (NLP) library (e.g., spaCy or NLTK) and extracts important keywords (e.g., "delivery" or "delay").

[1437] Data processing: Tokenize the text data, tag each token with a part of speech, and extract important keywords.

[1438] Output: A list of keywords extracted as a result of the analysis.

[1439] Step 3:

[1440] emotion recognition

[1441] Input: Text data of the inquiry and extracted keyword list.

[1442] Specific operation: Using an emotion engine (e.g., Google Cloud Natural Language API or AWS Comprehend), the server recognizes the customer's emotion from the inquiry content.

[1443] Data processing: The emotion engine analyzes the text data and calculates an emotion score. For example, it makes a judgment such as "The customer is angry" based on the emotion score.

[1444] Output: Analysis results including recognized emotion information (e.g., "angry").

[1445] Step 4:

[1446] Generate auto-replies

[1447] Input: Analysis results (keyword list and sentiment information).

[1448] What it does: The server generates an appropriate auto-reply based on the analysis results, choosing a basic reply from a template and adjusting the content based on the sentiment.

[1449] Data processing: Generate an automatic reply by combining template sentences and analysis results. For example, based on "Delivery delayed" and "Angry," generate a reply such as "Sorry for the wait. Delivery is currently in progress. We will deliver it to you soon."

[1450] Output: The generated auto-reply.

[1451] Step 5:

[1452] Sending automatic replies

[1453] Input: The generated auto-response.

[1454] Specific operation: Generates an API request to notify the customer's smartphone app of the automatic reply message generated by the server.

[1455] Data processing: Auto-reply messages are embedded into API requests and sent to the customer's app.

[1456] Output: The automated reply that will be displayed on the customer's smartphone app.

[1457] Step 6:

[1458] Record of inquiries and automatic replies

[1459] Input: Enquiry and generated auto-reply.

[1460] Specific operation: The server records the inquiry and the automatic reply in a database.

[1461] Data processing: Convert the text data into database format and save it along with the inquiry ID.

[1462] Output: A record of the inquiry and auto-reply stored in the database.

[1463] Step 7:

[1464] Report to the person in charge

[1465] Input: Inquiry details, auto-reply details and emotional information recorded in the database.

[1466] Specific operation: The server generates a notification to report to the agent and sends it to the agent's chat application.

[1467] Data processing: Compile inquiry details, automatic response details, and emotional information and convert them into a report format.

[1468] Output: Report notification sent to the responsible party.

[1469] This series of steps makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[1470] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1471] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1472] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1473] [Fourth embodiment]

[1474] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1475] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1476] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1477] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1478] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1479] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1480] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1481] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1482] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1483] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1484] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1485] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1486] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1487] The present invention is a system for responding to customer inquiries, and an embodiment thereof will be described below. This system includes a means for receiving customer inquiries, a means for analyzing the content of the received inquiry, a means for generating an automatic reply based on the analysis results, a means for sending the generated automatic reply to the customer, a means for recording the content of the inquiry and the automatic reply, and a means for reporting the recorded information to a person in charge.

[1488] System Overview

[1489] This system improves customer satisfaction and streamlines responses by having the server receive customer inquiries and then appropriately analyze and respond to the content. Furthermore, the system also has the function of recording the content of the automatic response and reporting it to the person in charge, allowing the person in charge to follow up promptly and appropriately.

[1490] Program processing explanation

[1491] 1. Receiving Inquiries

[1492] The server receives customer inquiries from the mail server or web form.

[1493] Example: The server receives an email from a customer ("I want to know the delivery status of my item").

[1494] 2. Analysis of inquiry content

[1495] The server uses natural language processing techniques to analyze the received query.

[1496] Example: Detect the keyword "delivery status" in the inquiry text and determine that it is a question about delivery.

[1497] 3. Generate Auto-Reply

[1498] The server generates an appropriate automatic reply based on the analysis results.

[1499] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[1500] 4. Sending Auto-Reply

[1501] The server sends the generated auto-reply to the customer.

[1502] Example: Send the generated text to the customer's email address (customer@example.com).

[1503] 5. Recording of inquiries and automated responses

[1504] The server records the inquiry and the automatic reply in a database.

[1505] Example: Save the inquiry and auto-reply message along with the customer ID.

[1506] 6. Report to the person in charge

[1507] The server sends a notification to the agent's chat application to report the recorded information to the agent.

[1508] For example, a message will be sent to the agent's chat window saying, "Customer A has inquired about the delivery status. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details.'"

[1509] Specific examples

[1510] 1. Specific examples of receiving inquiries

[1511] A customer sends an email asking, "What's the delivery status of my item?"

[1512] The server receives this email and begins analysis according to the analysis plan.

[1513] 2. Example of generating and sending an auto-reply

[1514] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[1515] The server immediately sends this automatic reply email to the customer.

[1516] 3. Examples of Recording and Reporting

[1517] The server records the inquiry and the automatic reply in a database along with information such as "Customer A, delivery status, reply content, time."

[1518] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[1519] Specific examples of follow-up processes

[1520] Once the report is received, the responsible person will review the situation and obtain further information if necessary (e.g., contact the delivery company).

[1521] The person in charge will then give the customer a specific reply, saying, "We apologize for the wait. Your order is scheduled to be delivered on the next business day."

[1522] In this way, by implementing this system, it becomes possible to respond quickly and appropriately to customer inquiries, thereby improving customer satisfaction (CS) and operational efficiency.

[1523] The processing flow will be explained below.

[1524] Step 1:

[1525] The server receives inquiries from customers.

[1526] When a customer submits an inquiry via email or web form, the server receives the content.

[1527] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[1528] Step 2:

[1529] The server analyzes the received inquiry.

[1530] Natural language processing (NLP) technology is used to analyze keywords and intent in inquiries.

[1531] Example: Extract important keywords such as "delivery status" and "order number".

[1532] Step 3:

[1533] The server generates an automatic reply based on the analysis results.

[1534] Choose the appropriate response template for your inquiry and customize it as needed.

[1535] Example: Generates the sentence, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[1536] Step 4:

[1537] The server sends the generated auto-reply to the customer.

[1538] The generated reply is sent to the customer's email address.

[1539] Example: An auto-reply email is sent to customer@example.com.

[1540] Step 5:

[1541] The server records the inquiry and the automatic reply.

[1542] Information such as inquiry details, auto-reply messages, and timestamps is stored in a database.

[1543] Example: Customer ID, inquiry, reply, and timestamp are recorded in the database.

[1544] Step 6:

[1545] The server reports the recorded information to the person in charge.

[1546] Send a report message to the agent's chat application about the inquiry and an automated reply.

[1547] For example, "Customer A has inquired about the status of their delivery. An automated reply has been sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[1548] Step 7:

[1549] The person in charge will review the report and take additional action if necessary.

[1550] We will review reports via the chat application and take further action if necessary.

[1551] Example: Contact the delivery company to confirm the specific delivery date and time, and then contact the customer again to say, "Sorry for the wait. Your order is expected to arrive the next business day."

[1552] Example 1

[1553] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1554] Conventional inquiry response systems required manual responses to customer inquiries, which was time-consuming and labor-intensive. Furthermore, the automated responses were formulaic and inflexible, which posed a risk of lowering customer satisfaction. This made it difficult to balance the efficiency and quality of customer responses.

[1555] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1556] In this invention, the server includes means for receiving inquiries from customers, means for analyzing the content of the received inquiries using natural language processing technology, means for generating an automatic reply using a generative AI model based on the analysis results, means for sending the generated automatic reply to the customer, means for recording the content of the inquiry and the content of the automatic reply in a database, and means for reporting the recorded information to a person in charge using a chat application. This enables quick and appropriate responses to customer inquiries, thereby improving the efficiency and quality of customer service.

[1557] "Means of receiving inquiries from customers" refers to the function by which the server receives the content of inquiries from customers via a mail server, web form, etc.

[1558] "Means of analyzing the content of received inquiries using natural language processing technology" refers to the function of extracting and analyzing the intent and keywords of the inquiry using text analysis tools and libraries (e.g., NLTK, spaCy) from the content of received inquiries.

[1559] "Means for generating an automatic reply using a generative AI model based on the analysis results" refers to a function that uses a generative AI model (e.g., GPT-3) to generate an optimal automatic reply based on the analyzed inquiry content.

[1560] "Means for sending generated auto-replies to customers" refers to the function of sending generated auto-replies to customers' email addresses using the SMTP protocol.

[1561] "Means for recording the inquiry content and the automatic reply content in a database" refers to the function of saving the received inquiry content and the generated automatic reply content in a database (e.g., MySQL, MongoDB) along with a unique ID and timestamp.

[1562] "Means of reporting recorded information to the person in charge using a chat application" refers to the function of notifying and reporting recorded information to the chat application used by the person in charge (e.g., Slack, Microsoft Teams).

[1563] "Means for inputting a prompt sentence into a generative AI model" refers to the operation of inputting a customized prompt sentence into a generative AI model and generating an automatic reply sentence.

[1564] "Means for a person in charge to receive reports through a chat application" refers to a function for a person in charge to receive reports regarding automatic replies through a chat application used by the person in charge.

[1565] The present invention is a system for responding promptly and appropriately to customer inquiries, and a specific embodiment thereof is described below. This system includes functions for receiving customer inquiries, analyzing them, generating automatic replies using a generative AI model, recording the replies, and reporting them to a responsible person.

[1566] System Configuration

[1567] Hardware Configuration

[1568] The main hardware configuration of this system consists of a server, a mail server, end-user terminals, and terminals for personnel. The server is a machine equipped with a high-performance processor and large memory, and links the mail server with a database to process various inquiries.

[1569] Software Configuration

[1570] The software configuration of this system is as follows:

[1571] 1. Mail server software: Mail server software such as Postfix or Sendmail is used to receive inquiry emails from customers.

[1572] 2. Web Forms (Backend): The front end is built with HTML and JavaScript, and the back end is implemented using Python languages ​​such as Flask and Django.

[1573] 3. Natural Language Processing Libraries: Use natural language processing libraries such as NLTK or spaCy.

[1574] 4. Generative AI models: Generative AI models, such as GPT-3, are used to generate automated replies.

[1575] 5. Database: Use MySQL or MongoDB to record inquiries and auto-reply content.

[1576] 6. Chat applications: Use APIs such as Slack or Microsoft Teams to send reports to the appropriate person.

[1577] Specific examples

[1578] 1. Specific examples of receiving inquiries

[1579] A customer sends an email asking, "What is the status of my delivery?" The server receives this email through the Postfix mail server and starts processing it as a new inquiry.

[1580] 2. Analysis of inquiry content

[1581] The server analyzes the received email using a natural language processing library (e.g., NLTK). This analysis involves tokenizing the sentence, tagging parts of speech, and extracting keywords. For example, the keyword "delivery status" is extracted, and it is determined that this is a delivery inquiry.

[1582] 3. Generate Auto-Reply

[1583] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. An example of a prompt sentence is given as input: "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." As a result, the reply sentence generated is "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with details shortly."

[1584] 4. Sending Auto-Reply

[1585] The server sends the generated auto-reply to the customer's email address using the SMTP protocol.

[1586] 5. Recording of inquiries and automated responses

[1587] The server records the inquiry and automatic reply in a MySQL database along with a unique ID and timestamp.

[1588] 6. Report to the person in charge

[1589] The server sends the recorded information to the chat application (e.g., Slack) used by the person in charge. The message "A new inquiry has been received" is displayed along with the inquiry details and an automatic reply.

[1590] As described above, the system of the present invention is designed to support prompt and appropriate responses to customer inquiries, and by using this system, customer satisfaction and response efficiency can be significantly improved.

[1591] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1592] Step 1:

[1593] The server receives customer inquiries from a mail server or web form. The input is the customer's inquiry, which is received via a mail server (Postfix or Sendmail). The server detects new emails and saves them in storage. Specifically, the server monitors ports 25 and 587 and saves new emails in storage.

[1594] Step 2:

[1595] The server analyzes the received inquiry using natural language processing technology. Specifically, it takes the received text data (inquiry content) as input and uses natural language processing libraries (NLTK or spaCy) to tokenize, tag parts of speech, and extract keywords. This process outputs the intent of the inquiry and key keywords. Specifically, the server splits the received text, extracts keywords such as "delivery status," and saves them as the analysis results.

[1596] Step 3:

[1597] Based on the analysis results, the server uses a generative AI model (GPT-3) to generate an automatic reply. The input is the analysis results and a prompt sentence, which is fed into the generative AI model to output the automatic reply. A specific example of a prompt sentence is, "A customer has inquired, 'Please tell me the delivery status of my item.' Please generate a reply in response to that." In concrete terms, the server inputs the prompt sentence into the generative AI model and generates an appropriate reply sentence (e.g., "Thank you for contacting us. Your order is currently being prepared for shipping.").

[1598] Step 4:

[1599] The server sends the generated auto-reply to the customer using the SMTP protocol. The input is the generated auto-reply message and the customer's email address, and this is sent via the SMTP server. Specifically, the server inserts the generated reply message into the body of the email and sends it to the customer's email address.

[1600] Step 5:

[1601] The server records the inquiry and auto-reply content in a database. The input is the inquiry, auto-reply content, and related metadata (e.g., unique ID and timestamp), which are then saved in a database (MySQL or MongoDB). Specifically, the server executes an INSERT query to save the inquiry and auto-reply content in the database.

[1602] Step 6:

[1603] The server reports the recorded information to the person in charge using a chat application. The input is the recorded inquiry content and the automatic reply content, and this is notified to the person in charge's chat application (Slack or Microsoft Teams). Specifically, the server uses the chat API (Webhook URL) to send a message saying "A new inquiry has been received."

[1604] (Application example 1)

[1605] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1606] Traditional customer support systems require a lot of time and effort to respond to inquiries, which can lead to poor customer satisfaction. Furthermore, the lack of real-time feedback makes it difficult for agents to follow up appropriately. This can result in poor service quality and lead to customer attrition.

[1607] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1608] In this invention, the server includes a means for receiving inquiries from customers, a means for analyzing the content of the received inquiries using natural language processing technology, and a means for generating an automatic reply using a generative AI model based on the analysis results, thereby enabling quick and accurate responses to customer inquiries and enabling personnel to receive reports in real time via a chat notification system.

[1609] "Means for receiving inquiries" refers to methods for receiving inquiries from customers, such as email, a web form, or a smartphone application.

[1610] The "means for analyzing the content of the inquiry" is a method for understanding and analyzing the content of the received inquiry using natural language processing technology.

[1611] "Generative AI model" means an artificial intelligence model used to generate appropriate automated replies based on the analysis results.

[1612] "Means for generating automatic replies" refers to methods for automatically creating replies based on the analyzed inquiry content using pre-prepared templates or generative AI models.

[1613] The "means for sending automatic replies" refers to the method for sending the generated automatic replies to customers, which may be email, a smartphone app, or other means of communication.

[1614] The "means for recording the contents of inquiries and automatic replies" is a method for storing the contents of received inquiries and the automatic replies thereto in a database.

[1615] The "means for reporting the recorded information to the person in charge" is a method for notifying the person in charge of the recorded information, and is carried out via a chat notification system.

[1616] "Means of interacting with customers through smartphone applications" refers to a method by which customers can make inquiries and receive replies via their smartphones.

[1617] The present invention can be applied to a customer support system for a mail-order site. A specific embodiment of this system will be described below.

[1618] Hardware and software used

[1619] Server: Receives and analyzes inquiries, and generates and sends automatic replies.

[1620] Node.js, Express: Server Application Frameworks

[1621] MySQL: Database Management System

[1622] NLTK, SpaCy: Natural Language Processing Libraries

[1623] Firebase: Notification System

[1624] Smartphone: Customers submit inquiries and receive automated responses.

[1625] Customer applications: contact form, inbox

[1626] System processing flow

[1627] 1. Receiving Inquiries

[1628] A user submits an inquiry using the inquiry form of the smartphone application.

[1629] The server receives this query and analyzes the content using natural language processing techniques.

[1630] Example: A customer sends an inquiry asking, "What is the delivery status of my item?"

[1631] 2. Analysis of inquiry content

[1632] The server parses the query using an NLP library (NLTK or SpaCy).

[1633] Identify the content of the inquiry based on keywords and context.

[1634] Example: Extract the keyword "delivery status" to determine that this is an inquiry about delivery.

[1635] 3. Generate Auto-Reply

[1636] Based on the analysis results, the server generates an automatic reply using a generative AI model.

[1637] Customize your reply based on templates as needed.

[1638] For example, generate an automated reply that says, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more information shortly."

[1639] 4. Sending Auto-Reply

[1640] The server sends an automated reply generated using Firebase to the customer's smartphone application.

[1641] Example: A generated auto-reply is displayed in the customer's app.

[1642] 5. Recording of inquiries and automated responses

[1643] The server records the inquiry and the automatic reply in a MySQL database.

[1644] This includes data such as customer ID, inquiry content, auto-reply content, and time of sending.

[1645] Example: Customer A's inquiries and automatic replies are stored in a database.

[1646] 6. Report to the person in charge

[1647] The server notifies the responsible party through Firebase and reports the recorded information.

[1648] Based on this, the person in charge will carry out the necessary follow-up.

[1649] Example: "Customer A contacted us regarding the status of their delivery. We sent an automated reply with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.'"

[1650] Examples of concrete examples and prompts

[1651] Specific examples

[1652] 1. Specific examples of receiving inquiries

[1653] A customer sends an email asking, "What's the delivery status of my item?"

[1654] The server receives this email and begins analyzing it.

[1655] 2. Example of generating and sending an auto-reply

[1656] The server detects the keyword "delivery status" and generates a reply saying, "Thank you for contacting us. Your order is currently being prepared for shipping. We will contact you with more details shortly."

[1657] The server immediately sends this automatic reply email to the customer.

[1658] 3. Examples of Recording and Reporting

[1659] The server records the inquiry and the automatic reply in a database along with information such as customer ID, delivery status, reply content, and time.

[1660] The server sends a report message to the agent's chat application saying, "Customer A has made an inquiry. An automatic reply has been sent with the following content..."

[1661] Prompt Sentence Examples

[1662] A customer asks, "What's the status of my delivery?" You want to generate an automated reply with the following information:

[1663] Delivery status

[1664] Thank you for contacting me

[1665] Expression to contact you later

[1666] This allows the system to respond to customer inquiries quickly and accurately, and enables staff to grasp information in real time.

[1667] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1668] Step 1:

[1669] Receiving inquiries

[1670] A user submits a question using the inquiry form of the smartphone application, and the server receives the inquiry. The input is the inquiry information submitted by the user, and the output is the raw inquiry data stored in the server.

[1671] Step 2:

[1672] Analysis of inquiry content

[1673] The server analyzes the received query using natural language processing techniques. Specifically, it uses libraries such as NLTK and SpaCy to tokenize the text and extract keywords and context. The input is the raw query data, and the output is the analyzed query data (e.g., including the keyword "delivery status").

[1674] Step 3:

[1675] Generate auto-replies

[1676] The server generates an auto-reply message using a generative AI model based on the analysis results, including customizing pre-defined templates. The input is the analyzed inquiry data and template information, and the output is the generated auto-reply message.

[1677] Step 4:

[1678] Sending automatic replies

[1679] The server sends the generated auto-reply to the customer's smartphone app. Notifications are sent using Firebase. The input is the generated auto-reply message, and the output is the reply message displayed on the user's device.

[1680] Step 5:

[1681] Record of inquiries and automatic replies

[1682] The server logs queries and generated auto-replies in a MySQL database. The input is the raw query data and the generated auto-reply messages, and the output is the logged data stored in the database.

[1683] Step 6:

[1684] Report to the person in charge

[1685] The server sends notifications to the person in charge via Firebase and reports the recorded information. The input is the recorded data and the content of the notification message, and the output is the report message displayed on the person's terminal.

[1686] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1687] The present invention is a system for responding to customer inquiries efficiently and quickly, thereby improving customer satisfaction. By combining an emotion engine, the present invention can recognize customer emotions and adjust the content of automatic replies accordingly.

[1688] System Overview

[1689] The system includes the following means:

[1690] 1. A means of receiving customer inquiries.

[1691] 2. A means of analyzing the inquiries received.

[1692] 3. A means of generating automated responses based on the analysis results.

[1693] 4. A means of sending generated auto-replies to customers.

[1694] 5. A means of recording inquiries and automated responses.

[1695] 6. A means of reporting recorded information to personnel.

[1696] 7. An emotion engine that recognizes customer emotions from the inquiries received.

[1697] 8. A means to tailor the content of auto-replies based on perceived customer sentiment.

[1698] 9. A means of reporting customer emotions recognized by the emotion engine to the agent.

[1699] Program processing explanation

[1700] 1. Receiving Inquiries

[1701] The server receives customer inquiries from the mail server or web form.

[1702] Example: The mail server detects new mail and forwards the inquiry mail to the system.

[1703] 2. Analysis of inquiry content

[1704] The server uses natural language processing techniques to analyze the received query.

[1705] Example: Extract important keywords such as "delivery status" and "order number".

[1706] 3. Customer Emotion Recognition

[1707] The emotion engine analyzes customer emotions from the inquiry, for example recognizing whether the customer is angry, distressed, or happy.

[1708] The results of the emotion recognition are recorded as information such as "confused" or "angry."

[1709] 4. Generate Auto-Reply

[1710] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[1711] For example: Generate a basic reply like "Thank you for contacting us. Your order is currently being prepared for shipping." But if the customer is upset, add an apology like "We're sorry for the inconvenience. We're working to resolve the issue soon."

[1712] 5. Sending Auto-Reply

[1713] The server sends the generated auto-reply to the customer.

[1714] Example: Send the generated text to the customer's email address (customer@example.com).

[1715] 6. Recording of inquiries and automated responses

[1716] The server records the inquiry and the automatic reply in a database.

[1717] Example: Save the inquiry and auto-reply message along with the customer ID.

[1718] 7. Report to the person in charge

[1719] The server reports the recorded information and the customer's emotions recognized by the emotion engine to the agent.

[1720] For example: "Customer A inquired about the status of their delivery. An automated reply was sent with the following message: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'"

[1721] The agent receives the report through a chat application and takes further appropriate action based on it.

[1722] Specific examples

[1723] 1. Specific examples of receiving inquiries

[1724] A customer sends an email asking, "Please let me know the delivery status of my item."

[1725] The server receives this email and begins analyzing it.

[1726] 2. Specific examples of emotion recognition

[1727] The emotion engine analyzes the content of the email and recognizes the emotion "confusion."

[1728] The result of the emotion recognition is recorded as "The customer is confused."

[1729] 3. Example of generating and sending an auto-reply

[1730] Based on the keyword "delivery status," the server generates an automatic reply message saying, "Thank you for contacting us. Your order is currently being prepared for shipping."

[1731] Because the customer's emotion is recognized as "confused," an additional sentence is also generated: "We apologize for the inconvenience."

[1732] The generated reply is sent to the customer's email address.

[1733] 4. Examples of Recording and Reporting

[1734] The server records the inquiry content, the content of the automatic reply, and the results of emotion recognition in a database.

[1735] The server reports to the person in charge, "Customer A made an inquiry and an automatic reply was sent with the following content. The customer's emotion was also recognized as 'confused.'"

[1736] 5. Specific examples of follow-up processes

[1737] The person in charge receives the report and checks the situation.

[1738] The person in charge will contact the delivery company to confirm the specific delivery date and time.

[1739] The person in charge will contact the customer again and say, "Sorry for the wait. Your order is scheduled to be delivered on the next business day."

[1740] According to the present invention, customer satisfaction can be further improved by recognizing the emotions of customers and taking appropriate measures accordingly.

[1741] The processing flow will be explained below.

[1742] Step 1:

[1743] The server receives customer inquiries from the mail server or web form.

[1744] When a customer sends an inquiry email, the server receives the content.

[1745] Example: A customer sends an email asking, "Please let me know the status of my delivery."

[1746] Step 2:

[1747] The server analyzes the received inquiry.

[1748] Natural language processing (NLP) technology is used to analyze the keywords and intent of inquiries.

[1749] Example: Identifying the "Delivery Status" and "Order Number" to identify that this is a delivery inquiry.

[1750] Step 3:

[1751] The emotion engine recognizes customer emotions from the content of the inquiry.

[1752] Extract emotions from customer text and determine whether they are angry, confused, happy, etc.

[1753] Example: Recognizing the emotion of "confusion" from a customer's writing.

[1754] Step 4:

[1755] The server generates an appropriate automatic reply based on the analysis results and emotion recognition results.

[1756] In addition to basic reply templates, add sentiment-based phrases.

[1757] Example: "Thank you for contacting us. Your order is currently being prepared for shipping." Add an apology such as "We apologize for any inconvenience caused."

[1758] Step 5:

[1759] The server sends the generated auto-reply to the customer.

[1760] The generated reply is sent to the customer's email address.

[1761] Example: An auto-reply email is sent to customer@example.com.

[1762] Step 6:

[1763] The server records the inquiry content, the automatic reply content, and the emotion recognition results.

[1764] The inquiry content, automatic reply message, timestamp, and emotion recognition results are stored in a database.

[1765] For example, customer ID, inquiry content, reply content, emotion recognition results, and timestamp are recorded.

[1766] Step 7:

[1767] The server reports the recorded information to the person in charge.

[1768] A report message containing the inquiry, automatic reply content, and emotion recognition results is sent to the agent's chat application.

[1769] For example, "Customer A has inquired about the delivery status. An automated reply was sent with the following content: 'Thank you for contacting us. Your order is currently being prepared for shipment. We will contact you with more information.' The customer's emotion was also identified as 'confused.'" This message is sent to the agent.

[1770] Step 8:

[1771] The person in charge will review the report and take additional action if necessary.

[1772] We will review the report via the chat application and conduct any further investigation or action required.

[1773] Example: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then contacts the customer again to say, "Sorry for the wait. Your order is expected to be delivered on the next business day."

[1774] By implementing this series of steps, it is possible to respond to customer inquiries quickly and appropriately, thereby improving customer satisfaction.

[1775] Example 2

[1776] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1777] Modern customer support systems are required to respond to customer inquiries quickly and accurately. However, with a wide range of customer inquiries, it is not easy to generate appropriate responses while understanding the customer's feelings, which can lead to delayed responses or inappropriate replies. In such cases, there is a risk of customer satisfaction decreasing, so an effective system to resolve this issue is needed.

[1778] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for receiving an inquiry from a customer, a means for analyzing the received inquiry using natural language processing technology, and a means for generating an automatic reply based on the analysis results and a means for analyzing the customer's emotions. This enables accurate analysis of the customer's inquiry and generation of an appropriate automatic reply based on the analysis results and the recognized emotions. Furthermore, by combining a means for recording the inquiry content and the automatic reply content in a database and reporting the recorded information and the customer's emotion recognition results to the staff, reporting to the staff and follow-up can be smoothly performed. Furthermore, by using a generative AI model when generating the automatic reply and including a means for creating prompt sentences, more advanced and flexible reply content can be provided. This system can improve customer satisfaction.

[1779] "Means of receiving customer inquiries" refers to the function of receiving customer inquiries via email or web forms.

[1780] "Means for analyzing the content of received inquiries using natural language processing technology" refers to the function of analyzing the content of received inquiries using natural language processing technology and extracting important keywords and intent.

[1781] "Means for analyzing analysis results and customer emotions" refers to the function of recognizing customer emotions using the analyzed inquiry content and an emotion analysis engine.

[1782] "Means for generating an automatic reply" refers to a function that generates an automatic reply based on the analysis results and recognized sentiment.

[1783] A "generative AI model" is an artificial intelligence model used for text generation, such as GPT (Generative Pre-trained Transformer).

[1784] A "prompt" is a text sentence that provides specific context or instructions for input to a generative AI model.

[1785] "Means for sending automated responses to customers" refers to the ability to send generated automated responses to customers' email addresses or other means of communication.

[1786] "Means for recording the inquiry content and the automatic reply content in a database" refers to a function for saving the inquiry content and the generated automatic reply in a database.

[1787] "Means for reporting the recorded information and the emotion recognition results of the customer to the person in charge" refers to a function for reporting the information recorded in the database and the emotion recognition results to the person in charge.

[1788] "Communication Applications" refers to chat applications and messaging services used by personnel to receive reports.

[1789] This invention is a system for efficiently and quickly responding to customer inquiries and improving customer satisfaction. By combining natural language processing technology and emotion recognition, this invention can analyze the content of customer inquiries, recognize the customer's emotions, and adjust the content of automatic replies accordingly. It also uses a generative AI model to generate high-quality automatic replies.

[1790] The system uses the following hardware and software:

[1791] Hardware: Servers, mail servers, database servers

[1792] Software: Natural language processing APIs (Google Cloud Natural Language API, Amazon Comprehend), emotion recognition engines (IBM Watson Tone Analyzer, Microsoft Azure Text Analytics), generative AI models (OpenAI GPT-3, GPT-4), email sending services (SendGrid, Amazon SES), communication applications (Slack, Microsoft Teams)

[1793] 1. Receiving Inquiries

[1794] The server receives customer inquiries from the mail server or web form. For example, the mail server detects new emails and saves them in a specified directory on the system.

[1795] 2. Analysis of inquiry content

[1796] The server reads the query information from the specified directory and analyzes it using natural language processing techniques, such as Google Cloud Natural Language API and Amazon Comprehend, to extract important keywords and intent.

[1797] 3. Customer Emotion Recognition

[1798] The server sends the inquiry to an emotion engine (such as IBM Watson Tone Analyzer or Microsoft Azure Text Analytics), which analyzes the inquiry and recognizes the customer's emotions. The results are recorded as tags such as "confused" or "angry."

[1799] 4. Generate Auto-Reply

[1800] The server generates an automated reply based on the analysis and emotion recognition results. It uses a generative AI model (such as OpenAI's GPT-3) to create a prompt and input it into the model. Examples of prompts include:

[1801] The customer inquired, "Please let me know the status of my delivery," and it was recognized that the customer was confused.

[1802] This is then input into a generative AI model to generate high-quality automated replies.

[1803] 5. Sending Auto-Reply

[1804] The server generates an automatic reply and sends it to the customer. It uses an email sending service such as SendGrid or Amazon SES to send the automatic reply to the customer's email address.

[1805] 6. Recording of inquiries and automated responses

[1806] The server records the inquiry and the automatic reply in a database. Database software such as MySQL or MongoDB is used to store the inquiry, reply, and emotion tag.

[1807] 7. Report to the person in charge

[1808] The server reports the recorded information and emotion recognition results to the person in charge, sending the information to the person in charge's chat room using a communication application such as Slack or Microsoft Teams, and the person in charge can then take necessary follow-up action based on the information.

[1809] As described above, the present invention makes it possible to respond to customer inquiries quickly and accurately, and is expected to improve customer satisfaction.

[1810] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1811] Step 1: Receiving an inquiry

[1812] The server receives inquiries from customers, specifically from a mail server or a web form, and temporarily stores the inquiry information in a specified directory.

[1813] Input: Customer inquiry emails and web form data

[1814] Output: Query data saved in temporary storage directory

[1815] Specific operation: When a new email is detected on the mail server (for example, general mail server software), the system saves the email in the 'inbox' directory.

[1816] Step 2: Analyzing the inquiry

[1817] The server reads the query information from the specified directory, analyzes the data using natural language processing technology, and extracts important keywords and intent.

[1818] Input: Query data read from temporary storage directory

[1819] Output: Extracted keywords and intents (e.g., "Delivery status," "Order number," etc.)

[1820] What it does: It uses a Python script to call the Google Cloud Natural Language API to extract key keywords and intent from the query.

[1821] Step 3: Recognize customer emotions

[1822] The server sends the inquiry to the emotion engine, which analyzes the inquiry and recognizes the customer's emotion tag.

[1823] Input: Inquiry details

[1824] Output: Recognized emotion tag (e.g., "confused," "angry," etc.)

[1825] Specific behavior: Uses IBM Watson Tone Analyzer to analyze inquiries and identify "confusion" and "anger."

[1826] Step 4: Generate an auto-reply

[1827] The server generates an automatic reply based on the analysis results and emotion recognition results, and uses a generative AI model to create prompts and generate reply content.

[1828] Input: Extracted keywords, intent, and sentiment tags

[1829] Output: Generated autoresponder

[1830] Specific operation: The server generates a prompt sentence such as: "A customer inquired, 'Please tell me the delivery status of my item,' and it was recognized that the customer was confused." This is input into the generative AI model to obtain a high-quality automatic reply sentence.

[1831] Step 5: Sending Auto-Reply

[1832] The server uses an email sending service to send the generated auto-reply to the customer.

[1833] Input: Generated autoresponder and customer email address

[1834] Output: Email sent to customer

[1835] Specific behavior: Calls SendGrid's API and sends the generated reply to the customer's email address (e.g., customer@example.com).

[1836] Step 6: Record your inquiries and automated responses

[1837] The server records the inquiry content, the automatic reply content, and the emotion recognition results in a database.

[1838] Input: Inquiry content, auto-reply content, emotion tag

[1839] Output: Records stored in the database

[1840] Specific behavior: Connects to a MySQL database and saves the inquiry content, response content, and emotion tag as records.

[1841] Step 7: Report to the person in charge

[1842] The server reports the recorded information and emotion recognition results to the person in charge, who receives the report through a communication application.

[1843] Input: Information stored in the database and emotion recognition results

[1844] Output: Report message sent to the agent

[1845] Specific behavior: Using the Slack API, a message is sent to the agent's chat room stating, "Customer A has made an inquiry and an automated reply has been sent with the following content. The customer's sentiment was recognized as 'confused'."

[1846] Step 8: Handle follow-up

[1847] The responsible person receives the report and manually follows up as needed.

[1848] Input: Report message received by the agent

[1849] Output: Result of additional customer actions

[1850] Specific actions: The person in charge contacts the delivery company to confirm the specific delivery date and time, and then uses that information to contact the customer again, saying, "Sorry for the wait. Your order is expected to be delivered the next business day."

[1851] (Application example 2)

[1852] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1853] Conventional food delivery services have issues with delayed responses to customer inquiries and low customer satisfaction due to uniform replies that do not consider customer feelings. In addition, it takes time and effort for staff to manually analyze the content of inquiries, understand the feelings of customers, and provide appropriate replies, which is inefficient.

[1854] The specific processing 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 receiving inquiries from customers, means for analyzing the content of the received inquiries, and means for generating an automatic reply based on the analysis results. This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's emotions, thereby improving customer satisfaction. In addition, by using an emotion engine to recognize the customer's emotions and adjusting the content of the automatic reply according to the recognized emotions, more appropriate and personalized responses can be provided.

[1855] "Customer" means any person who purchases or uses a product or service.

[1856] An "inquiry" is when a customer raises a question or concern about a product or service.

[1857] "Means for receiving" refers to the method or device by which the system collects customer inquiries.

[1858] "Means for analysis" refers to the method or device by which the system understands and analyzes the content of the query.

[1859] "Means for generating an automatic reply" refers to a method or device by which the system automatically generates a reply based on the analysis results.

[1860] "Means for sending" refers to a method or device for delivering the generated reply to the customer.

[1861] "Recording means" refers to a method or device for saving the contents of inquiries and automatic replies.

[1862] "Reporting means" means the method or device by which recorded information is communicated to responsible personnel.

[1863] The "emotion engine" is a system that recognizes customer emotions from the content of inquiries.

[1864] "Adjusting means" is a method or device for varying the content of the automated response based on perceived customer sentiment.

[1865] The present invention relates to a customer support system for food delivery services. This system uses an emotion engine to recognize customer emotions and generate automatic replies accordingly, in order to respond quickly and accurately to customer inquiries.

[1866] System Configuration

[1867] The system includes the following major components:

[1868] 1. Server

[1869] Receiving inquiries, analyzing them, and generating and sending automatic replies

[1870] 2. Database

[1871] Record of inquiries and automatic replies

[1872] 3. Emotion Engine

[1873] Customer Emotion Recognition

[1874] 4. Natural Language Processing (NLP) Libraries

[1875] Analysis of inquiry content

[1876] 5. Smartphone App

[1877] An interface for customers to make inquiries

[1878] System Operation Details

[1879] Hardware and Software Usage

[1880] The server (e.g., AWS EC2, Google Cloud Compute Engine) handles all the processes from receiving inquiries to sending automated replies. The server receives inquiries via a mail server or web form and analyzes them using a natural language processing (NLP) library (e.g., spaCy, NLTK).

[1881] An emotion engine (e.g., Google Cloud Natural Language API, AWS Comprehend) recognizes customer emotions from the content of inquiries and identifies emotions such as "confused" or "angry."

[1882] The database (e.g., MySQL, PostgreSQL) records the received inquiries, the generated auto-replies, and the emotion recognition results.

[1883] The smartphone app is developed using frameworks such as React Native and provides a user interface for customers to submit inquiries.

[1884] Details of data processing and calculation

[1885] 1. Receiving Inquiries

[1886] The server receives customer inquiries via email or web form, and the inquiries are stored in a database in text format.

[1887] 2. Analysis of inquiry content

[1888] The server uses an NLP library to analyze the inquiry and extract important keywords such as delivery delays and order cancellations.

[1889] 3. Emotion recognition

[1890] The server uses an emotion engine to recognize the customer's emotions from the analyzed inquiry, for example, identifying emotions such as "angry" or "confused."

[1891] 4. Generate Auto-Reply

[1892] The server generates an automatic reply based on the analysis results and emotion recognition results. For example, if a delivery is delayed and the customer is angry, the server generates a reply that includes an apology, such as "We apologize for the wait. Your delivery is currently on its way. We will deliver it to you soon."

[1893] 5. Sending Auto-Reply

[1894] The server notifies the customer's app of the generated auto-reply.

[1895] 6. Records

[1896] The server records the inquiry and the automatic reply in a database and shares it with the customer support team.

[1897] 7. Report to the person in charge

[1898] The server reports the automatically generated reply and customer sentiment to the agent, and if necessary, further follow-up is carried out.

[1899] Examples of concrete examples and prompts

[1900] Specific examples

[1901] Examples of receiving inquiries:

[1902] A customer sends an inquiry saying, "My pizza order hasn't arrived yet, what's going on?"

[1903] Examples of emotion recognition:

[1904] The emotion engine analyzes the content of the email and recognizes the emotion "anger."

[1905] Examples of generating and sending auto-replies:

[1906] Based on the keyword "delayed delivery" and the emotion "angry," the server generates an automatic reply saying, "We apologize for the wait. Your order is currently being delivered. We will deliver it to you soon." and sends it to the customer.

[1907] Prompt Sentence Examples

[1908] Prompts to recognize customer sentiment:

[1909] "Customer asks: 'My pizza order hasn't arrived yet, what's going on?' Recognize customer sentiment and generate appropriate automated responses."

[1910] This makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[1911] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1912] Step 1:

[1913] Receiving inquiries

[1914] Input: A customer submits an inquiry through a smartphone app.

[1915] Specific behavior: A customer uses the smartphone app's inquiry form or chat function to enter and send an inquiry in text format, such as, "The item I ordered hasn't arrived yet. What's going on?"

[1916] Data processing: The smartphone app generates an API request to send the inquiry to the server.

[1917] Output: The server receives the API request and saves the query in text format.

[1918] Step 2:

[1919] Analysis of inquiry content

[1920] Input: Text data of the query received by the server.

[1921] What it does: The server parses the query using a natural language processing (NLP) library (e.g., spaCy or NLTK) and extracts important keywords (e.g., "delivery" or "delay").

[1922] Data processing: Tokenize the text data, tag each token with a part of speech, and extract important keywords.

[1923] Output: A list of keywords extracted as a result of the analysis.

[1924] Step 3:

[1925] emotion recognition

[1926] Input: Text data of the inquiry and extracted keyword list.

[1927] Specific operation: Using an emotion engine (e.g., Google Cloud Natural Language API or AWS Comprehend), the server recognizes the customer's emotion from the inquiry content.

[1928] Data processing: The emotion engine analyzes the text data and calculates an emotion score. For example, it makes a judgment such as "The customer is angry" based on the emotion score.

[1929] Output: Analysis results including recognized emotion information (e.g., "angry").

[1930] Step 4:

[1931] Generate auto-replies

[1932] Input: Analysis results (keyword list and sentiment information).

[1933] What it does: The server generates an appropriate auto-reply based on the analysis results, choosing a basic reply from a template and adjusting the content based on the sentiment.

[1934] Data processing: Generate an automatic reply by combining template sentences and analysis results. For example, based on "Delivery delayed" and "Angry," generate a reply such as "Sorry for the wait. Delivery is currently in progress. We will deliver it to you soon."

[1935] Output: The generated auto-reply.

[1936] Step 5:

[1937] Sending automatic replies

[1938] Input: The generated auto-response.

[1939] Specific operation: Generates an API request to notify the customer's smartphone app of the automatic reply message generated by the server.

[1940] Data processing: Auto-reply messages are embedded into API requests and sent to the customer's app.

[1941] Output: The automated reply that will be displayed on the customer's smartphone app.

[1942] Step 6:

[1943] Record of inquiries and automatic replies

[1944] Input: Enquiry and generated auto-reply.

[1945] Specific operation: The server records the inquiry and the automatic reply in a database.

[1946] Data processing: Convert the text data into database format and save it along with the inquiry ID.

[1947] Output: A record of the inquiry and auto-reply stored in the database.

[1948] Step 7:

[1949] Report to the person in charge

[1950] Input: Inquiry details, auto-reply details and emotional information recorded in the database.

[1951] Specific operation: The server generates a notification to report to the agent and sends it to the agent's chat application.

[1952] Data processing: Compile inquiry details, automatic response details, and emotional information and convert them into a report format.

[1953] Output: Report notification sent to the responsible party.

[1954] This series of steps makes it possible to quickly and accurately grasp the content of the inquiry and the customer's feelings, thereby improving customer satisfaction.

[1955] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1956] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1957] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1958] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1959] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1960] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1961] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1962] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1963] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1964] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1965] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1966] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1967] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1968] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1969] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1970] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1971] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1972] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1973] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1974] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1975] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1976] The following is further disclosed regarding the above embodiment.

[1977] (Claim 1)

[1978] a means of receiving customer enquiries;

[1979] means for analyzing received inquiries;

[1980] means for generating an automated reply based on the analysis results;

[1981] A means for sending the generated auto-reply to the customer;

[1982] a means for recording the contents of inquiries and automatic replies;

[1983] a means for reporting the recorded information to a responsible person;

[1984] A system including:

[1985] (Claim 2)

[1986] 2. The system according to claim 1, further comprising means for selecting and customizing the content of the automatic reply from pre-prepared templates.

[1987] (Claim 3)

[1988] 10. The system of claim 1, further comprising means for the representative to receive the report through a chat application.

[1989] "Example 1"

[1990] (Claim 1)

[1991] a means of receiving customer enquiries;

[1992] A means for analyzing the content of the received inquiry using natural language processing technology;

[1993] a means for generating an automated reply using a generative AI model based on the analysis results;

[1994] A means for sending the generated auto-reply to the customer;

[1995] A means for recording the contents of inquiries and automatic replies in a database;

[1996] a means for reporting the recorded information to a person in charge using a chat application;

[1997] A system including:

[1998] (Claim 2)

[1999] 10. The system of claim 1, further comprising means for inputting prompt sentences to the generative AI model for generating content for the automatic reply.

[2000] (Claim 3)

[2001] 2. The system according to claim 1, further comprising means for the person in charge to receive a report of the content of the generated automatic reply through a chat application.

[2002] "Application Example 1"

[2003] (Claim 1)

[2004] a means of receiving customer enquiries;

[2005] A means for analyzing the content of the received inquiry using natural language processing technology;

[2006] A means for generating an automatic reply using a generative AI model based on the analysis results;

[2007] A means for sending the generated auto-reply to the customer;

[2008] A means for recording the contents of inquiries and automatic replies in a database;

[2009] A means for reporting the recorded information to personnel via a chat notification system;

[2010] A system that includes a means of interacting with customers through a smartphone application.

[2011] (Claim 2)

[2012] 10. The system of claim 1, further comprising means for selecting content of the automatic reply from pre-defined templates and customizing it using a generative AI model.

[2013] (Claim 3)

[2014] 10. The system of claim 1, further comprising means for a representative to receive reports in real time through a chat notification system.

[2015] "Example 2: Combining Emotion Engines"

[2016] (Claim 1)

[2017] a means of receiving customer enquiries;

[2018] A means for analyzing the content of the received inquiry using natural language processing technology;

[2019] means for generating an automatic reply based on the analysis results and the means for analyzing customer sentiment;

[2020] A means for sending the generated auto-reply to the customer;

[2021] A means for recording the contents of inquiries and automatic replies in a database;

[2022] a means for reporting the recorded information and the customer emotion recognition results to the agent;

[2023] A system including:

[2024] (Claim 2)

[2025] 10. The system of claim 1, further comprising means for using a generative AI model in generating the automated reply and for generating prompt sentences.

[2026] (Claim 3)

[2027] 10. The system of claim 1, further comprising means for a person to receive the report through a communication application.

[2028] "Application example 2 when combining emotion engines"

[2029] (Claim 1)

[2030] a means of receiving customer enquiries;

[2031] means for analyzing received inquiries;

[2032] means for generating an automated reply based on the analysis results;

[2033] A means for sending the generated auto-reply to the customer;

[2034] a means for recording the contents of inquiries and automatic replies;

[2035] a means for reporting the recorded information to a responsible person;

[2036] a means including an emotion engine for recognizing customer emotions from the content of an inquiry;

[2037] A means for tailoring the content of automated responses based on perceived customer sentiment; and

[2038] A system including:

[2039] (Claim 2)

[2040] 2. The system according to claim 1, further comprising means for selecting and customizing the content of the automatic reply from pre-prepared templates.

[2041] (Claim 3)

[2042] 10. The system of claim 1, further comprising means for a person to receive the report through a communication application. [Explanation of symbols]

[2043] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means of receiving customer enquiries; a means for analyzing the content of the received inquiry; means for generating an automated reply based on the analysis results; A means for sending the generated auto-reply to the customer; a means for recording the contents of inquiries and automatic replies; a means for reporting the recorded information to a responsible person; A system including:

2. 2. The system according to claim 1, further comprising means for selecting and customizing the content of the automatic reply from pre-prepared templates.

3. The system of claim 1 further comprising means for the representative to receive the report through a chat application.

Images