system

Abstract

To provide a system.SOLUTION: A system includes first means for receiving an inquiry from a user, second means for generating a first prompt sentence for outputting a first response to the inquiry based on the received inquiry from the user, third means for inputting the generated first prompt sentence into a generative AI model to generate the first response, fourth means for calculating a reliability score for the generated first response, and fifth means for displaying the generated first response and the calculated reliability score on a terminal of a person in charge.SELECTED DRAWING: Figure 1

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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] [[ID=2」

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a problem that there is no immediate response for a window staff or an operator to respond to a claim. As a result, the claim may go viral and the response time may be prolonged. In addition, in the case of responding based on one's own judgment or consultation with a supervisor, the mental burden is large, there are variations in response skills, and problems related to employment continuation also occur.

Means for Solving the Problems

[0005] The present invention is a claim response support system for window staff and operators, and provides means for providing an appropriate response to a claim, means for preventing further spread, means for shortening the response time. Furthermore, it provides means for providing a mental sense of security and means for realizing the leveling of response skills, and supports the solution of problems related to employment continuation. [Brief explanation of the drawing]

[0006] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 1 of Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment 2. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Form Example 2. [Figure 15]It is a sequence diagram showing the processing flow of the data processing system in Example 3 of Form Example 3. [Figure 16] It is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Form Example 3. [Figure 17] It is a sequence diagram showing the processing flow of the data processing system in Example 1 of Form Example 1 when combined with an emotion engine. [Figure 18] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1 when combined with an emotion engine. [Figure 19] It is a sequence diagram showing the processing flow of the data processing system in Example 2 of Form Example 2 when combined with an emotion engine. [[ID=...]] [Figure 20] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Form Example 2 when combined with an emotion engine. [Figure 21] It is a sequence diagram showing the processing flow of the data processing system in Example 3 of Form Example 3 when combined with an emotion engine. [Figure 22] It is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Form Example 3 when combined with an emotion engine.

Embodiments for Carrying Out the Invention

[0007] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained. [[ID=...]]

[0009] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)), etc.

[0010] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0011] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0012] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0014] [First Embodiment]

[0015] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0016] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0019] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0022] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0026] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0027] "Example of form 1"

[0028] One embodiment of the present invention is a complaint handling support system for counter staff and operators. This system incorporates an AI (artificial intelligence)-based automatic response generation module as a means of providing appropriate responses to complaints. This module analyzes the content of the complaint and generates an appropriate response. Furthermore, it includes a function to predict complaint escalation as a means of preventing further escalation. In addition, it employs an optimization algorithm to speed up response generation as a means of shortening response time. "Embodiment Example 2"

[0029] As an example of the present invention, there is a system that further includes means for providing a sense of security and means for achieving standardization of response skills. As means for providing a sense of security, the system has a function to display a confidence score for the response generated by the AI. This allows the operator to confirm the reliability of the response proposed by the AI ​​and gain a sense of security. Furthermore, as means for achieving standardization of response skills, the system allows operators to refer to the responses generated by the AI, thereby reducing the variation in response skills among operators.

[0030] "Example of form 3"

[0031] As a third embodiment of the present invention, there is a system that further includes means to support the resolution of the challenge of continued employment. Specifically, it has a function to evaluate the quality of responses generated by AI and to support the improvement of operators' skills. This helps to improve operators' skills and maintain their motivation, thereby supporting the resolution of the challenge of continued employment.

[0032] The following describes the processing flow for each example of the form.

[0033] "Example of form 1"

[0034] Step 1: The customer service representative or operator receives the complaint.

[0035] Step 2: The system analyzes the content of the claim.

[0036] Step 3: An AI (artificial intelligence) automated response generation module generates an appropriate response.

[0037] Step 4: The system predicts the escalation of a claim.

[0038] Step 5: An optimization algorithm is activated to speed up response generation and reduce response time.

[0039] "Example of form 2"

[0040] Step 1: The customer service representative or operator reviews the response generated by the AI.

[0041] Step 2: The system displays a confidence score for the AI-generated response.

[0042] Step 3: The customer service representative or operator decides on the final response based on the confidence score.

[0043] Step 4: Customer service staff and operators use AI-generated responses as a reference to standardize their response skills.

[0044] "Example of form 3"

[0045] Step 1: The customer service representative or operator uses the AI-generated response.

[0046] Step 2: The system evaluates the quality of the response generated by the AI.

[0047] Step 3: The system provides feedback to help operators improve their skills.

[0048] Step 4: Customer service staff and operators use feedback to improve their skills and maintain their motivation.

[0049] (Example 1)

[0050] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0051] Traditional complaint handling systems often lacked sufficient support for customer service representatives and operators to properly address complaints, resulting in inconsistencies in the quality and speed of responses. Furthermore, there was no way to predict the risk of complaint escalation, increasing the likelihood of escalation due to delays. Moreover, generating quick and appropriate responses was difficult amidst the pressure to shorten response times. A system is needed to address these challenges.

[0052] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0053] In this invention, the server includes means for providing appropriate responses to claims, means for preventing further escalation, means for shortening response times, means for analyzing the content of claims, means for generating automated responses, means for predicting escalation, and means for accelerating response generation. This improves the quality and speed of claim handling, reduces the risk of escalation, and enables the provision of quick and appropriate responses.

[0054] "Means of providing appropriate responses to complaints" refers to a function that generates and provides appropriate responses to complaints from users.

[0055] "Means to prevent further escalation of the issue" refers to functions designed to prevent complaints from escalating and the problem from spreading.

[0056] "Means for shortening response time" refers to functions that shorten response time by quickly generating and providing responses to complaints.

[0057] "Means for analyzing complaint content" refers to a function that analyzes the content of user complaints to identify the type of problem and the emotions involved.

[0058] "Means for generating automatic responses" refers to a function that automatically generates an appropriate response based on the analysis results.

[0059] "Means for predicting escalation" refers to functions for predicting the likelihood of a claim escalating and for assessing the associated risks.

[0060] "Means for speeding up response generation" refers to functions that optimize the response generation process and provide responses quickly.

[0061] This invention is a support system for customer service representatives and operators to appropriately handle customer complaints. The system includes multiple means for providing appropriate responses to complaints, preventing escalation, and shortening response times.

[0062] System Configuration

[0063] Hardware and software

[0064] Server: Analyzes claim details, generates automated responses, predicts escalations, and speeds up response generation.

[0065] Terminal: A device (such as a PC or smartphone) used by the user to enter a complaint and receive a response from the server.

[0066] Natural Language Processing (NLP) technology: Used to analyze the content of the claims. Specifically, Google® Cloud Natural Language API and IBM Watson® Natural Language Understanding® are used.

[0067] Generative AI Model: Used to generate automated responses. Specifically, it utilizes OpenAI's GPT-3® model.

[0068] Machine learning algorithms: Used to predict escalation. Specifically, scikit-learn® and TENSORFLOW® are used.

[0069] Cache technology and parallel processing: Used to speed up response generation.

[0070] System operation

[0071] 1. Entering a complaint

[0072] The user enters the details of the complaint from their device.

[0073] Example: The user enters "The product quality is poor."

[0074] 2. Analysis of the complaint content

[0075] The server receives the complaint details sent by the user.

[0076] The server uses the Google Cloud Natural Language API to parse the claim content.

[0077] Specifically, the text of the complaint is sent to the API for sentiment analysis and keyword extraction.

[0078] Example: Extract the keyword "quality problem" from the text "The product quality is poor" and determine that the emotion is "negative".

[0079] 3. Generating automated responses

[0080] The server uses a generative AI model (GPT-3) based on the analysis results to generate an appropriate response.

[0081] Specifically, the analysis results are input as prompts into the AI ​​model, which then generates response sentences.

[0082] Example: Enter the prompt "The user has submitted a complaint that the product quality is poor. Please generate an appropriate response." and retrieve the generated response "We apologize. Could you please tell us more about the product quality?".

[0083] 4. Escalation prediction

[0084] The server uses machine learning algorithms to predict the likelihood of a claim escalating.

[0085] Specifically, we will use scikit-learn to build a model that assesses the risk of escalation based on past claim data, and then calculate a risk score for the current claim.

[0086] Example: A high risk score is assigned to a complaint that states, "I have contacted them multiple times, but the issue remains unresolved."

[0087] 5. Providing a response

[0088] The server provides the generated response to the user.

[0089] Specifically, the generated response message is sent to the terminal and displayed on the user's screen.

[0090] Example: The user's device will display the response, "We apologize. Could you please tell us more about the quality of the product?"

[0091] 6. Speeding up response generation

[0092] The server uses caching technology and parallel processing to speed up response generation.

[0093] Specifically, past responses to the same type of claim are stored in a cache and reused. Furthermore, processing speed is improved by executing multiple claim processing tasks in parallel.

[0094] Example: If a response to a complaint such as "The product quality is poor" is stored in the cache, retrieve the response from the cache and provide it quickly.

[0095] Specific example

[0096] A user submits a complaint stating that "the product quality is poor."

[0097] The server sends the complaint details to the Google Cloud Natural Language API, which extracts the keywords "quality issues" and the sentiment "negative."

[0098] The server inputs the prompt "The user has entered a complaint that the product quality is poor. Please generate an appropriate response." into the generating AI model (GPT-3) and generates the response "We apologize. Could you please tell us more about the product quality?"

[0099] The server uses scikit-learn to calculate the escalation risk score and assigns a high risk score to the case.

[0100] The server sends the generated response to the user's terminal and displays it on the user's screen.

[0101] The server stores responses to the same type of claim in a cache to speed up response generation in subsequent instances.

[0102] The above describes the embodiments for carrying out this invention.

[0103] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0104] Step 1:

[0105] Entering a complaint

[0106] The user enters the details of the complaint from their device.

[0107] Input: The content of the complaint entered by the user (e.g., "The product quality is poor").

[0108] Output: The claim details are sent to the server.

[0109] Specific action: The user enters the complaint details into the input form on the device and presses the submit button.

[0110] Step 2:

[0111] Received the details of the complaint

[0112] The server receives the complaint details sent by the user.

[0113] Input: The content of the complaint submitted by the user.

[0114] Output: The claim details are saved on the server.

[0115] Specific operation: The server receives an HTTP request and saves the claim details to the database.

[0116] Step 3:

[0117] Analysis of complaint content

[0118] The server uses the Google Cloud Natural Language API to parse the claim content.

[0119] Input: Saved claim details.

[0120] Output: Analysis results (keywords, sentiment analysis results, etc.).

[0121] Specific operation: The server sends the claim details to the API and retrieves the analysis results returned by the API.

[0122] Step 4:

[0123] Generating an automated response

[0124] The server uses a generative AI model (GPT-3) based on the analysis results to generate an appropriate response.

[0125] Input: Analysis results.

[0126] Output: The generated response.

[0127] Specific operation: The server inputs the analysis results as prompts into the AI ​​model and retrieves the generated response.

[0128] Example: Enter the prompt "The user has submitted a complaint that the product quality is poor. Please generate an appropriate response." and generate the response "We apologize. Could you please tell us more about the product quality?"

[0129] Step 5:

[0130] Escalation prediction

[0131] The server uses machine learning algorithms to predict the likelihood of a claim escalating.

[0132] Input: Claim details and analysis results.

[0133] Output: Escalation risk score.

[0134] Specific operation: The server uses a model built on past claim data to calculate a risk score for the current claim.

[0135] Example: A high risk score is assigned to a complaint that states, "I have contacted them multiple times, but the issue remains unresolved."

[0136] Step 6:

[0137] Providing a response

[0138] The server provides the generated response to the user.

[0139] Input: The generated response.

[0140] Output: The response message is displayed on the user's terminal.

[0141] Specific operation: The server sends the generated response message to the user's terminal and displays it on the user's screen.

[0142] Example: The user's device will display the response, "We apologize. Could you please tell us more about the quality of the product?"

[0143] Step 7:

[0144] Speeding up response generation

[0145] The server uses caching technology and parallel processing to speed up response generation.

[0146] Input: Past complaint details and responses.

[0147] Output: Cached response statement.

[0148] Specific operation: The server caches and reuses past responses to the same type of claim. It also improves processing speed by executing multiple claim processing in parallel.

[0149] Example: If a response to a complaint such as "The product quality is poor" is stored in the cache, retrieve the response from the cache and provide it quickly.

[0150] (Application Example 1)

[0151] Next, we will describe Application Example 1 of Form 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."

[0152] Traditional customer complaint handling systems had several problems: long response times and decreased customer satisfaction because customer service representatives and operators handled complaints manually. Furthermore, it was difficult to predict complaint escalation, making it impossible to notify superiors at the appropriate time, increasing the risk of complaints escalating. In addition, inconsistencies in handling skills and significant mental stress among staff also created challenges in retaining employees.

[0153] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0154] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response times, means for analyzing complaint content and generating appropriate responses, means for predicting complaint escalation, means for automatically notifying senior personnel, means for generating responses using a generation AI model, and means for generating responses using prompt sentences. This enables more efficient and faster complaint handling, leading to improved customer satisfaction and reduced risk of complaint escalation. Furthermore, by standardizing response skills and providing a sense of security, it can also solve the challenge of retaining employees.

[0155] A "customer service representative" is someone whose job is to directly handle customer inquiries and complaints.

[0156] An "operator" is someone whose job is to operate systems or machines and perform specific tasks.

[0157] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[0158] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints.

[0159] "Means to prevent online firestorms" are functions designed to prevent complaints from escalating and problems from becoming serious.

[0160] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[0161] "Means for analyzing the content of a claim and generating an appropriate response" refers to a function that analyzes the content of a claim and generates an appropriate response based on that analysis.

[0162] "Methods for predicting complaint escalation" refer to functions that predict the likelihood of a complaint escalating to a higher-level employee.

[0163] The "automatic notification mechanism for senior personnel" is a function that automatically notifies senior personnel when a complaint escalates.

[0164] "Response generation means using a generative AI model" refers to a function that uses an artificial intelligence model to generate a response to a claim.

[0165] A "response generation means using a prompt statement" is a function for generating a response to a claim using a specific input statement (prompt statement).

[0166] As an example of how to implement this invention, we will describe a customer support application for an e-commerce site. This system can automatically generate appropriate responses to customer complaints and inquiries using AI and respond quickly. It also has a function to predict complaint escalation and automatically notify higher-level personnel as needed.

[0167] System Configuration

[0168] This system consists of the following main components:

[0169] 1. Claim Analysis Module: Analyzes the claim content and generates an appropriate response.

[0170] 2. Escalation prediction module: Predicts the escalation of claims.

[0171] 3. Notification Module: Automatically notifies senior personnel when an escalation is anticipated.

[0172] 4. Generative AI Model: An artificial intelligence model for generating responses to complaints.

[0173] 5. Prompt statement generation module: Generates a response using a specific input statement (prompt statement).

[0174] Hardware and software

[0175] This system uses an internet-connected smartphone or PC as hardware. The software uses Python® and the OpenAI API.

[0176] Processing flow

[0177] 1. Entering the complaint details: The user enters the details of the complaint.

[0178] 2. Response Generation: The claim analysis module analyzes the claim content and generates an appropriate response using a generation AI model.

[0179] 3. Escalation Prediction: The escalation prediction module predicts the likelihood of a claim escalating.

[0180] 4. Notifications: If an escalation is anticipated, the notification module will automatically notify the higher-level responsible party.

[0181] Specific example

[0182] For example, if a customer enters a complaint stating, "I did not receive the product, so I would like a refund," the system will operate as follows:

[0183] 1. Entering the complaint details: The user enters, "I did not receive the product, so I would like a refund."

[0184] 2. Response Generation: The complaint analysis module analyzes this content and uses a generation AI model to generate a response such as, "We apologize. We will immediately investigate the issue of the product not being delivered and proceed with the refund process."

[0185] 3. Escalation Prediction: The escalation prediction module predicts that this claim has a "high" likelihood of escalation.

[0186] 4. Notification: The notification module automatically notifies the senior person in charge, who then takes over the responsibility.

[0187] Example of a prompt

[0188] Please enter your complaint details: I did not receive the item and would like a refund.

[0189] In this way, customer support for e-commerce sites can be streamlined, and customer satisfaction can be improved.

[0190] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0191] Step 1:

[0192] The user enters the details of their complaint.

[0193] Input: The user enters, "I did not receive the product, so I would like a refund."

[0194] Specific action: The user enters the details of their complaint in text format into an input form on their smartphone or PC.

[0195] Step 2:

[0196] The server receives the claim details and sends them to the claim analysis module.

[0197] Input: The complaint details entered by the user.

[0198] Output: Claim details sent to the claim analysis module.

[0199] Specific operation: The server receives input from the user and sends its contents to the claims analysis module.

[0200] Step 3:

[0201] The claim analysis module analyzes the claim content and generates an appropriate response using a generative AI model.

[0202] Input: Claim details sent to the claim analysis module.

[0203] Output: The generated response text.

[0204] Specific operation: The claim analysis module analyzes the claim content using natural language processing technology and sends it as a prompt to the generative AI model. The generative AI model generates an appropriate response and returns it to the claim analysis module.

[0205] Step 4:

[0206] The server sends the generated response to the user.

[0207] Input: The generated response text.

[0208] Output: The response text displayed to the user.

[0209] Specific operation: The server receives the generated response and sends it to the user's terminal. The user's terminal displays the response.

[0210] Step 5:

[0211] The escalation prediction module predicts the likelihood of a claim escalating.

[0212] Input: Complaint details.

[0213] Output: Escalation prediction result (high / low).

[0214] Specific operation: The escalation prediction module analyzes the complaint content and uses a generative AI model to predict the likelihood of escalation. The prediction result is returned to the server.

[0215] Step 6:

[0216] The server receives the escalation prediction results and automatically notifies higher-level personnel as needed.

[0217] Input: Escalation prediction result.

[0218] Output: Notification to the superior in charge.

[0219] Specific operation: The server receives the escalation prediction result, and if the prediction result is "high," it sends a notification to the higher-level person in charge. The notification is sent via email or a dedicated application.

[0220] In this way, the system can streamline and expedite complaint handling, thereby improving customer satisfaction and reducing the risk of complaints escalating.

[0221] (Example 2)

[0222] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0223] Conventional complaint handling support systems lack sufficient support for operators to provide appropriate responses, and they lack means to provide a sense of security and standardize response skills. Furthermore, they lack effective means to shorten response times and prevent issues from escalating. As a result, operators face increased burdens and inconsistent response quality, which is a significant challenge.

[0224] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0225] In this invention, the server includes means for receiving inquiries, means for generating responses using a generative AI model, means for calculating a confidence score for the generated responses, means for displaying the responses and confidence scores, means for providing appropriate responses to complaints, means for preventing further escalation of issues, and means for shortening response times. This allows operators to verify the reliability of AI-generated responses and gain peace of mind, while also reducing variability in response skills, shortening response times, and preventing escalations of issues.

[0226] "Means for receiving inquiries" refers to the function that allows a server to receive inquiries from users.

[0227] "Means for generating responses using a generative AI model" refers to a function that uses a generative AI model to generate an appropriate response to an incoming inquiry.

[0228] "Means for calculating a confidence score for a generated response" refers to a function for evaluating the reliability of a generated response and calculating a confidence score.

[0229] "Means for displaying responses and confidence scores" refers to a function that visually displays the generated responses and their confidence scores to the user.

[0230] "Means of providing appropriate responses to complaints" refers to a function that provides responses in order to address complaints appropriately.

[0231] "Means to prevent further escalation of the issue" refers to a function that provides preventative measures to prevent problems from escalating during complaint handling.

[0232] "Means for shortening response time" refers to functions designed to improve efficiency in order to reduce the time spent handling complaints.

[0233] "Means of providing psychological reassurance" refers to a function that allows operators to gain psychological reassurance by verifying the reliability of the AI's responses.

[0234] "Means for achieving standardization of response skills" refers to functions that reduce variations in response skills among operators and achieve uniform responses.

[0235] "Means to support the resolution of employment continuity issues" refers to functions that provide support for resolving issues related to the continued employment of operators.

[0236] This invention provides operators with peace of mind and standardizes their response skills by generating responses using a generative AI model and displaying a confidence score in a claims handling support system. A specific embodiment of this system is described below.

[0237] Server Processing

[0238] The server receives a query from the user. The query is a prompt, such as "Please tell me the progress of the new project." The server inputs this query into a generative AI model (e.g., GPT-4®) and generates an appropriate response. For the generated response, the server calculates a confidence score. This confidence score is calculated using the AI ​​model's internal evaluation mechanism and additional evaluation algorithms. The server sends the generated response and the confidence score to the terminal.

[0239] Terminal processing

[0240] The terminal receives the response and confidence score sent from the server. The received response and confidence score are displayed on the terminal's screen. For example, it might say, "Current progress is on track. The next milestone is next week. (Confidence score: 85%)." The terminal uses visual aids to make it easy to understand, such as displaying a high confidence score in green and a low score in red.

[0241] User processing

[0242] The user checks the response and confidence score displayed on the device. They evaluate the reliability of the response based on the confidence score. For example, if the confidence score is 85%, the user can trust and accept the response. If the confidence score is low, the user can make another inquiry. When making a re-inquiry, they can add more specific questions based on the previous response.

[0243] Specific example

[0244] Specific example 1: A means of providing a sense of mental security

[0245] A user sends an inquiry asking, "Please tell me the progress of the new project." The server uses a generative AI model (e.g., GPT-4) to generate a response saying, "The current progress is on track. The next milestone is next week," and calculates a confidence score of 85%. The terminal displays this response and the confidence score, and the user feels reassured after checking the confidence score.

[0246] Specific example 2: A means to achieve standardization of response skills

[0247] Operator A receives an inquiry asking, "Could you please explain the procedure for handling customer complaints?" The server uses a generative AI model to generate a response saying, "First, listen carefully to the customer's story, then confirm the details of the problem, and finally, propose a solution." Operator A uses this response as a reference, and Operator B can use the same procedure as a reference for similar inquiries, thereby reducing variability in their response skills.

[0248] Example of a prompt

[0249] "Please tell me about the progress of the new project."

[0250] "Please explain the procedure for handling customer complaints."

[0251] By inputting these prompt sentences into a generating AI model, appropriate responses and confidence scores can be obtained.

[0252] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0253] Step 1:

[0254] The server receives inquiries from users. When a user enters a prompt message such as "Please tell me the progress of the new project" through their terminal and presses the send button, the server receives the inquiry. The input is the user's inquiry, and the output is the inquiry data stored on the server.

[0255] Step 2:

[0256] The server inputs the received query into the generative AI model. Specifically, the server sends the query content to the generative AI model (e.g., GPT-4) via an API. The input is the query data, and the output is the prompt message sent to the generative AI model.

[0257] Step 3:

[0258] The generative AI model generates a response based on a prompt. The generative AI model uses an internal natural language processing algorithm to generate an appropriate response. The input is the prompt, and the output is the generated response text.

[0259] Step 4:

[0260] The server calculates a confidence score for the generated response. The server uses the AI ​​model's internal evaluation mechanism and additional evaluation algorithms to assess the reliability of the response and calculate the confidence score. The input is the generated response text, and the output is the confidence score.

[0261] Step 5:

[0262] The server sends the generated response and confidence score to the terminal. The server combines the response text and confidence score into a single data packet and sends it to the terminal. The input is the response text and confidence score, and the output is the data packet sent to the terminal.

[0263] Step 6:

[0264] The terminal receives the response and confidence score sent from the server. The terminal analyzes the received data packets and extracts the response text and confidence score. The input is the data packet sent from the server, and the output is the analyzed response text and confidence score.

[0265] Step 7:

[0266] The terminal displays the received response and confidence score to the user. The terminal screen displays "Current progress is on track. The next milestone is next week. (Confidence score: 85%)." The input is the parsed response text and confidence score, and the output is the information displayed to the user.

[0267] Step 8:

[0268] The user checks the response and confidence score displayed on the device. The user then evaluates the reliability of the response based on the confidence score. The input is the displayed response text and confidence score, and the output is the user's evaluation result.

[0269] Step 9:

[0270] The user either accepts the response or makes a new inquiry. If the confidence score is high, the user trusts and accepts the response. If the confidence score is low, the user makes a new inquiry with more specific questions. The input is the user's evaluation result, and the output is the new inquiry or the acceptance of the response.

[0271] (Application Example 2)

[0272] Next, we will describe Application Example 2 of Form 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."

[0273] In the conventional claim response support system, when an operator provides an appropriate response to a claim, the mental burden is large, and variations in response skills often occur. In addition, when a security operator receives an AI proposal, there is a lack of means to confirm its reliability, making it difficult to respond quickly and appropriately. As a result, there is a problem that the response time is prolonged and the risk of a fire spreading increases.

[0274] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following respective means.

[0275] In this invention, the server includes means for providing an appropriate response to a claim, means for preventing further fire spread, means for shortening the response time, means for displaying a reliability score for a response generated by AI, means for reducing variations in response skills by allowing an operator to refer to the response generated by AI, and means for a security operator to receive an AI proposal and respond while confirming its reliability score. As a result, the operator can respond quickly and appropriately while obtaining a sense of mental security.

[0276] A "front-line staff member" is a person who has the duty of directly responding to inquiries and claims from customers and users.

[0277] An "operator" is a person who operates a system or machine and has the duty to perform specific tasks.

[0278] A "claim response support system" is a system for supporting appropriate responses to claims from customers and users.​​​​​​​​

[0281] The "means for shortening response time" is a function for reducing the time taken for claim handling.

[0282] The "means for displaying a reliability score for a response generated by AI" is a function for quantifying the reliability of a response generated by AI and displaying it to the operator.

[0283] The "means for reducing the variation in response skills by allowing the operator to refer to the response generated by AI" is a function for reducing the differences in response skills among operators by allowing the operator to refer to the response generated by AI.

[0284] A "security operator" is a person who is responsible for security-related work and has the duty of ensuring the security of systems and facilities.

[0285] A "reliability score" is a numerical value for evaluating the reliability of a response generated by AI.

[0286] As a form for implementing this invention, a claim handling support system used by security operators will be specifically described.

[0287] Configuration of the System

[0288] This system is composed of the following main components.

[0289] 1. Server: Hosts an AI model and generates responses to queries.

[0290] 2. Terminal: Devices such as smartphones and tablets used by security operators.

[0291] 3. User Interface: An application for the operator to input queries and view the responses and reliability scores of AI.

[0292] Hardware and Software to be Used

[0293] Hardware: Smartphones, Tablets, Servers

[0294] Software: Flask (registered trademark) (a web framework for Python), requests (a library for sending HTTP requests), Generative AI Model

[0295] Data Processing and Data Calculation

[0296] The server receives a query from the security operator and sends a prompt to the generative AI model. The generative AI model generates a response to the query and returns the response to the server. The server calculates the confidence score of the response and sends it to the operator's terminal.

[0297] Process Flow

[0298] 1. Query Input: The security operator inputs a query through the user interface of the terminal.

[0299] 2. Query Transmission: The terminal sends the query to the server.

[0300] 3. Response Generation: The server sends the query to the generative AI model and generates a response.

[0301] 4. Confidence Score Calculation: The server calculates the confidence score of the generated response.

[0302] 5. Response and Score Display: The server sends the response and the confidence score to the terminal so that the operator can view them.

[0303] Specific Example

[0304] For example, a security operator might enter a query such as, "What should I do if an intruder enters the building?" This query is sent to a server, and a generative AI model generates an appropriate response. The server calculates a confidence score for the generated response and displays the response and score on the operator's terminal.

[0305] Example of a prompt

[0306] "Please tell me how to respond if a suspicious person enters a building."

[0307] In this way, security operators can take quick and appropriate action by referring to AI suggestions. This allows them to gain peace of mind while reducing variability in response skills.

[0308] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0309] Step 1:

[0310] The security operator enters the query through the terminal's user interface.

[0311] Input: Query entered by the security operator (e.g., "Please tell me how to respond if an intruder enters the building.")

[0312] Output: The query is entered into the terminal.

[0313] Specific actions: The security operator opens the application on their smartphone or tablet, enters the query in the text box, and presses the "Submit" button.

[0314] Step 2:

[0315] The terminal sends a query to the server.

[0316] Input: Query entered on the terminal

[0317] Output: The query is sent to the server.

[0318] Specific operation: The application on the terminal sends the entered query to the server as an HTTP request.

[0319] Step 3:

[0320] The server sends queries to the generated AI model and generates responses.

[0321] Input: Query sent to the server

[0322] Output: Response from the generative AI model

[0323] Specific operation: The server sends the received query as a prompt to the generating AI model, and the generating AI model generates a response to the query and returns it to the server.

[0324] Step 4:

[0325] The server calculates a confidence score for the response it generates.

[0326] Input: Response from a generative AI model

[0327] Output: Confidence score

[0328] Specific operation: The server analyzes the content of the generated response and applies an algorithm to calculate a confidence score.

[0329] Step 5:

[0330] The server sends the response and confidence score to the terminal.

[0331] Input: Responses and confidence scores from the generated AI model

[0332] Output: Response and confidence score sent to the terminal

[0333] Specific operation: The server sends the response and confidence score together as an HTTP response to the terminal.

[0334] Step 6:

[0335] The terminal displays the response and confidence score to the security operator.

[0336] Input: Response sent from the server and confidence score

[0337] Output: Response and confidence score displayed to the security operator

[0338] Specific operation: The terminal application displays the received response and confidence score on the screen for the security operator to review.

[0339] (Example 3)

[0340] Next, we will describe Embodiment 3 of Embodiment Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0341] Conventional complaint handling support systems have been insufficient in providing adequate support for customer service staff and operators to respond appropriately, making it difficult to shorten response times, provide a sense of security, and standardize response skills. Furthermore, it has been difficult to improve operators' skills and maintain their motivation, posing challenges to continued employment.

[0342] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0343] In this invention, the server includes means for generating responses to prompt sentences using a generative AI model, means for evaluating the quality of the generated responses, and means for providing the evaluation results as feedback. This enables operators to respond appropriately and quickly, improving their skills and maintaining motivation. It also provides a sense of security, standardizes response skills, and solves the challenge of retaining employees.

[0344] A "customer service representative" is an employee whose role is to directly handle customer inquiries and complaints.

[0345] An "operator" is a person responsible for operating the system and handling customer inquiries and complaints.

[0346] A "customer complaint handling support system" is a system that provides support for appropriately responding to customer complaints.

[0347] A "generative AI model" is a model that uses artificial intelligence to generate responses to input prompt sentences.

[0348] A "prompt message" is text, such as a question or instruction, that a user enters into the system.

[0349] A "response" is the text of an answer or suggestion that a generative AI model generates based on a prompt.

[0350] A "quality evaluation algorithm" is an algorithm used to evaluate the quality of a generated response.

[0351] "Feedback" refers to information provided to operators based on evaluation results, intended to encourage skill development and improvement.

[0352] "Skill improvement" refers to an operator improving their own abilities and skills.

[0353] "Maintaining motivation" refers to operators maintaining their enthusiasm and motivation for their work.

[0354] "Mental security" refers to the sense of stability and security that operators feel while performing their work.

[0355] "Standardizing response skills" refers to ensuring that multiple operators possess the same level of response skills.

[0356] "Continued employment" refers to an operator continuing to work at the same workplace for an extended period.

[0357] This invention is a support system for counter staff and operators to appropriately handle customer complaints. This system uses a generative AI model to generate responses to prompt sentences, evaluates the quality of those responses, and provides the evaluation results as feedback, thereby aiming to improve operators' skills and maintain their motivation.

[0358] System Configuration

[0359] hardware

[0360] Server: Provides computing resources for running the generated AI model and quality evaluation algorithm.

[0361] Terminal: A device used by the user to input prompts and display responses and evaluation results from the server.

[0362] software

[0363] Generative AI model: For example, use a natural language generation model such as GPT-4.

[0364] Quality evaluation algorithm: For example, use a BERT®-based evaluation model.

[0365] Program processing

[0366] The server receives a prompt message entered by the user from the terminal. Next, it generates a response to the prompt message using a generative AI model. The generated response is evaluated by a quality evaluation algorithm, and the evaluation result is sent to the terminal as feedback. The terminal displays the evaluation result and the generated response to the user.

[0367] Specific example

[0368] For example, if a user enters the prompt "Please tell me the appropriate way to handle customer complaints," the server will process it as follows:

[0369] 1. The server receives the prompt message "Please tell me the appropriate way to handle customer complaints."

[0370] 2. Using a generative AI model (e.g., GPT-4), generate a response like this: "First, it's important to listen to the customer completely and show empathy. Then, you need to propose concrete solutions and respond quickly."

[0371] 3. The server evaluates this response using a quality evaluation algorithm (for example, a BERT-based evaluation model) and determines, for example, that "the response quality is high."

[0372] 4. Send these evaluation results to the operator as feedback to support their skill development.

[0373] Other examples of prompt statements include the following:

[0374] "Please briefly explain the features of the new product."

[0375] "Please tell me the appropriate way to handle customer complaints."

[0376] "Please suggest ways to boost team motivation."

[0377] This system allows operators to objectively evaluate the quality of their responses, thereby promoting skill development and maintaining motivation, and supporting the resolution of employment retention challenges. The flow of a specific process in Example 3 will be explained using Figure 15.

[0378] Step 1:

[0379] The user enters a prompt message.

[0380] The user enters prompt text, such as a question or instruction, into the terminal's input field. For example, they might enter, "Please tell me the appropriate way to handle a customer complaint." The entered prompt text is temporarily stored in the terminal's memory.

[0381] Step 2:

[0382] The terminal sends a prompt message to the server.

[0383] The terminal sends the prompt text entered by the user to the server. The prompt text is sent to the server in an appropriate format (e.g., JSON). The input is the prompt text, and the output is the data sent to the server.

[0384] Step 3:

[0385] The server generates a response using a generated AI model.

[0386] The server inputs the received prompt into a generating AI model (e.g., GPT-4) to generate a response. The input is the prompt, and the output is the generated response. For example, a response like, "First, it is important to listen to the customer completely and show empathy. After that, you are required to propose concrete solutions and respond quickly," might be generated.

[0387] Step 4:

[0388] The response generated by the server is evaluated using a quality evaluation algorithm.

[0389] The server inputs the generated response into a quality evaluation algorithm (e.g., a BERT-based evaluation model) to evaluate the quality of the response. The input is the generated response, and the output is the evaluation result. For example, a score or comment such as "Response quality is high" might be output.

[0390] Step 5:

[0391] The server sends the evaluation results to the terminal.

[0392] The server sends the evaluation results to the terminal. The evaluation results are sent along with the response. The input consists of the evaluation results and the generated response, while the output is the data sent to the terminal.

[0393] Step 6:

[0394] The device displays the evaluation results to the user.

[0395] The terminal displays the evaluation results received from the server and the generated responses to the user. The input is the evaluation results and generated responses, and the output is the data displayed to the user. The user can use the displayed evaluation results as a reference to improve their own skills.

[0396] (Application Example 3)

[0397] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0398] Conventional customer complaint support systems fail to adequately address operator skill development and motivation maintenance, leaving the challenge of continued employment unresolved. Furthermore, the lack of a means to evaluate the quality of AI-generated responses results in inconsistent operator response quality. Additionally, the system fails to consider the convenience of being a smartphone application.

[0399] In Application Example 3, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of the issue, means for shortening response times, means for evaluating the quality of responses generated by the AI, means for supporting operator skill improvement, means for maintaining motivation, and means for supporting the resolution of employment retention issues. This enables operator skill improvement and motivation maintenance, and resolves employment retention issues. Furthermore, by evaluating the quality of responses generated by the AI, the quality of responses is made more uniform. In addition, the convenience of being an application installed on a smartphone is also provided.

[0400] A "customer service representative" is an employee whose job is to directly handle customer inquiries and complaints.

[0401] An "operator" is an employee whose job is to handle customer inquiries using communication methods such as telephone or chat.

[0402] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[0403] "Means of providing appropriate responses" refers to the function of generating and providing appropriate answers to customer complaints.

[0404] "Means to prevent online backlash" refer to functions that prevent problems from escalating during complaint handling.

[0405] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[0406] "Means for evaluating the quality of AI-generated responses" refers to a function for evaluating the quality of responses generated by artificial intelligence.

[0407] "Means to support operator skill development" refers to functions designed to improve operators' response skills.

[0408] "Means for maintaining motivation" refers to functions that maintain the operator's enthusiasm for their work.

[0409] "Means to support the resolution of employment continuity challenges" refers to functions that address the challenges of maintaining the employment of operators.

[0410] "Means of providing psychological reassurance" refers to functions that provide psychological reassurance to operators.

[0411] "Means for achieving standardization of response skills" refers to functions for standardizing the response skills of operators.

[0412] An "application installed on a smartphone" is software that is installed on a smartphone and used by the user.

[0413] The system for carrying out this invention is configured as a claims handling support system. The system includes a server, a terminal used by the operator (such as a smartphone), and a generating AI model.

[0414] The server includes means of providing appropriate responses to complaints, preventing online crises, shortening response times, evaluating the quality of AI-generated responses, supporting operator skill development, maintaining motivation, and addressing challenges related to continued employment.

[0415] The terminals used by operators include an application installed on their smartphones. This application has the function of evaluating the quality of AI-generated responses in real time and providing feedback when operators handle customer complaints.

[0416] Specifically, when an operator handles a complaint, they input both the user's input and the response generated by the AI ​​into the application. The server uses a generative AI model (e.g., GPT-3) to evaluate the quality of the AI-generated response. The evaluation results are displayed as feedback on the operator's terminal. Based on this feedback, the operator can improve their skills.

[0417] The hardware used includes servers and operator smartphones. The software used includes Python, the OpenAI API, and smartphone applications.

[0418] As a concrete example, the following prompt statements can be used.

[0419] Example of a prompt:

[0420] User input: How do I return an item?

[0421] AI response: To return an item, first log in to your account and select the item you wish to return from your order history. Then, select the reason for the return and complete the return process.

[0422] Please rate the quality of this response.

[0423] Using this prompt, the server evaluates the quality of the response using a generative AI model and provides feedback to the operator. This allows the operator to improve their skills, maintain motivation, and address challenges in retaining employment.

[0424] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[0425] Step 1:

[0426] When a user handles a complaint, they input both their own information and the AI-generated response into the terminal. The input data consists of the user's complaint and the AI's response text.

[0427] Step 2:

[0428] The terminal sends the user's entered complaint details and the AI's response text to the server. The transmitted data consists of the user's input text and the AI's response text.

[0429] Step 3:

[0430] The server sends the received user input text and the AI's response text as prompts to the generating AI model. An example of a prompt is: "User input: How do I return an item? AI response: To return an item, first log in to your account and select the item you wish to return from your order history. Then, select a reason for return and complete the return process. Please rate the quality of this response."

[0431] Step 4:

[0432] The generative AI model evaluates the quality of the AI's response based on the prompt text. The evaluation result is feedback text regarding the quality of the response.

[0433] Step 5:

[0434] The server sends the evaluation results received from the generated AI model to the terminal. The transmitted data is feedback text regarding the quality of the response.

[0435] Step 6:

[0436] The terminal displays the received evaluation results to the user. The displayed data is feedback text regarding the quality of the response. The user can use this feedback to improve their skills.

[0437] Through the above processing steps, users can evaluate the quality of AI-generated responses during complaint handling in real time and receive feedback. This enables users to improve their skills and maintain their motivation, thus solving the challenge of retaining employees.

[0438] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0439] "Example of form 1"

[0440] One embodiment of the present invention provides a complaint handling support system that incorporates an emotion engine. When a customer service representative or operator receives a complaint, this system uses an emotion engine to recognize the user's emotions and analyze the user's emotional state. Based on the analysis results, the system generates an appropriate response to prevent further escalation. Specifically, if the system senses that the user is angry or dissatisfied, it generates a response that offers an apology or a solution. If the system senses that the user is confused, it generates a response that provides a clear explanation or guidance. As a result, customer service representatives and operators can quickly provide the most appropriate response according to the user's emotional state, shortening response times and improving service quality.

[0441] "Example of form 2"

[0442] Another embodiment of the present invention provides a complaint handling support system that incorporates an emotion engine. When a customer service representative or operator receives a complaint, this system uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state. Based on the analysis results, the system generates an appropriate response to prevent further escalation of the complaint. Furthermore, this system provides customer service representatives and operators with a sense of security and standardizes their response skills. Specifically, the system provides the optimal response according to the user's emotional state, enabling customer service representatives and operators to respond with confidence. In addition, the responses provided by the system eliminate differences in response skills among customer service representatives and operators, maintaining consistency in service quality.

[0443] "Example of form 3"

[0444] Furthermore, as another embodiment of the present invention, a complaint handling support system combined with an emotion engine is provided. When a customer service representative or operator receives a complaint, this system uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state. Based on the analysis results, the system generates an appropriate response to prevent further escalation. In addition, this system provides a sense of security to customer service representatives and operators and helps to standardize their response skills. Moreover, this system helps to solve the challenge of retaining employees. Specifically, the responses and feedback provided by the system help to improve the skills and maintain the motivation of customer service representatives and operators. This reduces the turnover rate of customer service representatives and operators and promotes continued employment.

[0445] The following describes the processing flow for each example of the form.

[0446] "Example of form 1"

[0447] Step 1: A customer service representative or operator receives the complaint.

[0448] Step 2: The system uses an emotion engine to analyze the user's emotional state.

[0449] Step 3: Based on the analysis results, the system generates an appropriate response.

[0450] Step 4: The customer service representative or operator uses the generated response to assist the user.

[0451] "Example of form 2"

[0452] Step 1: The customer service representative or operator receives the complaint.

[0453] Step 2: The system uses an emotion engine to analyze the user's emotional state.

[0454] Step 3: Based on the analysis results, the system generates an appropriate response.

[0455] Step 4: The customer service representative or operator uses the generated response to assist the user.

[0456] Step 5: Customer service staff and operators use the feedback received from the system to improve their customer service skills.

[0457] "Example of form 3"

[0458] Step 1: The customer service representative or operator receives the complaint.

[0459] Step 2: The system uses an emotion engine to analyze the user's emotional state.

[0460] Step 3: Based on the analysis results, the system generates an appropriate response.

[0461] Step 4: The customer service representative or operator uses the generated response to assist the user.

[0462] Step 5: Customer service staff and operators use the feedback received from the system to improve their customer service skills.

[0463] Step 6: Customer service staff and operators use feedback from the system to maintain their motivation and reduce employee turnover.

[0464] (Example 1)

[0465] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0466] In handling customer complaints, it is difficult for customer service representatives and operators to provide appropriate responses quickly. Furthermore, there is a high risk of complaints escalating and becoming a major issue, and prolonged handling times lead to decreased customer satisfaction. Additionally, there is variability in the handling skills of individual representatives, resulting in significant mental strain. To address these challenges, a system is needed that accurately analyzes the content of complaints, understands the user's emotional state, and quickly generates appropriate responses.

[0467] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0468] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of the issue, means for shortening response times, means for analyzing the content of the complaint, means for analyzing the user's emotional state, means for generating an appropriate response, and means for displaying the generated response. This enables improved efficiency and quality in complaint handling.

[0469] A "complaint handling support system" is a system designed to provide appropriate responses to complaints, prevent them from escalating, and shorten response times.

[0470] "Means of providing appropriate responses" refers to a function that analyzes the content of a complaint and generates the optimal response based on the user's emotional state.

[0471] "Means to prevent online outrage" refer to functions designed to prevent complaints from escalating and becoming major problems.

[0472] "Means of shortening response times" refers to a function that quickly generates responses to complaints, enabling customer service staff and operators to respond promptly.

[0473] "Means for analyzing the content of a claim" refers to a function that analyzes the text data of a claim using natural language processing technology to understand the content of the claim.

[0474] "Means for analyzing the user's emotional state" refers to a function that extracts emotions from the user's complaint text and identifies emotional states such as anger, dissatisfaction, and confusion.

[0475] "Means for generating appropriate responses" refers to a function that generates the optimal response based on the user's emotional state, using analysis results.

[0476] "Means for displaying generated responses" refers to a function that sends the response generated from the server to a terminal and displays it so that the counter staff or operator can confirm it.

[0477] This invention relates to a complaint handling support system for customer service staff and operators. This system aims to provide appropriate responses to complaints, prevent escalation, and shorten response times. Specific embodiments of this system are described below.

[0478] Hardware and software to be used

[0479] server

[0480] The server plays a central role in the claims handling support system. The server uses the following software and hardware:

[0481] Hardware: General cloud servers (e.g., AWS® EC2 instances)

[0482] software:

[0483] Automatic response generation module (e.g., AI model generation)

[0484] Emotion engine (e.g., natural language processing engine)

[0485] Database (e.g., SQL database)

[0486] terminal

[0487] Terminals are devices used by counter staff and operators. These terminals utilize the following hardware and software:

[0488] Hardware: General personal computers (e.g., Windows® PC, iPad®)

[0489] software:

[0490] Claims input interface

[0491] Response display interface

[0492] Program processing

[0493] Entering and submitting a claim

[0494] When a user enters a complaint, the device sends this information to the server. For example, if a user enters "The product hasn't arrived," the device packages this text data in JSON format and sends it to the server using the HTTPS protocol.

[0495] Claim Analysis

[0496] The server analyzes the text data of the received complaint. Specifically, it uses a generative AI model to perform natural language processing (NLP) and understand the content of the complaint. For example, in response to a complaint that "the product did not arrive," the server classifies it as a "delivery-related problem."

[0497] Analysis of emotional states

[0498] The server uses an emotion engine to analyze the user's emotional state. It extracts emotions such as anger, frustration, and confusion from the text of the complaint. For example, in response to a complaint that "the product did not arrive," the server identifies the emotion of "frustration."

[0499] Response generation

[0500] The server generates an appropriate response based on the analysis results. Using a generative AI model, it generates responses that correspond to the user's emotional state. For example, it might generate a response such as, "We apologize. We will confirm your order. Please wait a moment."

[0501] Sending and displaying responses

[0502] The server sends the generated response to the terminal. The terminal displays the response received from the server on its screen. The customer service representative or operator reviews this response and takes appropriate action for the user.

[0503] Specific example

[0504] For example, if a user enters a complaint stating "the product hasn't arrived," the terminal sends this information to the server. The server analyzes the complaint using a generative AI model and identifies the user's emotional state using an emotion engine. Based on the analysis, the server generates a response such as "We apologize. We will check on your order. Please wait a moment," and sends it to the terminal. The customer service representative then provides this response to the user and works to resolve the issue.

[0505] Example of a prompt

[0506] "Please explain how the server generates a response when a user submits a complaint stating that 'the product has not arrived.'"

[0507] In this way, this invention improves the efficiency of claim handling and enhances quality.

[0508] The flow of the specific processing in Example 1 will be explained using Figure 17.

[0509] Step 1:

[0510] The user enters a complaint.

[0511] The user enters a complaint through the interface of the complaint handling support system. For example, they might enter "The product did not arrive." The entered complaint is sent to the terminal as text data. The input data is text containing the content of the complaint.

[0512] Step 2:

[0513] The device sends the claim to the server.

[0514] The terminal sends user-entered claims to the server in real time. Specifically, it packages the claim text data in JSON format and sends it to the server using the HTTPS protocol. The input data is the claim text data, and the output data is the JSON format data sent to the server.

[0515] Step 3:

[0516] The server analyzes the content of the claim.

[0517] The server analyzes the text data of the received complaint. Specifically, it uses a generative AI model to perform natural language processing (NLP) and understand the content of the complaint. For example, in response to a complaint that "the product did not arrive," the server classifies it as a "delivery-related problem." The input data is the text data of the complaint, and the output data is the classification information of the complaint as a result of the analysis.

[0518] Step 4:

[0519] The server analyzes the user's emotional state.

[0520] The server uses an emotion engine to analyze the user's emotional state. It extracts emotions such as anger, frustration, and confusion from the text of the complaint. For example, in response to a complaint that "the product did not arrive," the server identifies the emotion "frustration." The input data is the text data of the complaint, and the output data is the result of the emotion analysis.

[0521] Step 5:

[0522] The server generates an appropriate response.

[0523] The server generates an appropriate response based on the analysis results. Using a generative AI model, it generates responses that correspond to the user's emotional state. For example, it might generate a response such as, "We apologize. We will confirm your order. Please wait a moment." The input data consists of complaint classification information and sentiment analysis results, while the output data is the generated response text.

[0524] Step 6:

[0525] The server sends a response to the terminal.

[0526] The server sends the generated response to the terminal. Specifically, it packages the response text data in JSON format and sends it to the terminal using the HTTPS protocol. The input data is the generated response text, and the output data is the JSON data sent to the terminal.

[0527] Step 7:

[0528] The device displays a response.

[0529] The terminal displays the response received from the server on the screen. The customer service representative or operator reviews this response and takes appropriate action for the user. The input data is the response text sent from the server, and the output data is the response displayed on the screen.

[0530] Step 8:

[0531] The user receives a response.

[0532] The user receives a response from a customer service representative or operator. This allows the user to know the next steps toward resolving the problem. The input data is the response provided by the customer service representative or operator, and the output data is the response received by the user.

[0533] (Application Example 1)

[0534] Next, we will describe Application Example 1 of Form 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."

[0535] Traditional complaint handling systems faced challenges in accurately understanding customers' emotional states and providing prompt and appropriate responses. Furthermore, the inability to predict complaint escalations and take preventative measures led to delays and a risk of decreased customer satisfaction. Additionally, insufficient management of response history meant that past records could not be used to improve future responses.

[0536] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response time, means for analyzing the user's emotional state using an emotion engine, means for generating an appropriate response based on the analysis results, means for predicting the possibility of escalation, and means for managing the complaint handling history. This makes it possible to appropriately grasp the customer's emotional state and respond quickly and appropriately. Furthermore, since complaint escalation can be predicted and countermeasures can be taken in advance, it is expected that customer satisfaction will improve without delays in response. In addition, the management of response history is enhanced, and past response history can be utilized to help with future responses.

[0537] A "customer service representative" is an employee whose role is to directly receive and handle customer inquiries and complaints.

[0538] An "operator" is an employee whose role is to operate systems or machines and perform specific tasks.

[0539] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[0540] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints to customers.

[0541] "Means to prevent online firestorms" are functions designed to prevent complaints from escalating and problems from becoming serious.

[0542] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[0543] An "emotion engine" is a technology used to analyze a user's emotional state.

[0544] "Means for analyzing the user's emotional state" refers to a function that uses an emotion engine to analyze the user's emotions.

[0545] "Means for generating appropriate responses" refers to a function that generates appropriate responses based on the results of emotion analysis.

[0546] "Means for predicting the possibility of escalation" refers to a function that predicts the likelihood of a problem becoming more serious based on the content of the complaint and the emotional state of the person making it.

[0547] "Means for managing complaint handling history" refers to a function for recording and managing past complaint handling history.

[0548] As an example of how to implement this invention, a customer complaint handling support system for an e-commerce site will be used. This system operates through the cooperation of three parties: a server, a terminal, and a user.

[0549] System Configuration

[0550] 1. Server:

[0551] Emotion Engine: This is a software module for analyzing a user's emotional state. Specifically, it uses Python sentiment analysis libraries (e.g., TextBlob® or VADER®).

[0552] Generative AI Model: This is an AI model designed to generate appropriate responses to claims. Specifically, it uses generative AI models such as GPT-3.

[0553] Escalation Prediction Module: This is a machine learning model for predicting the likelihood of escalation based on the content and emotional state of a complaint. Specifically, it uses machine learning libraries such as scikit-learn.

[0554] History Management System: This is a database system for managing the history of customer complaints. Specifically, it uses database management systems such as MySQL (registered trademark) or PostgreSQL (registered trademark).

[0555] 2. Terminal:

[0556] Smartphone app: This is an application for users to input complaints and communicate with a server. The application provides a user interface for inputting complaints, displaying sentiment analysis results, and receiving responses.

[0557] 3. User:

[0558] Customer: Responsible for entering complaints and receiving responses from the system.

[0559] System operation

[0560] 1. Complaint Reception:

[0561] Users submit complaints through a smartphone app. For example, they might enter, "I am very dissatisfied because the product has not arrived!"

[0562] 2. Emotion analysis:

[0563] The server's emotion engine analyzes the complaint text to determine the user's emotional state. For example, it might detect "anger" from the submitted complaint.

[0564] 3. Automated response generation:

[0565] The server's AI model generates an appropriate response based on the sentiment analysis results. For example, it might generate a response such as, "We apologize for the inconvenience. We will investigate immediately and provide you with a solution."

[0566] 4. Escalation prediction:

[0567] The server's escalation prediction module predicts the risk of escalation based on the content and emotional state of the complaint. For example, it might determine it to be "high risk."

[0568] 5. History management:

[0569] The server's history management system stores complaints and their response history in a database. This allows for future support and management.

[0570] Specific examples and prompt statements

[0571] Specific example:

[0572] Complaint: "I am extremely dissatisfied because the product has not arrived!"

[0573] Emotion analysis result: Anger

[0574] Automated response: "We apologize for the inconvenience. We will investigate immediately and provide you with a solution."

[0575] Escalation risk: High

[0576] Example of a prompt:

[0577] Customer complaint: "I am extremely dissatisfied because the product has not arrived!"

[0578] Emotion analysis result: Anger

[0579] Please generate an appropriate response.

[0580] In this way, the server, terminal, and user work together to accurately understand the customer's emotional state and respond quickly and appropriately. Furthermore, it becomes possible to predict complaint escalations and take preventative measures, thus preventing delays in responses and improving customer satisfaction. In addition, the management of response history is enhanced, allowing past response records to be used to improve future responses.

[0581] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[0582] Step 1:

[0583] A user submits a complaint through a smartphone app. The entered complaint text is sent to the server. For example, a complaint might read, "I am very dissatisfied because the product has not arrived!"

[0584] Step 2:

[0585] The server's sentiment engine receives the claim text and performs sentiment analysis. It analyzes the input claim text and determines the user's emotional state. For example, it uses a sentiment analysis library (e.g., TextBlob or VADER) to detect the emotion "anger." The output is the emotional state (e.g., anger).

[0586] Step 3:

[0587] The server's generative AI model receives the sentiment analysis results and generates an appropriate response. Taking the emotional state (e.g., anger) and complaint text as input, the generative AI model (e.g., GPT-3) is given a prompt to generate a response. For example, a response such as "We apologize for the inconvenience. We will investigate immediately and provide you with a solution" might be generated. The output is the generated response.

[0588] Step 4:

[0589] The server's escalation prediction module receives the claim text and sentiment state and predicts the risk of escalation. Using the claim text and sentiment state as input, a machine learning model (e.g., scikit-learn) is used to predict the escalation risk. For example, it might be judged as "high risk." The output is the escalation risk.

[0590] Step 5:

[0591] The server's history management system stores claims and their response history in a database. Claim text, emotional state, generated response, and escalation risk are taken as input and stored in the database management system (e.g., MySQL or PostgreSQL). The output is the stored history data.

[0592] Step 6:

[0593] The server sends the generated response to the terminal. The generated response is then sent to the user's smartphone app as input. For example, the user might see a response such as, "We apologize for the inconvenience. We will investigate immediately and provide you with a solution." The output is the response displayed to the user.

[0594] Step 7:

[0595] The user reviews the response via a smartphone app and enters any additional complaints or feedback as needed. If the user reviews the response and is satisfied, the complaint handling is complete. If there are additional complaints or feedback, the process is repeated from step 1. The output includes the user's satisfaction level and any additional complaints.

[0596] (Example 2)

[0597] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0598] Traditional complaint handling systems make it difficult for customer service representatives and operators to provide appropriate responses to customer inquiries, resulting in significant mental strain. Furthermore, inconsistencies in response skills and a lack of consistency in service quality are major challenges. Additionally, while complaint handling requires accurately understanding customer emotions and preventing escalation, there is a lack of effective means to achieve this.

[0599] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0600] In this invention, the server includes means for generating an appropriate response to an inquiry, means for calculating a confidence score for the generated response, means for displaying the confidence score, means for analyzing the user's emotional state, means for generating an appropriate response based on the analysis results, means for providing a sense of security, and means for achieving standardization of response skills. This makes it possible for customer service representatives and operators to provide consistent, high-quality responses to customers while gaining a sense of security.

[0601] "Means for generating appropriate responses to inquiries" refers to technologies that automatically generate appropriate answers based on the content of inquiries from users.

[0602] "Means for calculating a confidence score for a generated response" refers to techniques for evaluating and quantifying the accuracy and reliability of a generated response.

[0603] "Means for displaying confidence scores" refers to technologies for visually displaying calculated confidence scores to users.

[0604] "Means for analyzing a user's emotional state" refers to technologies for reading and analyzing a user's emotions from their words and actions.

[0605] "Means for generating appropriate responses based on analysis results" refers to technologies for generating optimal responses for users based on the results of sentiment analysis.

[0606] "Means of providing psychological reassurance" refers to technologies that enable users to feel safe and confident when using a system.

[0607] "Means for achieving standardization of response skills" refers to technologies that reduce variations in response skills among different users and achieve consistent response quality.

[0608] This invention is a system for enabling customer service representatives and operators to appropriately handle customer inquiries and complaints. This system uses a generative AI model to generate responses to inquiries, calculates and displays a confidence score, thereby providing users with a sense of security and standardizing response skills. Furthermore, it improves the quality of complaint handling by analyzing the user's emotional state using an emotion engine and generating appropriate responses.

[0609] Hardware and software to be used

[0610] Server: Provides computing resources for running generated AI models (e.g., OpenAI's GPT-4).

[0611] Terminal: The device the user operates on (e.g., PC, tablet, smartphone) displays the response sent from the server and the confidence score.

[0612] Emotion engine: Software used to analyze a user's emotional state (e.g., Microsoft's Azure Cognitive Services).

[0613] Program processing

[0614] 1. The user receives an inquiry from a customer. For example, the customer asks, "What is the warranty period for this product?"

[0615] 2. The server uses a generative AI model to input the prompt "What is the warranty period for this product?" and generates an appropriate response.

[0616] 3. The server calculates a confidence score for the generated response. For example, suppose the generated response is "This product has a one-year warranty" and the confidence score is 90%.

[0617] 4. The server sends the generated response and confidence score to the terminal.

[0618] 5. The device displays the received response and confidence score to the user. For example, it might display "Response: This product has a 1-year warranty (Confidence score: 90%)" on the screen.

[0619] 6. Users can review the displayed response and confidence score, and provide responses to customers with confidence.

[0620] Specific example

[0621] The user (operator) receives an inquiry from a customer.

[0622] The server uses a generative AI model to generate a response to the inquiry, "What is the warranty period for this product?"

[0623] The server calculates a confidence score for the generated response, "This product has a one-year warranty," and sends a message to the terminal indicating that the confidence score is 90%.

[0624] The device displays the response and confidence score to the user.

[0625] Users can check the confidence score and provide responses to customers with peace of mind.

[0626] Example of a prompt

[0627] "Generate an appropriate response to a customer inquiry. The inquiry is: 'What is the warranty period for this product?'"

[0628] In this way, this system utilizes AI technology to provide a sense of security and standardize response skills.

[0629] The flow of the specific processing in Example 2 will be explained using Figure 19.

[0630] Step 1:

[0631] The user receives an inquiry from a customer.

[0632] Input: Customer inquiry (e.g., "What is the warranty period for this product?")

[0633] Specific operation: The user (operator) receives customer inquiries via means such as phone or chat.

[0634] Step 2:

[0635] The server generates a response using a generated AI model.

[0636] Input: Inquiry received from the user (e.g., "What is the warranty period for this product?")

[0637] Data processing: Input the query content as a prompt message into a generative AI model (e.g., OpenAI's GPT-4).

[0638] Output: Generated response (e.g., "This product has a one-year warranty")

[0639] Specific operation: The server inputs the query details into a generation AI model and generates an appropriate response.

[0640] Step 3:

[0641] The server calculates a confidence score for the response it generates.

[0642] Input: Generated response (e.g., "This product has a one-year warranty")

[0643] Data processing: The response confidence score is calculated using an internal algorithm.

[0644] Output: Confidence score (e.g., 90%)

[0645] Specific operation: The server evaluates the accuracy and reliability of the generated response and calculates a confidence score.

[0646] Step 4:

[0647] The server sends the response and confidence score to the terminal.

[0648] Input: Generated response (e.g., "This product has a one-year warranty") and confidence score (e.g., 90%)

[0649] Data processing: Combine the response and confidence score into a single data packet.

[0650] Output: Data packets sent to the terminal

[0651] Specific operation: The server bundles the generated response and confidence score into a data packet and sends it to the terminal.

[0652] Step 5:

[0653] The device displays the response and confidence score to the user.

[0654] Input: Data packets sent from the server (response and confidence score)

[0655] Data processing: Analyze data packets and extract responses and confidence scores.

[0656] Output: The response and confidence score displayed to the user (e.g., "Response: This product has a 1-year warranty (Confidence score: 90%)")

[0657] Specific operation: The terminal analyzes the received data packets and displays the response and confidence score on the screen.

[0658] Step 6:

[0659] The user checks the displayed response and confidence score.

[0660] Input: The response and confidence score displayed on the device (e.g., "Response: This product has a 1-year warranty (Confidence score: 90%)")

[0661] Specific operation: The user (operator) checks the displayed response and confidence score to determine the reliability of the response.

[0662] Step 7:

[0663] The user provides a response to the customer.

[0664] Input: Confirmed response (e.g., "This product has a one-year warranty")

[0665] Specific actions: The user (operator) responds to the customer based on the confirmed response. For example, they might tell the customer, "The warranty period for this product is one year."

[0666] In this way, by performing specific actions at each step, the system achieves both the provision of psychological reassurance and the standardization of response skills.

[0667] (Application Example 2)

[0668] Next, we will describe Application Example 2 of Form 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."

[0669] Traditional customer service systems often failed to adequately understand the emotional state of users, leading to delays or inappropriate responses. Furthermore, the lack of a way to verify the reliability of AI-generated responses frequently caused anxiety among operators. Additionally, inconsistencies in response skills resulted in a lack of consistent service quality.

[0670] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0671] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of the issue, means for shortening response times, means for displaying a confidence score for AI-generated responses, means for analyzing the user's emotional state using an emotion engine, and means for generating appropriate responses based on the analysis results. This enables customer service representatives and operators to appropriately understand the user's emotional state and provide reliable responses quickly. Furthermore, by providing a sense of security and standardizing response skills, consistency in service quality can be maintained.

[0672] A "customer service representative" is an employee whose role is to directly receive and handle customer inquiries and complaints.

[0673] An "operator" is an employee whose role is to operate systems or machines and perform specific tasks.

[0674] A "customer complaint handling support system" is a system that provides support for appropriately responding to customer complaints.

[0675] "Means of providing appropriate responses" refers to the function of generating and providing appropriate answers to customer inquiries and complaints.

[0676] "Means to prevent online outrage" refer to functions that take measures to prevent customer dissatisfaction and anger from escalating.

[0677] "Means of shortening response times" refers to functions that enable quick responses to complaints and inquiries.

[0678] "A means of displaying a confidence score for AI-generated responses" refers to a function that quantifies and displays the reliability of AI-generated answers.

[0679] An "emotion engine" is software or hardware used to analyze a user's emotional state.

[0680] "Means for analyzing a user's emotional state" refers to a function that reads and analyzes a user's emotions from their words and actions.

[0681] "Means for generating appropriate responses based on analysis results" refers to a function that generates the optimal response based on the results of sentiment analysis.

[0682] "Means of providing a sense of security" refers to functions that provide support to enable operators to perform their duties with peace of mind.

[0683] "Means for achieving standardization of response skills" refers to a function that reduces variations in response skills among operators and maintains consistent service quality.

[0684] To implement this invention, the following hardware and software are required. The hardware requires a smartphone. The software requires Python, the OpenAI API, and the emotion_recognition library.

[0685] The server first receives user complaints and inquiries as input. Next, it uses the emotion_recognition library to analyze the user's emotional state. Based on this analysis, it uses the OpenAI API to generate an AI response to the user input. It obtains a confidence score for the generated response and displays it to the operator.

[0686] For example, if a user enters a complaint such as, "Recently, the security system has been malfunctioning, and I'm having trouble with it. What should I do?", the sentiment analysis result will be "anger." Based on this analysis result, the AI ​​will generate a response such as, "We apologize for the inconvenience. Could you please try restarting the system?" and display a confidence score of 0.85.

[0687] An example of a prompt message is as follows:

[0688] Please enter your user complaint or inquiry: Recently, our security system has been malfunctioning, and I'm experiencing problems. What should I do?

[0689] In this way, the server can accurately understand the user's emotional state and provide reliable responses quickly. Furthermore, by providing a sense of security and standardizing response skills, it can maintain consistency in service quality.

[0690] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[0691] Step 1:

[0692] Users enter complaints or inquiries using their smartphones. The entered text is sent to the server. The input data consists of the content of the user's complaint or inquiry.

[0693] Step 2:

[0694] The server passes the received user input to the emotion_recognition library, which analyzes the user's emotional state. The input data is the text of the user's complaint or inquiry, and the output data is the analyzed emotional state (e.g., "anger," "sadness," "joy," etc.).

[0695] Step 3:

[0696] The server uses the OpenAI API to generate AI responses to user input based on the analyzed emotional state. The input data consists of the user's complaint or inquiry text and the analyzed emotional state, while the output data is the text of the generated AI response.

[0697] Step 4:

[0698] The server obtains a confidence score for the generated AI response. The input data is the text of the generated AI response, and the output data is the confidence score (e.g., 0.85).

[0699] Step 5:

[0700] The server displays the generated AI response and confidence score on the operator's smartphone. The input data is the text of the generated AI response and its confidence score, while the output data is the information displayed on the operator's smartphone.

[0701] Step 6:

[0702] The operator reviews the displayed AI response and confidence score, and takes appropriate action for the user as needed. The input data is the text and confidence score of the AI ​​response reviewed by the operator, and the output data is the final response provided by the operator to the user.

[0703] In this way, the server can accurately understand the user's emotional state and provide reliable responses quickly. Furthermore, by providing a sense of security and standardizing response skills, it can maintain consistency in service quality.

[0704] (Example 3)

[0705] Next, we will describe Embodiment 3 of Embodiment Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0706] Traditional customer complaint handling systems often resulted in inconsistent quality of service due to the difficulty of call center staff and operators providing appropriate responses. Furthermore, they struggled to accurately understand user emotions during complaint handling, making it difficult to prevent escalations. Additionally, insufficient skill development and motivation maintenance for operators led to employment retention challenges. To address these issues, a system is needed that improves the quality of complaint handling, shortens response times, and supports operator skill development and motivation maintenance.

[0707] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0708] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of complaints, means for shortening response times, means for analyzing the user's emotional state using an emotion analysis engine, means for generating appropriate responses using a generative AI model, means for evaluating the quality of the generated responses, means for providing feedback to operators based on the evaluation results, and means for supporting the skill development of operators. This makes it possible to improve the quality of complaint handling, shorten response times, and support the skill development and motivation maintenance of operators.

[0709] A "customer service representative" is an employee whose role is to directly receive and handle customer inquiries and complaints.

[0710] An "operator" is a person responsible for operating the system and handling customer inquiries and complaints.

[0711] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[0712] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to customer complaints.

[0713] "Means to prevent online firestorms" refer to functions that appropriately handle customer dissatisfaction and anger, and prevent problems from escalating.

[0714] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[0715] An "emotion analysis engine" is software or hardware that analyzes a customer's emotional state and provides the results.

[0716] A "generative AI model" is an artificial intelligence model that generates an appropriate response based on a given prompt.

[0717] "Means for evaluating the quality of responses" refers to functions for evaluating the appropriateness and effectiveness of the generated responses.

[0718] "Means of providing feedback" refers to a function that provides operators with suggestions for improvement and advice based on the results of the response quality evaluation.

[0719] "Means of supporting skill development" refers to the function of providing training and educational programs to improve operators' response skills.

[0720] "Means of providing psychological reassurance" refers to functions that reduce the stress and anxiety that operators experience while handling complaints, and provide them with a sense of security.

[0721] "Means for achieving standardization of response skills" refers to functions that reduce variations in response skills among operators and maintain a consistent level of quality.

[0722] "Means to support the resolution of employment retention challenges" refer to functions that reduce the turnover rate of operators and support long-term employment retention.

[0723] This invention is a support system for counter staff and operators to appropriately handle customer complaints. The aim of this system is to improve the quality of complaint handling, reduce response times, and support the skill development and motivation maintenance of operators.

[0724] Hardware and software to be used

[0725] Hardware: Servers, terminals (PCs, tablets, smartphones)

[0726] Software: Generative AI models (e.g., GPT-4), sentiment analysis engines (e.g., Affectiva®), database management systems (e.g., MySQL)

[0727] Specific operation of the system

[0728] Receiving a complaint

[0729] User: Customer service representatives and operators receive customer complaints.

[0730] Terminal: The customer service representative or operator enters the details of the complaint into the terminal. For example, if a customer complains that "the product has not arrived," they would enter the details in text format.

[0731] Emotion analysis

[0732] Terminal: Sends the entered claim details to the server.

[0733] Server: Receives the complaint details and sends them to the sentiment analysis engine.

[0734] Emotion analysis engine: Analyzes the content of a complaint to identify the user's emotional state. For example, it might be analyzed as "anger."

[0735] Server: Receives analysis results from the emotion analysis engine and proceeds to the next processing step.

[0736] Response generation

[0737] Server: Based on the sentiment analysis results, it sends prompt messages to the generative AI model. For example, it might send a prompt message such as, "The customer is angry because their product hasn't arrived. Generate an appropriate response."

[0738] Generative AI model: Generates appropriate responses based on prompt text. For example, it can generate a response such as, "We are very sorry, customer. We will investigate immediately and address the issue as soon as possible."

[0739] Server: Receives the generated response and proceeds to the next processing step.

[0740] Response quality evaluation

[0741] Server: Executes algorithms to evaluate the quality of the generated response. For example, it evaluates the appropriateness of the response and its impact on customer satisfaction.

[0742] Server: Saves evaluation results to a database and generates feedback.

[0743] Feedback and skill development support

[0744] Server: Based on the evaluation results, it generates feedback for the operator. For example, it might generate feedback such as, "The response was appropriate this time, but it would be even better if you could provide a more specific solution."

[0745] Terminal: Operators receive feedback and participate in training programs to improve their skills. For example, they may conduct simulation training to improve the quality of their responses.

[0746] Examples of specific cases and prompt statements

[0747] Specific example: When a customer service representative receives a complaint from a customer that "the product has not arrived."

[0748] Terminal: The counter staff member enters the details of the complaint.

[0749] Server: Sends the complaint details to the sentiment analysis engine.

[0750] Emotion analysis engine: Analyzes the customer's emotional state as "anger".

[0751] Server: Sends a prompt message to the AI ​​model stating, "The customer is angry because the product has not arrived."

[0752] Generative AI model: Generates the response, "We are very sorry, customer. We will investigate immediately and take immediate action."

[0753] Server: Evaluates the quality of the response and provides feedback.

[0754] Example of a prompt:

[0755] "The customer is angry because their product hasn't arrived. Please generate an appropriate response."

[0756] "The customer is dissatisfied with the quality of service. Please generate an appropriate response."

[0757] This system helps improve operators' skills and maintain their motivation, thus resolving the challenge of continued employment. The flow of a specific process in Example 3 will be explained using Figure 21.

[0758] Step 1: Receiving the complaint

[0759] User: A customer service representative or operator receives a complaint from a customer. For example, a customer complains that "the product hasn't arrived."

[0760] Terminal: The customer service representative or operator enters the details of the complaint into the terminal. The entered complaint details are saved in text format.

[0761] Input: Customer complaint details (e.g., "The product did not arrive").

[0762] Output: Claims data in text format.

[0763] Step 2: Emotion Analysis

[0764] Terminal: Sends the entered claim details to the server.

[0765] Server: Receives the complaint details and sends them to the sentiment analysis engine.

[0766] Emotion analysis engine: Analyzes the content of a complaint to identify the user's emotional state. For example, it might be analyzed as "anger."

[0767] Input: Claim data in text format.

[0768] Output: Emotion analysis results (e.g., "anger").

[0769] Step 3: Response Generation

[0770] Server: Based on the sentiment analysis results, it sends prompt messages to the generative AI model. For example, it might send a prompt message such as, "The customer is angry because their product hasn't arrived. Generate an appropriate response."

[0771] Generative AI model: Generates appropriate responses based on prompt text. For example, it can generate a response such as, "We are very sorry, customer. We will investigate immediately and address the issue as soon as possible."

[0772] Input: Sentiment analysis results and prompt text.

[0773] Output: The generated response.

[0774] Step 4: Evaluate the quality of the response

[0775] Server: Executes algorithms to evaluate the quality of the generated response. For example, it evaluates the appropriateness of the response and its impact on customer satisfaction.

[0776] Server: Saves evaluation results to a database and generates feedback.

[0777] Input: The generated response.

[0778] Output: Response quality evaluation results and feedback.

[0779] Step 5: Feedback and Skill Development Support

[0780] Server: Based on the evaluation results, it generates feedback for the operator. For example, it might generate feedback such as, "The response was appropriate this time, but it would be even better if you could provide a more specific solution."

[0781] Terminal: Operators receive feedback and participate in training programs to improve their skills. For example, they may conduct simulation training to improve the quality of their responses.

[0782] Input: Response quality evaluation results and feedback.

[0783] Output: Operator skill development and implementation of training programs.

[0784] (Application Example 3)

[0785] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0786] Traditional customer complaint handling systems often lacked sufficient support for call center staff and operators to properly address complaints, resulting in inconsistent quality of service. Furthermore, the lack of functionality to accurately analyze user emotions and generate responses accordingly led to inappropriate complaint handling, sometimes resulting in escalating customer complaints. Additionally, the systems placed a significant mental burden on operators, posing challenges to their continued employment. To address these issues, a system is needed that analyzes user emotions and generates appropriate responses.

[0787] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0788] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response times, means for analyzing user emotions, means for generating appropriate responses based on the analysis results, means for evaluating the quality of the generated responses, means for providing feedback on the evaluation results, means for providing a sense of security, means for standardizing response skills, and means for supporting the resolution of employment retention issues. As a result, customer service staff and operators can provide appropriate responses in accordance with user emotions, improving the quality of complaint handling, preventing escalation, reducing mental burden, and promoting employment retention.

[0789] A "customer service representative" is an employee whose role is to directly receive and handle inquiries and complaints from customers and users.

[0790] An "operator" is an employee whose role is to operate systems or machines and perform specific tasks.

[0791] A "complaint handling support system" is a system designed to assist in appropriately responding to complaints from customers and users.

[0792] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints.

[0793] "Means to prevent online backlash" refer to functions designed to prevent problems from escalating due to inappropriate handling of complaints.

[0794] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[0795] "Means for analyzing user emotions" refers to functions that analyze user emotions from their statements and actions.

[0796] "Means for generating appropriate responses" refers to the function of generating appropriate responses based on analyzed emotions.

[0797] "Means for evaluating the quality of the generated response" refers to a function for evaluating the quality of the generated response.

[0798] "Means for providing feedback on evaluation results" refers to a function for providing feedback to the operator on the quality of the evaluated response.

[0799] "Means of providing psychological reassurance" refers to functions that provide psychological reassurance to operators.

[0800] "Means for achieving standardization of response skills" refers to functions for standardizing the response skills of operators.

[0801] "Means to support the resolution of employment retention challenges" refers to functions that support operators' skill development and motivation maintenance, thereby resolving employment retention challenges.

[0802] In order to implement this invention, it is necessary to build a claims handling support system. This system will help customer service staff and operators to respond to claims appropriately and will use the following hardware and software.

[0803] hardware

[0804] Server: Plays a central role in the customer complaint handling support system. It processes and analyzes data and generates responses.

[0805] Terminal: A smartphone or computer used by a customer service representative or operator. It receives user input and displays responses from the server.

[0806] software

[0807] OpenAI API: Provides generative AI models used for response generation.

[0808] EmotionEngine: A library for analyzing user emotions.

[0809] Database: A system for storing claim data and response data.

[0810] Processing flow

[0811] 1. Receiving user input

[0812] Users enter complaints through their devices. For example, they might say, "My security alarm keeps going off and it's causing me problems."

[0813] 2. Emotion analysis

[0814] The server uses EmotionEngine to analyze emotions from user input. For example, it might recognize the emotion "confusion" from user input.

[0815] 3. Response generation

[0816] The server uses the OpenAI API to generate an appropriate response based on the analyzed emotions. An example of a prompt is as follows:

[0817] User's emotion: Confusion

[0818] User input: I'm having trouble with a security alarm that keeps going off.

[0819] Please generate an appropriate response.

[0820] 4. Response Quality Evaluation

[0821] The server evaluates the quality of the generated response. For example, it checks whether it contains keywords such as "thank you" or "gratitude."

[0822] 5. Provide feedback

[0823] The server provides feedback on the evaluation results to the operator, allowing them to improve their response skills.

[0824] Specific example

[0825] If a user enters "I'm having trouble because my security alarm keeps going off," the server uses EmotionEngine to analyze the emotion "confusion." Then, it uses the OpenAI API to generate a response like the following:

[0826] "It seems you're having trouble. We'll address the security alarm issue immediately. Please wait a moment."

[0827] By evaluating the quality of the generated responses and providing feedback to the operators, it is possible to improve the operators' response skills.

[0828] In this way, customer service representatives and operators can respond appropriately to users' emotions, improving the quality of complaint handling, preventing escalations, reducing mental stress, and promoting continued employment.

[0829] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[0830] Step 1:

[0831] The user enters a complaint via their device. The entered complaint might be something like, "I'm having trouble because the security alarm keeps going off." This input data is sent to the server.

[0832] Step 2:

[0833] The server passes the user's input data to the EmotionEngine, which analyzes the user's emotions. The EmotionEngine analyzes the input data and recognizes, for example, an emotion such as "confusion." This analysis result is then returned to the server.

[0834] Step 3:

[0835] Based on the analyzed emotions, the server uses the OpenAI API to generate an appropriate response. The server generates a prompt message similar to the following and sends it to the OpenAI API.

[0836] User's emotion: Confusion

[0837] User input: I'm having trouble with a security alarm that keeps going off.

[0838] Please generate an appropriate response.

[0839] The OpenAI API generates a response based on this prompt and returns it to the server.

[0840] Step 4:

[0841] The server evaluates the quality of the generated response. For example, it checks whether the response contains keywords such as "thank you" or "gratitude." This evaluation result is stored within the server.

[0842] Step 5:

[0843] The server provides feedback on the evaluation results to the operator. This feedback is displayed to the operator via a terminal. This allows the operator to improve their response skills.

[0844] Step 6:

[0845] The operator will respond to the user appropriately based on the feedback provided by the server. For example, they might respond with, "It seems you're having trouble. We'll address the security alarm issue immediately. Please wait a moment."

[0846] In this way, customer service representatives and operators can respond appropriately to users' emotions, improving the quality of complaint handling, preventing escalations, reducing mental stress, and promoting continued employment.

[0847] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0848] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0849] Other examples of generative AI include Gemini® (registered trademark) (Internet search). <url: https: gemini.google.com ?hl="ja">) are some examples.

[0850] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0851] [Second Embodiment]

[0852] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0853] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0854] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0855] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0856] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0857] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0858] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0859] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0860] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0861] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0862] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0863] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0864] "Example of form 1"

[0865] One embodiment of the present invention is a complaint handling support system for counter staff and operators. This system incorporates an AI (artificial intelligence)-based automatic response generation module as a means of providing appropriate responses to complaints. This module analyzes the content of the complaint and generates an appropriate response. Furthermore, it includes a function to predict complaint escalation as a means of preventing further escalation. In addition, it employs an optimization algorithm to speed up response generation as a means of shortening response time. "Embodiment Example 2"

[0866] As an example of the present invention, there is a system that further includes means for providing a sense of security and means for achieving standardization of response skills. As means for providing a sense of security, the system has a function to display a confidence score for the response generated by the AI. This allows the operator to confirm the reliability of the response proposed by the AI ​​and gain a sense of security. Furthermore, as means for achieving standardization of response skills, the system allows operators to refer to the responses generated by the AI, thereby reducing the variation in response skills among operators.

[0867] "Example of form 3"

[0868] As a third embodiment of the present invention, there is a system that further includes means to support the resolution of the challenge of continued employment. Specifically, it has a function to evaluate the quality of responses generated by AI and to support the improvement of operators' skills. This helps to improve operators' skills and maintain their motivation, thereby supporting the resolution of the challenge of continued employment.

[0869] The following describes the processing flow for each example of the form.

[0870] "Example of form 1"

[0871] Step 1: A customer service representative or operator receives the complaint.

[0872] Step 2: The system analyzes the content of the claim.

[0873] Step 3: An AI (artificial intelligence) automated response generation module generates an appropriate response.

[0874] Step 4: The system predicts the escalation of a claim.

[0875] Step 5: An optimization algorithm is activated to speed up response generation and reduce response time.

[0876] "Example of form 2"

[0877] Step 1: The customer service representative or operator reviews the response generated by the AI.

[0878] Step 2: The system displays a confidence score for the AI-generated response.

[0879] Step 3: The customer service representative or operator decides on the final response based on the confidence score.

[0880] Step 4: Customer service staff and operators use AI-generated responses as a reference to standardize their response skills.

[0881] "Example of form 3"

[0882] Step 1: The customer service representative or operator uses the AI-generated response.

[0883] Step 2: The system evaluates the quality of the response generated by the AI.

[0884] Step 3: The system provides feedback to help operators improve their skills.

[0885] Step 4: Customer service staff and operators use feedback to improve their skills and maintain their motivation.

[0886] (Example 1)

[0887] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0888] Traditional complaint handling systems often lacked sufficient support for customer service representatives and operators to properly address complaints, resulting in inconsistencies in the quality and speed of responses. Furthermore, there was no way to predict the risk of complaint escalation, increasing the likelihood of escalation due to delays. Moreover, generating quick and appropriate responses was difficult amidst the pressure to shorten response times. A system is needed to address these challenges.

[0889] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0890] In this invention, the server includes means for providing appropriate responses to claims, means for preventing further escalation, means for shortening response times, means for analyzing the content of claims, means for generating automated responses, means for predicting escalation, and means for accelerating response generation. This improves the quality and speed of claim handling, reduces the risk of escalation, and enables the provision of quick and appropriate responses.

[0891] "Means of providing appropriate responses to complaints" refers to a function that generates and provides appropriate responses to complaints from users.

[0892] "Means to prevent further escalation of the issue" refers to functions designed to prevent complaints from escalating and the problem from spreading.

[0893] "Means for shortening response time" refers to functions that shorten response time by quickly generating and providing responses to complaints.

[0894] "Means for analyzing complaint content" refers to a function that analyzes the content of user complaints to identify the type of problem and the emotions involved.

[0895] "Means for generating automatic responses" refers to a function that automatically generates an appropriate response based on the analysis results.

[0896] "Means for predicting escalation" refers to functions for predicting the likelihood of a claim escalating and for assessing the associated risks.

[0897] "Means for speeding up response generation" refers to functions that optimize the response generation process and provide responses quickly.

[0898] This invention is a support system for customer service representatives and operators to appropriately handle customer complaints. The system includes multiple means for providing appropriate responses to complaints, preventing escalation, and shortening response times.

[0899] System Configuration

[0900] Hardware and software

[0901] Server: Analyzes claim details, generates automated responses, predicts escalations, and speeds up response generation.

[0902] Terminal: A device (such as a PC or smartphone) used by the user to enter a complaint and receive a response from the server.

[0903] Natural Language Processing (NLP) techniques: Used to analyze the content of the claims. Specifically, Google Cloud Natural Language API and IBM Watson Natural Language Understanding will be used.

[0904] Generative AI model: Used to generate automated responses. Specifically, it utilizes OpenAI's GPT-3.

[0905] Machine learning algorithms: Used to perform escalation prediction. Specifically, scikit-learn and TensorFlow are used.

[0906] Cache technology and parallel processing: Used to speed up response generation.

[0907] System operation

[0908] 1. Entering a complaint

[0909] The user enters the details of the complaint from their device.

[0910] Example: The user enters "The product quality is poor."

[0911] 2. Analysis of the complaint content

[0912] The server receives the complaint details sent by the user.

[0913] The server uses the Google Cloud Natural Language API to parse the claim content.

[0914] Specifically, the text of the complaint is sent to the API for sentiment analysis and keyword extraction.

[0915] Example: Extract the keyword "quality problem" from the text "The product quality is poor" and determine that the emotion is "negative".

[0916] 3. Generating automated responses

[0917] The server uses a generative AI model (GPT-3) based on the analysis results to generate an appropriate response.

[0918] Specifically, the analysis results are input as prompts into the AI ​​model, which then generates response sentences.

[0919] Example: Enter the prompt "The user has submitted a complaint that the product quality is poor. Please generate an appropriate response." and retrieve the generated response "We apologize. Could you please tell us more about the product quality?".

[0920] 4. Escalation prediction

[0921] The server uses machine learning algorithms to predict the likelihood of a claim escalating.

[0922] Specifically, we will use scikit-learn to build a model that assesses the risk of escalation based on past claim data, and then calculate a risk score for the current claim.

[0923] Example: A high risk score is assigned to a complaint that states, "I have contacted them multiple times, but the issue remains unresolved."

[0924] 5. Providing a response

[0925] The server provides the generated response to the user.

[0926] Specifically, the generated response message is sent to the terminal and displayed on the user's screen.

[0927] Example: The user's device will display the response, "We apologize. Could you please tell us more about the quality of the product?"

[0928] 6. Speeding up response generation

[0929] The server uses caching technology and parallel processing to speed up response generation.

[0930] Specifically, past responses to the same type of claim are stored in a cache and reused. Furthermore, processing speed is improved by executing multiple claim processing tasks in parallel.

[0931] Example: If a response to a complaint such as "The product quality is poor" is stored in the cache, retrieve the response from the cache and provide it quickly.

[0932] Specific example

[0933] A user submits a complaint stating that "the product quality is poor."

[0934] The server sends the complaint details to the Google Cloud Natural Language API, which extracts the keywords "quality issues" and the sentiment "negative."

[0935] The server inputs the prompt "The user has entered a complaint that the product quality is poor. Please generate an appropriate response." into the generating AI model (GPT-3) and generates the response "We apologize. Could you please tell us more about the product quality?"

[0936] The server uses scikit-learn to calculate the escalation risk score and assigns a high risk score to the case.

[0937] The server sends the generated response to the user's terminal and displays it on the user's screen.

[0938] The server stores responses to the same type of claim in a cache to speed up response generation in subsequent instances.

[0939] The above describes the embodiments for carrying out this invention.

[0940] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0941] Step 1:

[0942] Entering a complaint

[0943] The user enters the details of the complaint from their device.

[0944] Input: The content of the complaint entered by the user (e.g., "The product quality is poor").

[0945] Output: The claim details are sent to the server.

[0946] Specific action: The user enters the complaint details into the input form on the device and presses the submit button.

[0947] Step 2:

[0948] Received the details of the complaint

[0949] The server receives the complaint details sent by the user.

[0950] Input: The content of the complaint submitted by the user.

[0951] Output: The claim details are saved on the server.

[0952] Specific operation: The server receives an HTTP request and saves the claim details to the database.

[0953] Step 3:

[0954] Analysis of complaint content

[0955] The server uses the Google Cloud Natural Language API to parse the claim content.

[0956] Input: Saved claim details.

[0957] Output: Analysis results (keywords, sentiment analysis results, etc.).

[0958] Specific operation: The server sends the claim details to the API and retrieves the analysis results returned by the API.

[0959] Step 4:

[0960] Generating an automated response

[0961] The server uses a generative AI model (GPT-3) based on the analysis results to generate an appropriate response.

[0962] Input: Analysis results.

[0963] Output: The generated response.

[0964] Specific operation: The server inputs the analysis results as prompts into the AI ​​model and retrieves the generated response.

[0965] Example: Enter the prompt "The user has submitted a complaint that the product quality is poor. Please generate an appropriate response." and generate the response "We apologize. Could you please tell us more about the product quality?"

[0966] Step 5:

[0967] Escalation prediction

[0968] The server uses machine learning algorithms to predict the likelihood of a claim escalating.

[0969] Input: Claim details and analysis results.

[0970] Output: Escalation risk score.

[0971] Specific operation: The server uses a model built on past claim data to calculate a risk score for the current claim.

[0972] Example: A high risk score is assigned to a complaint that states, "I have contacted them multiple times, but the issue remains unresolved."

[0973] Step 6:

[0974] Providing a response

[0975] The server provides the generated response to the user.

[0976] Input: The generated response.

[0977] Output: The response message is displayed on the user's terminal.

[0978] Specific operation: The server sends the generated response message to the user's terminal and displays it on the user's screen.

[0979] Example: The user's device will display the response, "We apologize. Could you please tell us more about the quality of the product?"

[0980] Step 7:

[0981] Speeding up response generation

[0982] The server uses caching technology and parallel processing to speed up response generation.

[0983] Input: Past complaint details and responses.

[0984] Output: Cached response statement.

[0985] Specific operation: The server stores past responses to the same type of claim in a cache and reuses them. It also improves processing speed by executing multiple claim processing in parallel.

[0986] Example: If a response to a complaint such as "The product quality is poor" is stored in the cache, retrieve the response from the cache and provide it quickly.

[0987] (Application Example 1)

[0988] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0989] Traditional customer complaint handling systems had several problems: long response times and decreased customer satisfaction because customer service representatives and operators handled complaints manually. Furthermore, it was difficult to predict complaint escalation, making it impossible to notify superiors at the appropriate time, increasing the risk of complaints escalating. In addition, inconsistencies in handling skills and significant mental stress among staff also created challenges in retaining employees.

[0990] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0991] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response times, means for analyzing complaint content and generating appropriate responses, means for predicting complaint escalation, means for automatically notifying senior personnel, means for generating responses using a generation AI model, and means for generating responses using prompt sentences. This enables more efficient and faster complaint handling, leading to improved customer satisfaction and reduced risk of complaint escalation. Furthermore, by standardizing response skills and providing a sense of security, it can also solve the challenge of retaining employees.

[0992] A "customer service representative" is someone whose job is to directly handle customer inquiries and complaints.

[0993] An "operator" is someone whose job is to operate systems or machines and perform specific tasks.

[0994] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[0995] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints.

[0996] "Means to prevent online firestorms" are functions designed to prevent complaints from escalating and problems from becoming serious.

[0997] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[0998] "Means for analyzing the content of a claim and generating an appropriate response" refers to a function that analyzes the content of a claim and generates an appropriate response based on that analysis.

[0999] "Methods for predicting complaint escalation" refer to functions that predict the likelihood of a complaint escalating to a higher-level employee.

[1000] The "automatic notification mechanism for senior personnel" is a function that automatically notifies senior personnel when a complaint escalates.

[1001] "Response generation means using a generative AI model" refers to a function that uses an artificial intelligence model to generate a response to a claim.

[1002] A "response generation means using a prompt statement" is a function for generating a response to a claim using a specific input statement (prompt statement).

[1003] As an example of how to implement this invention, we will describe a customer support application for an e-commerce site. This system can automatically generate appropriate responses to customer complaints and inquiries using AI and respond quickly. It also has a function to predict complaint escalation and automatically notify higher-level personnel as needed.

[1004] System Configuration

[1005] This system consists of the following main components:

[1006] 1. Claim Analysis Module: Analyzes the claim content and generates an appropriate response.

[1007] 2. Escalation prediction module: Predicts the escalation of claims.

[1008] 3. Notification Module: Automatically notifies senior personnel when an escalation is anticipated.

[1009] 4. Generative AI Model: An artificial intelligence model for generating responses to complaints.

[1010] 5. Prompt statement generation module: Generates a response using a specific input statement (prompt statement).

[1011] Hardware and software

[1012] This system uses internet-connected smartphones or PCs as hardware. The software utilizes Python and the OpenAI API.

[1013] Processing flow

[1014] 1. Entering the complaint details: The user enters the details of the complaint.

[1015] 2. Response Generation: The claim analysis module analyzes the claim content and generates an appropriate response using a generation AI model.

[1016] 3. Escalation Prediction: The escalation prediction module predicts the likelihood of a claim escalating.

[1017] 4. Notifications: If an escalation is anticipated, the notification module will automatically notify the higher-level responsible party.

[1018] Specific example

[1019] For example, if a customer enters a complaint stating, "I did not receive the product, so I would like a refund," the system will operate as follows:

[1020] 1. Entering the complaint details: The user enters, "I did not receive the product, so I would like a refund."

[1021] 2. Response Generation: The complaint analysis module analyzes this content and uses a generation AI model to generate a response such as, "We apologize. We will immediately investigate the issue of the product not being delivered and proceed with the refund process."

[1022] 3. Escalation Prediction: The escalation prediction module predicts that this claim has a "high" likelihood of escalation.

[1023] 4. Notification: The notification module automatically notifies the senior person in charge, who then takes over the responsibility.

[1024] Example of a prompt

[1025] Please enter your complaint details: I did not receive the item and would like a refund.

[1026] In this way, customer support for e-commerce sites can be streamlined, and customer satisfaction can be improved.

[1027] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1028] Step 1:

[1029] The user enters the details of their complaint.

[1030] Input: The user enters, "I did not receive the product, so I would like a refund."

[1031] Specific action: The user enters the details of their complaint in text format into an input form on their smartphone or PC.

[1032] Step 2:

[1033] The server receives the claim details and sends them to the claim analysis module.

[1034] Input: The complaint details entered by the user.

[1035] Output: Claim details sent to the claim analysis module.

[1036] Specific operation: The server receives input from the user and sends its contents to the claims analysis module.

[1037] Step 3:

[1038] The claim analysis module analyzes the claim content and generates an appropriate response using a generative AI model.

[1039] Input: Claim details sent to the claim analysis module.

[1040] Output: The generated response text.

[1041] Specific operation: The claim analysis module analyzes the claim content using natural language processing technology and sends it as a prompt to the generative AI model. The generative AI model generates an appropriate response and returns it to the claim analysis module.

[1042] Step 4:

[1043] The server sends the generated response to the user.

[1044] Input: The generated response text.

[1045] Output: The response text displayed to the user.

[1046] Specific operation: The server receives the generated response and sends it to the user's terminal. The user's terminal displays the response.

[1047] Step 5:

[1048] The escalation prediction module predicts the likelihood of a claim escalating.

[1049] Input: Complaint details.

[1050] Output: Escalation prediction result (high / low).

[1051] Specific operation: The escalation prediction module analyzes the complaint content and uses a generative AI model to predict the likelihood of escalation. The prediction result is returned to the server.

[1052] Step 6:

[1053] The server receives the escalation prediction results and automatically notifies higher-level personnel as needed.

[1054] Input: Escalation prediction result.

[1055] Output: Notification to the superior in charge.

[1056] Specific operation: The server receives the escalation prediction result, and if the prediction result is "high," it sends a notification to the higher-level person in charge. The notification is sent via email or a dedicated application.

[1057] In this way, the system can streamline and expedite complaint handling, thereby improving customer satisfaction and reducing the risk of complaints escalating.

[1058] (Example 2)

[1059] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[1060] Conventional complaint handling support systems lack sufficient support for operators to provide appropriate responses, and they lack means to provide a sense of security and standardize response skills. Furthermore, they lack effective means to shorten response times and prevent issues from escalating. As a result, operators face increased burdens and inconsistent response quality, which is a significant challenge.

[1061] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1062] In this invention, the server includes means for receiving inquiries, means for generating responses using a generative AI model, means for calculating a confidence score for the generated responses, means for displaying the responses and confidence scores, means for providing appropriate responses to complaints, means for preventing further escalation of issues, and means for shortening response times. This allows operators to verify the reliability of AI-generated responses and gain peace of mind, while also reducing variability in response skills, shortening response times, and preventing escalations of issues.

[1063] "Means for receiving inquiries" refers to the function that allows a server to receive inquiries from users.

[1064] "Means for generating responses using a generative AI model" refers to a function that uses a generative AI model to generate an appropriate response to an incoming inquiry.

[1065] "Means for calculating a confidence score for a generated response" refers to a function for evaluating the reliability of a generated response and calculating a confidence score.

[1066] "Means for displaying responses and confidence scores" refers to a function that visually displays the generated responses and their confidence scores to the user.

[1067] "Means of providing appropriate responses to complaints" refers to a function that provides responses in order to address complaints appropriately.

[1068] "Means to prevent further escalation of the issue" refers to a function that provides preventative measures to prevent problems from escalating during complaint handling.

[1069] "Means for shortening response time" refers to functions designed to improve efficiency in order to reduce the time spent handling complaints.

[1070] "Means of providing psychological reassurance" refers to a function that allows operators to gain psychological reassurance by verifying the reliability of the AI's responses.

[1071] "Means for achieving standardization of response skills" refers to functions that reduce variations in response skills among operators and achieve uniform responses.

[1072] "Means to support the resolution of employment continuity issues" refers to functions that provide support for resolving issues related to the continued employment of operators.

[1073] This invention provides operators with peace of mind and standardizes their response skills by generating responses using a generative AI model and displaying a confidence score in a claims handling support system. A specific embodiment of this system is described below.

[1074] Server Processing

[1075] The server receives a query from the user. The query is a prompt, such as "Please tell me the progress of the new project." The server inputs this query into a generative AI model (e.g., GPT-4) and generates an appropriate response. For the generated response, the server calculates a confidence score. This confidence score is calculated using the AI ​​model's internal evaluation mechanism and additional evaluation algorithms. The server sends the generated response and the confidence score to the terminal.

[1076] Terminal processing

[1077] The terminal receives the response and confidence score sent from the server. The received response and confidence score are displayed on the terminal's screen. For example, it might say, "Current progress is on track. The next milestone is next week. (Confidence score: 85%)." The terminal uses visual aids to make it easy to understand, such as displaying a high confidence score in green and a low score in red.

[1078] User processing

[1079] The user checks the response and confidence score displayed on the device. They evaluate the reliability of the response based on the confidence score. For example, if the confidence score is 85%, the user can trust and accept the response. If the confidence score is low, the user can make another inquiry. When making a re-inquiry, they can add more specific questions based on the previous response.

[1080] Specific example

[1081] Specific example 1: A means of providing a sense of mental security

[1082] A user sends an inquiry asking, "Please tell me the progress of the new project." The server uses a generative AI model (e.g., GPT-4) to generate a response saying, "The current progress is on track. The next milestone is next week," and calculates a confidence score of 85%. The terminal displays this response and the confidence score, and the user feels reassured after checking the confidence score.

[1083] Specific example 2: A means to achieve standardization of response skills

[1084] Operator A receives an inquiry asking, "Could you please explain the procedure for handling customer complaints?" The server uses a generative AI model to generate a response saying, "First, listen carefully to the customer's story, then confirm the details of the problem, and finally, propose a solution." Operator A uses this response as a reference, and Operator B can use the same procedure as a reference for similar inquiries, thereby reducing variability in their response skills.

[1085] Example of a prompt

[1086] "Please tell me about the progress of the new project."

[1087] "Please explain the procedure for handling customer complaints."

[1088] By inputting these prompt sentences into a generating AI model, appropriate responses and confidence scores can be obtained.

[1089] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1090] Step 1:

[1091] The server receives inquiries from users. When a user enters a prompt message such as "Please tell me the progress of the new project" through their terminal and presses the send button, the server receives the inquiry. The input is the user's inquiry, and the output is the inquiry data stored on the server.

[1092] Step 2:

[1093] The server inputs the received query into the generative AI model. Specifically, the server sends the query content to the generative AI model (e.g., GPT-4) via an API. The input is the query data, and the output is the prompt message sent to the generative AI model.

[1094] Step 3:

[1095] The generative AI model generates a response based on a prompt. The generative AI model uses an internal natural language processing algorithm to generate an appropriate response. The input is the prompt, and the output is the generated response text.

[1096] Step 4:

[1097] The server calculates a confidence score for the generated response. The server uses the AI ​​model's internal evaluation mechanism and additional evaluation algorithms to assess the reliability of the response and calculate the confidence score. The input is the generated response text, and the output is the confidence score.

[1098] Step 5:

[1099] The server sends the generated response and confidence score to the terminal. The server combines the response text and confidence score into a single data packet and sends it to the terminal. The input is the response text and confidence score, and the output is the data packet sent to the terminal.

[1100] Step 6:

[1101] The terminal receives the response and confidence score sent from the server. The terminal analyzes the received data packets and extracts the response text and confidence score. The input is the data packet sent from the server, and the output is the analyzed response text and confidence score.

[1102] Step 7:

[1103] The terminal displays the received response and confidence score to the user. The terminal screen displays "Current progress is on track. The next milestone is next week. (Confidence score: 85%)." The input is the parsed response text and confidence score, and the output is the information displayed to the user.

[1104] Step 8:

[1105] The user checks the response and confidence score displayed on the device. The user then evaluates the reliability of the response based on the confidence score. The input is the displayed response text and confidence score, and the output is the user's evaluation result.

[1106] Step 9:

[1107] The user either accepts the response or makes a new inquiry. If the confidence score is high, the user trusts and accepts the response. If the confidence score is low, the user makes a new inquiry with more specific questions. The input is the user's evaluation result, and the output is the new inquiry or the acceptance of the response.

[1108] (Application Example 2)

[1109] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[1110] Traditional complaint handling support systems often resulted in significant mental strain on operators when providing appropriate responses to complaints, leading to inconsistencies in their response skills. Furthermore, security operators lacked a reliable means to verify the reliability of AI-generated suggestions, making quick and appropriate responses difficult. This resulted in prolonged response times and an increased risk of the issue escalating.

[1111] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1112] This invention includes a server that provides appropriate responses to complaints, prevents further escalation of the issue, shortens response times, displays a confidence score for AI-generated responses, reduces variations in response skills by allowing operators to refer to AI-generated responses, and enables security operators to receive AI suggestions and respond while checking their confidence scores. This allows operators to respond quickly and appropriately while gaining peace of mind.

[1113] A "customer service representative" is a person whose job is to directly handle inquiries and complaints from customers and users.

[1114] An "operator" is a person whose job is to operate a system or machine and perform a specific task.

[1115] A "complaint handling support system" is a system designed to assist in appropriately responding to complaints from customers and users.

[1116] "Means of providing appropriate responses" refers to the function of generating accurate and effective responses to complaints and providing them to operators.

[1117] "Means to prevent online backlash" refers to the function of taking measures to prevent problems from escalating in handling complaints.

[1118] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1119] "A means of displaying a confidence score for AI-generated responses" refers to a function that quantifies the reliability of AI-generated responses and displays it to the operator.

[1120] "A means of reducing variations in response skills by having operators refer to responses generated by AI" refers to a function that reduces differences in response skills among operators by having operators refer to responses generated by AI.

[1121] A "security operator" is a person whose job is to handle security-related tasks and ensure the safety of systems and facilities.

[1122] A "confidence score" is a numerical value used to evaluate the reliability of a response generated by AI.

[1123] As an embodiment of this invention, a claims handling support system used by security operators will be specifically described.

[1124] System Configuration

[1125] This system consists of the following main components:

[1126] 1. Server: Hosts the AI ​​model and generates responses to queries.

[1127] 2. Terminal: A device such as a smartphone or tablet used by a security operator.

[1128] 3. User Interface: An application for operators to input queries and view AI responses and confidence scores.

[1129] Hardware and software to be used

[1130] Hardware: Smartphones, tablets, servers

[1131] Software: Flask (Python web framework), requests (library for sending HTTP requests), generative AI model

[1132] Data processing and data calculation

[1133] The server receives queries from security operators and sends prompts to the generative AI model. The generative AI model generates responses to the queries and returns those responses to the server. The server calculates a confidence score for the responses and sends it to the operator's terminal.

[1134] Processing flow

[1135] 1. Query Input: The security operator inputs the query through the terminal's user interface.

[1136] 2. Sending the query: The terminal sends the query to the server.

[1137] 3. Response generation: The server sends a query to the generating AI model and generates a response.

[1138] 4. Calculation of confidence score: The server calculates a confidence score for the generated response.

[1139] 5. Display of response and score: The server sends the response and confidence score to the terminal for the operator to review.

[1140] Specific example

[1141] For example, a security operator might enter a query such as, "What should I do if an intruder enters the building?" This query is sent to a server, and a generative AI model generates an appropriate response. The server calculates a confidence score for the generated response and displays the response and score on the operator's terminal.

[1142] Example of a prompt

[1143] "Please tell me how to respond if a suspicious person enters a building."

[1144] In this way, security operators can take quick and appropriate action by referring to AI suggestions. This allows them to gain peace of mind while reducing variability in response skills.

[1145] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1146] Step 1:

[1147] The security operator enters the query through the terminal's user interface.

[1148] Input: Query entered by the security operator (e.g., "Please tell me how to respond if an intruder enters the building.")

[1149] Output: The query is entered into the terminal.

[1150] Specific actions: The security operator opens the application on their smartphone or tablet, enters the query in the text box, and presses the "Submit" button.

[1151] Step 2:

[1152] The terminal sends a query to the server.

[1153] Input: Query entered on the terminal

[1154] Output: The query is sent to the server.

[1155] Specific operation: The application on the terminal sends the entered query to the server as an HTTP request.

[1156] Step 3:

[1157] The server sends queries to the generated AI model and generates responses.

[1158] Input: Query sent to the server

[1159] Output: Response from the generative AI model

[1160] Specific operation: The server sends the received query as a prompt to the generating AI model, and the generating AI model generates a response to the query and returns it to the server.

[1161] Step 4:

[1162] The server calculates a confidence score for the response it generates.

[1163] Input: Response from a generative AI model

[1164] Output: Confidence score

[1165] Specific operation: The server analyzes the content of the generated response and applies an algorithm to calculate a confidence score.

[1166] Step 5:

[1167] The server sends the response and confidence score to the terminal.

[1168] Input: Responses and confidence scores from the generated AI model

[1169] Output: Response and confidence score sent to the terminal

[1170] Specific operation: The server sends the response and confidence score together as an HTTP response to the terminal.

[1171] Step 6:

[1172] The terminal displays its response and confidence score to the security operator.

[1173] Input: Response sent from the server and confidence score

[1174] Output: Response and confidence score displayed to the security operator

[1175] Specific operation: The terminal application displays the received response and confidence score on the screen for the security operator to review.

[1176] (Example 3)

[1177] Next, we will describe Embodiment 3 of Embodiment Example 3. 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."

[1178] Conventional complaint handling support systems have been insufficient in providing adequate support for customer service staff and operators to respond appropriately, making it difficult to shorten response times, provide a sense of security, and standardize response skills. Furthermore, it has been difficult to improve operators' skills and maintain their motivation, posing challenges to continued employment.

[1179] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1180] In this invention, the server includes means for generating responses to prompt sentences using a generative AI model, means for evaluating the quality of the generated responses, and means for providing the evaluation results as feedback. This enables operators to respond appropriately and quickly, improving their skills and maintaining motivation. It also provides a sense of security, standardizes response skills, and solves the challenge of retaining employees.

[1181] A "customer service representative" is an employee whose role is to directly handle customer inquiries and complaints.

[1182] An "operator" is a person responsible for operating the system and handling customer inquiries and complaints.

[1183] A "customer complaint handling support system" is a system that provides support for appropriately responding to customer complaints.

[1184] A "generative AI model" is a model that uses artificial intelligence to generate responses to input prompt sentences.

[1185] A "prompt message" is text, such as a question or instruction, that a user enters into the system.

[1186] A "response" is the text of an answer or suggestion that a generative AI model generates based on a prompt.

[1187] A "quality evaluation algorithm" is an algorithm used to evaluate the quality of a generated response.

[1188] "Feedback" refers to information provided to operators based on evaluation results, intended to encourage skill development and improvement.

[1189] "Skill improvement" refers to an operator improving their own abilities and skills.

[1190] "Maintaining motivation" refers to operators maintaining their enthusiasm and motivation for their work.

[1191] "Mental security" refers to the sense of stability and security that operators feel while performing their work.

[1192] "Standardizing response skills" refers to ensuring that multiple operators possess the same level of response skills.

[1193] "Continued employment" refers to an operator continuing to work at the same workplace for an extended period.

[1194] This invention is a support system for counter staff and operators to appropriately handle customer complaints. This system uses a generative AI model to generate responses to prompt sentences, evaluates the quality of those responses, and provides the evaluation results as feedback, thereby aiming to improve operators' skills and maintain their motivation.

[1195] System Configuration

[1196] hardware

[1197] Server: Provides computing resources for running the generated AI model and quality evaluation algorithm.

[1198] Terminal: A device used by the user to input prompts and display responses and evaluation results from the server.

[1199] software

[1200] Generative AI model: For example, use a natural language generation model such as GPT-4.

[1201] Quality evaluation algorithm: For example, use a BERT-based evaluation model.

[1202] Program processing

[1203] The server receives a prompt message entered by the user from the terminal. Next, it generates a response to the prompt message using a generative AI model. The generated response is evaluated by a quality evaluation algorithm, and the evaluation result is sent to the terminal as feedback. The terminal displays the evaluation result and the generated response to the user.

[1204] Specific example

[1205] For example, if a user enters the prompt "Please tell me the appropriate way to handle customer complaints," the server will process it as follows:

[1206] 1. The server receives the prompt message "Please tell me the appropriate way to handle customer complaints."

[1207] 2. Using a generative AI model (e.g., GPT-4), generate a response like this: "First, it's important to listen to the customer completely and show empathy. Then, you need to propose concrete solutions and respond quickly."

[1208] 3. The server evaluates this response using a quality evaluation algorithm (for example, a BERT-based evaluation model) and determines, for example, that "the response quality is high."

[1209] 4. Send these evaluation results to the operator as feedback to support their skill development.

[1210] Other examples of prompt statements include the following:

[1211] "Please briefly explain the features of the new product."

[1212] "Please tell me the appropriate way to handle customer complaints."

[1213] "Please suggest ways to boost team motivation."

[1214] This system allows operators to objectively evaluate the quality of their responses, thereby promoting skill development and maintaining motivation, and supporting the resolution of employment retention challenges. The flow of a specific process in Example 3 will be explained using Figure 15.

[1215] Step 1:

[1216] The user enters a prompt message.

[1217] The user enters prompt text, such as a question or instruction, into the terminal's input field. For example, they might enter, "Please tell me the appropriate way to handle a customer complaint." The entered prompt text is temporarily stored in the terminal's memory.

[1218] Step 2:

[1219] The terminal sends a prompt message to the server.

[1220] The terminal sends the prompt text entered by the user to the server. The prompt text is sent to the server in an appropriate format (e.g., JSON). The input is the prompt text, and the output is the data sent to the server.

[1221] Step 3:

[1222] The server generates a response using a generated AI model.

[1223] The server inputs the received prompt into a generating AI model (e.g., GPT-4) to generate a response. The input is the prompt, and the output is the generated response. For example, a response like, "First, it is important to listen to the customer completely and show empathy. After that, you are required to propose concrete solutions and respond quickly," might be generated.

[1224] Step 4:

[1225] The response generated by the server is evaluated using a quality evaluation algorithm.

[1226] The server inputs the generated response into a quality evaluation algorithm (e.g., a BERT-based evaluation model) to evaluate the quality of the response. The input is the generated response, and the output is the evaluation result. For example, a score or comment such as "Response quality is high" might be output.

[1227] Step 5:

[1228] The server sends the evaluation results to the terminal.

[1229] The server sends the evaluation results to the terminal. The evaluation results are sent along with the response. The input consists of the evaluation results and the generated response, while the output is the data sent to the terminal.

[1230] Step 6:

[1231] The device displays the evaluation results to the user.

[1232] The terminal displays the evaluation results received from the server and the generated responses to the user. The input is the evaluation results and generated responses, and the output is the data displayed to the user. The user can use the displayed evaluation results as a reference to improve their own skills.

[1233] (Application Example 3)

[1234] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[1235] Conventional customer complaint support systems fail to adequately address operator skill development and motivation maintenance, leaving the challenge of continued employment unresolved. Furthermore, the lack of a means to evaluate the quality of AI-generated responses results in inconsistent operator response quality. Additionally, the system fails to consider the convenience of being a smartphone application.

[1236] In Application Example 3, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of the issue, means for shortening response times, means for evaluating the quality of responses generated by the AI, means for supporting operator skill improvement, means for maintaining motivation, and means for supporting the resolution of employment retention issues. This enables operator skill improvement and motivation maintenance, and resolves employment retention issues. Furthermore, by evaluating the quality of responses generated by the AI, the quality of responses is made more uniform. In addition, the convenience of being an application installed on a smartphone is also provided.

[1237] A "customer service representative" is an employee whose job is to directly handle customer inquiries and complaints.

[1238] An "operator" is an employee whose job is to handle customer inquiries using communication methods such as telephone or chat.

[1239] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[1240] "Means of providing appropriate responses" refers to the function of generating and providing appropriate answers to customer complaints.

[1241] "Means to prevent online backlash" refer to functions that prevent problems from escalating during complaint handling.

[1242] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1243] "Means for evaluating the quality of AI-generated responses" refers to a function for evaluating the quality of responses generated by artificial intelligence.

[1244] "Means to support operator skill development" refers to functions designed to improve operators' response skills.

[1245] "Means for maintaining motivation" refers to functions that maintain the operator's enthusiasm for their work.

[1246] "Means to support the resolution of employment continuity challenges" refers to functions that address the challenges of maintaining the employment of operators.

[1247] "Means of providing psychological reassurance" refers to functions that provide psychological reassurance to operators.

[1248] "Means for achieving standardization of response skills" refers to functions for standardizing the response skills of operators.

[1249] An "application installed on a smartphone" is software that is installed on a smartphone and used by the user.

[1250] The system for carrying out this invention is configured as a claims handling support system. The system includes a server, a terminal used by the operator (such as a smartphone), and a generating AI model.

[1251] The server includes means of providing appropriate responses to complaints, preventing online crises, shortening response times, evaluating the quality of AI-generated responses, supporting operator skill development, maintaining motivation, and addressing challenges related to continued employment.

[1252] The terminals used by operators include an application installed on their smartphones. This application has the function of evaluating the quality of AI-generated responses in real time and providing feedback when operators handle customer complaints.

[1253] Specifically, when an operator handles a complaint, they input both the user's input and the response generated by the AI ​​into the application. The server uses a generative AI model (e.g., GPT-3) to evaluate the quality of the AI-generated response. The evaluation results are displayed as feedback on the operator's terminal. Based on this feedback, the operator can improve their skills.

[1254] The hardware used includes servers and operator smartphones. The software used includes Python, the OpenAI API, and smartphone applications.

[1255] As a concrete example, the following prompt statements can be used.

[1256] Example of a prompt:

[1257] User input: How do I return an item?

[1258] AI response: To return an item, first log in to your account and select the item you wish to return from your order history. Then, select the reason for the return and complete the return process.

[1259] Please rate the quality of this response.

[1260] Using this prompt, the server evaluates the quality of the response using a generative AI model and provides feedback to the operator. This allows the operator to improve their skills, maintain motivation, and address challenges in retaining employment.

[1261] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[1262] Step 1:

[1263] When a user handles a complaint, they input both their own information and the AI-generated response into the terminal. The input data consists of the user's complaint and the AI's response text.

[1264] Step 2:

[1265] The terminal sends the user's entered complaint details and the AI's response text to the server. The transmitted data consists of the user's input text and the AI's response text.

[1266] Step 3:

[1267] The server sends the received user input text and the AI's response text as prompts to the generating AI model. An example of a prompt is: "User input: How do I return an item? AI response: To return an item, first log in to your account and select the item you wish to return from your order history. Then, select a reason for return and complete the return process. Please rate the quality of this response."

[1268] Step 4:

[1269] The generative AI model evaluates the quality of the AI's response based on the prompt text. The evaluation result is feedback text regarding the quality of the response.

[1270] Step 5:

[1271] The server sends the evaluation results received from the generated AI model to the terminal. The transmitted data is feedback text regarding the quality of the response.

[1272] Step 6:

[1273] The terminal displays the received evaluation results to the user. The displayed data is feedback text regarding the quality of the response. The user can use this feedback to improve their skills.

[1274] Through the above processing steps, users can evaluate the quality of AI-generated responses during complaint handling in real time and receive feedback. This enables users to improve their skills and maintain their motivation, thus solving the challenge of retaining employees.

[1275] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1276] "Example of form 1"

[1277] One embodiment of the present invention provides a complaint handling support system that incorporates an emotion engine. When a customer service representative or operator receives a complaint, this system uses an emotion engine to recognize the user's emotions and analyze the user's emotional state. Based on the analysis results, the system generates an appropriate response to prevent further escalation. Specifically, if the system senses that the user is angry or dissatisfied, it generates a response that offers an apology or a solution. If the system senses that the user is confused, it generates a response that provides a clear explanation or guidance. As a result, customer service representatives and operators can quickly provide the most appropriate response according to the user's emotional state, shortening response times and improving service quality.

[1278] "Example of form 2"

[1279] Another embodiment of the present invention provides a complaint handling support system that incorporates an emotion engine. When a customer service representative or operator receives a complaint, this system uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state. Based on the analysis results, the system generates an appropriate response to prevent further escalation of the complaint. Furthermore, this system provides customer service representatives and operators with a sense of security and standardizes their response skills. Specifically, the system provides the optimal response according to the user's emotional state, enabling customer service representatives and operators to respond with confidence. In addition, the responses provided by the system eliminate differences in response skills among customer service representatives and operators, maintaining consistency in service quality.

[1280] "Example of form 3"

[1281] Furthermore, as another embodiment of the present invention, a complaint handling support system combined with an emotion engine is provided. When a customer service representative or operator receives a complaint, this system uses an emotion engine that recognizes the user's emotions to analyze the user's emotional state. Based on the analysis results, the system generates an appropriate response to prevent further escalation. In addition, this system provides a sense of security to customer service representatives and operators and helps to standardize their response skills. Moreover, this system helps to solve the challenge of retaining employees. Specifically, the responses and feedback provided by the system help to improve the skills and maintain the motivation of customer service representatives and operators. This reduces the turnover rate of customer service representatives and operators and promotes continued employment.

[1282] The following describes the processing flow for each example of the form.

[1283] "Example of form 1"

[1284] Step 1: The customer service representative or operator receives the complaint.

[1285] Step 2: The system uses an emotion engine to analyze the user's emotional state.

[1286] Step 3: Based on the analysis results, the system generates an appropriate response.

[1287] Step 4: The customer service representative or operator uses the generated response to assist the user.

[1288] "Example of form 2"

[1289] Step 1: A customer service representative or operator receives the complaint.

[1290] Step 2: The system uses an emotion engine to analyze the user's emotional state.

[1291] Step 3: Based on the analysis results, the system generates an appropriate response.

[1292] Step 4: The customer service representative or operator uses the generated response to assist the user.

[1293] Step 5: Customer service staff and operators use the feedback received from the system to improve their customer service skills.

[1294] "Example of form 3"

[1295] Step 1: A customer service representative or operator receives the complaint.

[1296] Step 2: The system uses an emotion engine to analyze the user's emotional state.

[1297] Step 3: Based on the analysis results, the system generates an appropriate response.

[1298] Step 4: The customer service representative or operator uses the generated response to assist the user.

[1299] Step 5: Customer service staff and operators use the feedback received from the system to improve their customer service skills.

[1300] Step 6: Customer service staff and operators use feedback from the system to maintain their motivation and reduce employee turnover.

[1301] (Example 1)

[1302] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[1303] In handling customer complaints, it is difficult for customer service representatives and operators to provide appropriate responses quickly. Furthermore, there is a high risk of complaints escalating and becoming a major issue, and prolonged handling times lead to decreased customer satisfaction. Additionally, there is variability in the handling skills of individual representatives, resulting in significant mental strain. To address these challenges, a system is needed that accurately analyzes the content of complaints, understands the user's emotional state, and quickly generates appropriate responses.

[1304] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1305] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of the issue, means for shortening response times, means for analyzing the content of the complaint, means for analyzing the user's emotional state, means for generating an appropriate response, and means for displaying the generated response. This enables improved efficiency and quality in complaint handling.

[1306] A "complaint handling support system" is a system designed to provide appropriate responses to complaints, prevent them from escalating, and shorten response times.

[1307] "Means of providing appropriate responses" refers to a function that analyzes the content of a complaint and generates the optimal response based on the user's emotional state.

[1308] "Means to prevent online outrage" refer to functions designed to prevent complaints from escalating and becoming major problems.

[1309] "Means of shortening response times" refers to a function that quickly generates responses to complaints, enabling customer service staff and operators to respond promptly.

[1310] "Means for analyzing the content of a claim" refers to a function that analyzes the text data of a claim using natural language processing technology to understand the content of the claim.

[1311] "Means for analyzing the user's emotional state" refers to a function that extracts emotions from the user's complaint text and identifies emotional states such as anger, dissatisfaction, and confusion.

[1312] "Means for generating appropriate responses" refers to a function that generates the optimal response based on the user's emotional state, using analysis results.

[1313] "Means for displaying generated responses" refers to a function that sends the response generated from the server to a terminal and displays it so that the counter staff or operator can confirm it.

[1314] This invention relates to a complaint handling support system for customer service staff and operators. This system aims to provide appropriate responses to complaints, prevent escalation, and shorten response times. Specific embodiments of this system are described below.

[1315] Hardware and software to be used

[1316] server

[1317] The server plays a central role in the claims handling support system. The server uses the following software and hardware:

[1318] Hardware: Typical cloud servers (e.g., AWS EC2 instances)

[1319] software:

[1320] Automatic response generation module (e.g., AI model generation)

[1321] Emotion engine (e.g., natural language processing engine)

[1322] Database (e.g., SQL database)

[1323] terminal

[1324] Terminals are devices used by counter staff and operators. These terminals utilize the following hardware and software:

[1325] Hardware: A typical personal computer (e.g., Windows PC, iPad)

[1326] software:

[1327] Claims input interface

[1328] Response display interface

[1329] Program processing

[1330] Entering and submitting a claim

[1331] When a user enters a complaint, the device sends this information to the server. For example, if a user enters "The product hasn't arrived," the device packages this text data in JSON format and sends it to the server using the HTTPS protocol.

[1332] Claim Analysis

[1333] The server analyzes the text data of the received complaint. Specifically, it uses a generative AI model to perform natural language processing (NLP) and understand the content of the complaint. For example, in response to a complaint that "the product did not arrive," the server classifies it as a "delivery-related problem."

[1334] Analysis of emotional states

[1335] The server uses an emotion engine to analyze the user's emotional state. It extracts emotions such as anger, frustration, and confusion from the text of the complaint. For example, in response to a complaint that "the product did not arrive," the server identifies the emotion of "frustration."

[1336] Response generation

[1337] The server generates an appropriate response based on the analysis results. Using a generative AI model, it generates responses that correspond to the user's emotional state. For example, it might generate a response such as, "We apologize. We will confirm your order. Please wait a moment."

[1338] Sending and displaying responses

[1339] The server sends the generated response to the terminal. The terminal displays the response received from the server on its screen. The customer service representative or operator reviews this response and takes appropriate action for the user.

[1340] Specific example

[1341] For example, if a user enters a complaint stating "the product hasn't arrived," the terminal sends this information to the server. The server analyzes the complaint using a generative AI model and identifies the user's emotional state using an emotion engine. Based on the analysis, the server generates a response such as "We apologize. We will check on your order. Please wait a moment," and sends it to the terminal. The customer service representative then provides this response to the user and works to resolve the issue.

[1342] Example of a prompt

[1343] "Please explain how the server generates a response when a user submits a complaint stating that 'the product has not arrived.'"

[1344] In this way, this invention improves the efficiency of claim handling and enhances quality.

[1345] The flow of the specific processing in Example 1 will be explained using Figure 17.

[1346] Step 1:

[1347] The user enters a complaint.

[1348] The user enters a complaint through the interface of the complaint handling support system. For example, they might enter "The product did not arrive." The entered complaint is sent to the terminal as text data. The input data is text containing the content of the complaint.

[1349] Step 2:

[1350] The device sends the claim to the server.

[1351] The terminal sends user-entered claims to the server in real time. Specifically, it packages the claim text data in JSON format and sends it to the server using the HTTPS protocol. The input data is the claim text data, and the output data is the JSON format data sent to the server.

[1352] Step 3:

[1353] The server analyzes the content of the claim.

[1354] The server analyzes the text data of the received complaint. Specifically, it uses a generative AI model to perform natural language processing (NLP) and understand the content of the complaint. For example, in response to a complaint that "the product did not arrive," the server classifies it as a "delivery-related problem." The input data is the text data of the complaint, and the output data is the classification information of the complaint as a result of the analysis.

[1355] Step 4:

[1356] The server analyzes the user's emotional state.

[1357] The server uses an emotion engine to analyze the user's emotional state. It extracts emotions such as anger, frustration, and confusion from the text of the complaint. For example, in response to a complaint that "the product did not arrive," the server identifies the emotion "frustration." The input data is the text data of the complaint, and the output data is the result of the emotion analysis.

[1358] Step 5:

[1359] The server generates an appropriate response.

[1360] The server generates an appropriate response based on the analysis results. Using a generative AI model, it generates responses that correspond to the user's emotional state. For example, it might generate a response such as, "We apologize. We will confirm your order. Please wait a moment." The input data consists of complaint classification information and sentiment analysis results, while the output data is the generated response text.

[1361] Step 6:

[1362] The server sends a response to the terminal.

[1363] The server sends the generated response to the terminal. Specifically, it packages the response text data in JSON format and sends it to the terminal using the HTTPS protocol. The input data is the generated response text, and the output data is the JSON data sent to the terminal.

[1364] Step 7:

[1365] The device displays a response.

[1366] The terminal displays the response received from the server on the screen. The customer service representative or operator reviews this response and takes appropriate action for the user. The input data is the response text sent from the server, and the output data is the response displayed on the screen.

[1367] Step 8:

[1368] The user receives a response.

[1369] The user receives a response from a customer service representative or operator. This allows the user to know the next steps toward resolving the problem. The input data is the response provided by the customer service representative or operator, and the output data is the response received by the user.

[1370] (Application Example 1)

[1371] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[1372] Traditional complaint handling systems faced challenges in accurately understanding customers' emotional states and providing prompt and appropriate responses. Furthermore, the inability to predict complaint escalations and take preventative measures led to delays and a risk of decreased customer satisfaction. Additionally, insufficient management of response history meant that past records could not be used to improve future responses.

[1373] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response time, means for analyzing the user's emotional state using an emotion engine, means for generating an appropriate response based on the analysis results, means for predicting the possibility of escalation, and means for managing the complaint handling history. This makes it possible to appropriately grasp the customer's emotional state and respond quickly and appropriately. Furthermore, since complaint escalation can be predicted and countermeasures can be taken in advance, it is expected that customer satisfaction will improve without delays in response. In addition, the management of response history is enhanced, and past response history can be utilized to help with future responses.

[1374] A "customer service representative" is an employee whose role is to directly receive and handle customer inquiries and complaints.

[1375] An "operator" is an employee whose role is to operate systems or machines and perform specific tasks.

[1376] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[1377] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints to customers.

[1378] "Means to prevent online firestorms" are functions designed to prevent complaints from escalating and problems from becoming serious.

[1379] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1380] An "emotion engine" is a technology used to analyze a user's emotional state.

[1381] "Means for analyzing the user's emotional state" refers to a function that uses an emotion engine to analyze the user's emotions.

[1382] "Means for generating appropriate responses" refers to a function that generates appropriate responses based on the results of emotion analysis.

[1383] "Means for predicting the possibility of escalation" refers to a function that predicts the likelihood of a problem becoming more serious based on the content of the complaint and the emotional state of the person making it.

[1384] "Means for managing complaint handling history" refers to a function for recording and managing past complaint handling history.

[1385] As an example of how to implement this invention, a customer complaint handling support system for an e-commerce site will be used. This system operates through the cooperation of three parties: a server, a terminal, and a user.

[1386] System Configuration

[1387] 1. Server:

[1388] Emotion Engine: This is a software module for analyzing a user's emotional state. Specifically, it uses Python's emotion analysis libraries (e.g., TextBlob or VADER).

[1389] Generative AI Model: This is an AI model designed to generate appropriate responses to claims. Specifically, it uses generative AI models such as GPT-3.

[1390] Escalation Prediction Module: This is a machine learning model for predicting the likelihood of escalation based on the content and emotional state of a complaint. Specifically, it uses machine learning libraries such as scikit-learn.

[1391] History Management System: This is a database system for managing the history of customer complaint handling. Specifically, it uses database management systems such as MySQL or PostgreSQL.

[1392] 2. Terminal:

[1393] Smartphone app: This is an application for users to input complaints and communicate with a server. The application provides a user interface for inputting complaints, displaying sentiment analysis results, and receiving responses.

[1394] 3. User:

[1395] Customer: Responsible for entering complaints and receiving responses from the system.

[1396] System operation

[1397] 1. Complaint Reception:

[1398] Users submit complaints through a smartphone app. For example, they might enter, "I am very dissatisfied because the product has not arrived!"

[1399] 2. Emotion analysis:

[1400] The server's emotion engine analyzes the complaint text to determine the user's emotional state. For example, it might detect "anger" from the submitted complaint.

[1401] 3. Automated response generation:

[1402] The server's AI model generates an appropriate response based on the sentiment analysis results. For example, it might generate a response such as, "We apologize for the inconvenience. We will investigate immediately and provide you with a solution."

[1403] 4. Escalation prediction:

[1404] The server's escalation prediction module predicts the risk of escalation based on the content and emotional state of the complaint. For example, it might determine it to be "high risk."

[1405] 5. History management:

[1406] The server's history management system stores complaints and their response history in a database. This allows for future support and management.

[1407] Specific examples and prompt statements

[1408] Specific example:

[1409] Complaint: "I am extremely dissatisfied because the product has not arrived!"

[1410] Emotion analysis result: Anger

[1411] Automated response: "We apologize for the inconvenience. We will investigate immediately and provide you with a solution."

[1412] Escalation risk: High

[1413] Example of a prompt:

[1414] Customer complaint: "I am extremely dissatisfied because the product has not arrived!"

[1415] Emotion analysis result: Anger

[1416] Please generate an appropriate response.

[1417] In this way, the server, terminal, and user work together to accurately understand the customer's emotional state and respond quickly and appropriately. Furthermore, it becomes possible to predict complaint escalations and take preventative measures, thus preventing delays in responses and improving customer satisfaction. In addition, the management of response history is enhanced, allowing past response records to be used to improve future responses.

[1418] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[1419] Step 1:

[1420] A user submits a complaint through a smartphone app. The entered complaint text is sent to the server. For example, a complaint might read, "I am very dissatisfied because the product has not arrived!"

[1421] Step 2:

[1422] The server's sentiment engine receives the claim text and performs sentiment analysis. It analyzes the input claim text and determines the user's emotional state. For example, it uses a sentiment analysis library (e.g., TextBlob or VADER) to detect the emotion "anger." The output is the emotional state (e.g., anger).

[1423] Step 3:

[1424] The server's generative AI model receives the sentiment analysis results and generates an appropriate response. Taking the emotional state (e.g., anger) and complaint text as input, the generative AI model (e.g., GPT-3) is given a prompt to generate a response. For example, a response such as "We apologize for the inconvenience. We will investigate immediately and provide you with a solution" might be generated. The output is the generated response.

[1425] Step 4:

[1426] The server's escalation prediction module receives the claim text and sentiment state and predicts the risk of escalation. Using the claim text and sentiment state as input, a machine learning model (e.g., scikit-learn) is used to predict the escalation risk. For example, it might be judged as "high risk." The output is the escalation risk.

[1427] Step 5:

[1428] The server's history management system stores claims and their response history in a database. Claim text, emotional state, generated response, and escalation risk are taken as input and stored in the database management system (e.g., MySQL or PostgreSQL). The output is the stored history data.

[1429] Step 6:

[1430] The server sends the generated response to the terminal. The generated response is then sent to the user's smartphone app as input. For example, the user might see a response such as, "We apologize for the inconvenience. We will investigate immediately and provide you with a solution." The output is the response displayed to the user.

[1431] Step 7:

[1432] The user reviews the response via a smartphone app and enters any additional complaints or feedback as needed. If the user reviews the response and is satisfied, the complaint handling is complete. If there are additional complaints or feedback, the process is repeated from step 1. The output includes the user's satisfaction level and any additional complaints.

[1433] (Example 2)

[1434] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[1435] Traditional complaint handling systems make it difficult for customer service representatives and operators to provide appropriate responses to customer inquiries, resulting in significant mental strain. Furthermore, inconsistencies in response skills and a lack of consistency in service quality are major challenges. Additionally, while complaint handling requires accurately understanding customer emotions and preventing escalation, there is a lack of effective means to achieve this.

[1436] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1437] In this invention, the server includes means for generating an appropriate response to an inquiry, means for calculating a confidence score for the generated response, means for displaying the confidence score, means for analyzing the user's emotional state, means for generating an appropriate response based on the analysis results, means for providing a sense of security, and means for achieving standardization of response skills. This makes it possible for customer service representatives and operators to provide consistent, high-quality responses to customers while gaining a sense of security.

[1438] "Means for generating appropriate responses to inquiries" refers to technologies that automatically generate appropriate answers based on the content of inquiries from users.

[1439] "Means for calculating a confidence score for a generated response" refers to techniques for evaluating and quantifying the accuracy and reliability of a generated response.

[1440] "Means for displaying confidence scores" refers to technologies for visually displaying calculated confidence scores to users.

[1441] "Means for analyzing a user's emotional state" refers to technologies for reading and analyzing a user's emotions from their words and actions.

[1442] "Means for generating appropriate responses based on analysis results" refers to technologies for generating optimal responses for users based on the results of sentiment analysis.

[1443] "Means of providing psychological reassurance" refers to technologies that enable users to feel safe and confident when using a system.

[1444] "Means for achieving standardization of response skills" refers to technologies that reduce variations in response skills among different users and achieve consistent response quality.

[1445] This invention is a system for enabling customer service representatives and operators to appropriately handle customer inquiries and complaints. This system uses a generative AI model to generate responses to inquiries, calculates and displays a confidence score, thereby providing users with a sense of security and standardizing response skills. Furthermore, it improves the quality of complaint handling by analyzing the user's emotional state using an emotion engine and generating appropriate responses.

[1446] Hardware and software to be used

[1447] Server: Provides computing resources for running generated AI models (e.g., OpenAI's GPT-4).

[1448] Terminal: The device the user operates on (e.g., PC, tablet, smartphone) displays the response sent from the server and the confidence score.

[1449] Emotion engine: Software used to analyze a user's emotional state (e.g., Microsoft's Azure Cognitive Services).

[1450] Program processing

[1451] 1. The user receives an inquiry from a customer. For example, the customer asks, "What is the warranty period for this product?"

[1452] 2. The server uses a generative AI model to input the prompt "What is the warranty period for this product?" and generates an appropriate response.

[1453] 3. The server calculates a confidence score for the generated response. For example, suppose the generated response is "This product has a one-year warranty" and the confidence score is 90%.

[1454] 4. The server sends the generated response and confidence score to the terminal.

[1455] 5. The device displays the received response and confidence score to the user. For example, it might display "Response: This product has a 1-year warranty (Confidence score: 90%)" on the screen.

[1456] 6. Users can review the displayed response and confidence score, and provide responses to customers with confidence.

[1457] Specific example

[1458] The user (operator) receives an inquiry from a customer.

[1459] The server uses a generative AI model to generate a response to the inquiry, "What is the warranty period for this product?"

[1460] The server calculates a confidence score for the generated response, "This product has a one-year warranty," and sends a message to the terminal indicating that the confidence score is 90%.

[1461] The device displays the response and confidence score to the user.

[1462] Users can check the confidence score and provide responses to customers with peace of mind.

[1463] Example of a prompt

[1464] "Generate an appropriate response to a customer inquiry. The inquiry is: 'What is the warranty period for this product?'"

[1465] In this way, this system utilizes AI technology to provide a sense of security and standardize response skills.

[1466] The flow of the specific processing in Example 2 will be explained using Figure 19.

[1467] Step 1:

[1468] The user receives an inquiry from a customer.

[1469] Input: Customer inquiry (e.g., "What is the warranty period for this product?")

[1470] Specific operation: The user (operator) receives customer inquiries via means such as phone or chat.

[1471] Step 2:

[1472] The server generates a response using a generated AI model.

[1473] Input: Inquiry received from the user (e.g., "What is the warranty period for this product?")

[1474] Data processing: Input the query content as a prompt message into a generative AI model (e.g., OpenAI's GPT-4).

[1475] Output: Generated response (e.g., "This product has a one-year warranty")

[1476] Specific operation: The server inputs the query details into a generation AI model and generates an appropriate response.

[1477] Step 3:

[1478] The server calculates a confidence score for the response it generates.

[1479] Input: Generated response (e.g., "This product has a one-year warranty")

[1480] Data processing: The response confidence score is calculated using an internal algorithm.

[1481] Output: Confidence score (e.g., 90%)

[1482] Specific operation: The server evaluates the accuracy and reliability of the generated response and calculates a confidence score.

[1483] Step 4:

[1484] The server sends the response and confidence score to the terminal.

[1485] Input: Generated response (e.g., "This product has a one-year warranty") and confidence score (e.g., 90%)

[1486] Data processing: Combine the response and confidence score into a single data packet.

[1487] Output: Data packets sent to the terminal

[1488] Specific operation: The server bundles the generated response and confidence score into a data packet and sends it to the terminal.

[1489] Step 5:

[1490] The device displays the response and confidence score to the user.

[1491] Input: Data packets sent from the server (response and confidence score)

[1492] Data processing: Analyze data packets and extract responses and confidence scores.

[1493] Output: The response and confidence score displayed to the user (e.g., "Response: This product has a 1-year warranty (Confidence score: 90%)")

[1494] Specific operation: The terminal analyzes the received data packets and displays the response and confidence score on the screen.

[1495] Step 6:

[1496] The user checks the displayed response and confidence score.

[1497] Input: The response and confidence score displayed on the device (e.g., "Response: This product has a 1-year warranty (Confidence score: 90%)")

[1498] Specific operation: The user (operator) checks the displayed response and confidence score to determine the reliability of the response.

[1499] Step 7:

[1500] The user provides a response to the customer.

[1501] Input: Confirmed response (e.g., "This product has a one-year warranty")

[1502] Specific actions: The user (operator) responds to the customer based on the confirmed response. For example, they might tell the customer, "The warranty period for this product is one year."

[1503] In this way, by performing specific actions at each step, the system achieves both the provision of psychological reassurance and the standardization of response skills.

[1504] (Application Example 2)

[1505] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[1506] Traditional customer service systems often failed to adequately understand the emotional state of users, leading to delays or inappropriate responses. Furthermore, the lack of a way to verify the reliability of AI-generated responses frequently caused anxiety among operators. Additionally, inconsistencies in response skills resulted in a lack of consistent service quality.

[1507] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1508] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of the issue, means for shortening response times, means for displaying a confidence score for AI-generated responses, means for analyzing the user's emotional state using an emotion engine, and means for generating appropriate responses based on the analysis results. This enables customer service representatives and operators to appropriately understand the user's emotional state and provide reliable responses quickly. Furthermore, by providing a sense of security and standardizing response skills, consistency in service quality can be maintained.

[1509] A "customer service representative" is an employee whose role is to directly receive and handle customer inquiries and complaints.

[1510] An "operator" is an employee whose role is to operate systems or machines and perform specific tasks.

[1511] A "customer complaint handling support system" is a system that provides support for appropriately responding to customer complaints.

[1512] "Means of providing appropriate responses" refers to the function of generating and providing appropriate answers to customer inquiries and complaints.

[1513] "Means to prevent online outrage" refer to functions that take measures to prevent customer dissatisfaction and anger from escalating.

[1514] "Means of shortening response times" refers to functions that enable quick responses to complaints and inquiries.

[1515] "A means of displaying a confidence score for AI-generated responses" refers to a function that quantifies and displays the reliability of AI-generated answers.

[1516] An "emotion engine" is software or hardware used to analyze a user's emotional state.

[1517] "Means for analyzing a user's emotional state" refers to a function that reads and analyzes a user's emotions from their words and actions.

[1518] "Means for generating appropriate responses based on analysis results" refers to a function that generates the optimal response based on the results of sentiment analysis.

[1519] "Means of providing a sense of security" refers to functions that provide support to enable operators to perform their duties with peace of mind.

[1520] "Means for achieving standardization of response skills" refers to a function that reduces variations in response skills among operators and maintains consistent service quality.

[1521] To implement this invention, the following hardware and software are required. The hardware requires a smartphone. The software requires Python, the OpenAI API, and the emotion_recognition library.

[1522] The server first receives user complaints and inquiries as input. Next, it uses the emotion_recognition library to analyze the user's emotional state. Based on this analysis, it uses the OpenAI API to generate an AI response to the user input. It obtains a confidence score for the generated response and displays it to the operator.

[1523] For example, if a user enters a complaint such as, "Recently, the security system has been malfunctioning, and I'm having trouble with it. What should I do?", the sentiment analysis result will be "anger." Based on this analysis result, the AI ​​will generate a response such as, "We apologize for the inconvenience. Could you please try restarting the system?" and display a confidence score of 0.85.

[1524] An example of a prompt message is as follows:

[1525] Please enter your user complaint or inquiry: Recently, our security system has been malfunctioning, and I'm experiencing problems. What should I do?

[1526] In this way, the server can accurately understand the user's emotional state and provide reliable responses quickly. Furthermore, by providing a sense of security and standardizing response skills, it can maintain consistency in service quality.

[1527] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[1528] Step 1:

[1529] Users enter complaints or inquiries using their smartphones. The entered text is sent to the server. The input data consists of the content of the user's complaint or inquiry.

[1530] Step 2:

[1531] The server passes the received user input to the emotion_recognition library, which analyzes the user's emotional state. The input data is the text of the user's complaint or inquiry, and the output data is the analyzed emotional state (e.g., "anger," "sadness," "joy," etc.).

[1532] Step 3:

[1533] The server uses the OpenAI API to generate AI responses to user input based on the analyzed emotional state. The input data consists of the user's complaint or inquiry text and the analyzed emotional state, while the output data is the text of the generated AI response.

[1534] Step 4:

[1535] The server obtains a confidence score for the generated AI response. The input data is the text of the generated AI response, and the output data is the confidence score (e.g., 0.85).

[1536] Step 5:

[1537] The server displays the generated AI response and confidence score on the operator's smartphone. The input data is the text of the generated AI response and its confidence score, while the output data is the information displayed on the operator's smartphone.

[1538] Step 6:

[1539] The operator reviews the displayed AI response and confidence score, and takes appropriate action for the user as needed. The input data is the text and confidence score of the AI ​​response reviewed by the operator, and the output data is the final response provided by the operator to the user.

[1540] In this way, the server can accurately understand the user's emotional state and provide reliable responses quickly. Furthermore, by providing a sense of security and standardizing response skills, it can maintain consistency in service quality.

[1541] (Example 3)

[1542] Next, we will describe Embodiment 3 of Embodiment Example 3. 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".

[1543] Traditional customer complaint handling systems often resulted in inconsistent quality of service due to the difficulty of call center staff and operators providing appropriate responses. Furthermore, they struggled to accurately understand user emotions during complaint handling, making it difficult to prevent escalations. Additionally, insufficient skill development and motivation maintenance for operators led to employment retention challenges. To address these issues, a system is needed that improves the quality of complaint handling, shortens response times, and supports operator skill development and motivation maintenance.

[1544] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1545] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation of complaints, means for shortening response times, means for analyzing the user's emotional state using an emotion analysis engine, means for generating appropriate responses using a generative AI model, means for evaluating the quality of the generated responses, means for providing feedback to operators based on the evaluation results, and means for supporting the skill development of operators. This makes it possible to improve the quality of complaint handling, shorten response times, and support the skill development and motivation maintenance of operators.

[1546] A "customer service representative" is an employee whose role is to directly receive and handle customer inquiries and complaints.

[1547] An "operator" is a person responsible for operating the system and handling customer inquiries and complaints.

[1548] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[1549] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to customer complaints.

[1550] "Means to prevent online firestorms" refer to functions that appropriately handle customer dissatisfaction and anger, and prevent problems from escalating.

[1551] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1552] An "emotion analysis engine" is software or hardware that analyzes a customer's emotional state and provides the results.

[1553] A "generative AI model" is an artificial intelligence model that generates an appropriate response based on a given prompt.

[1554] "Means for evaluating the quality of responses" refers to functions for evaluating the appropriateness and effectiveness of the generated responses.

[1555] "Means of providing feedback" refers to a function that provides operators with suggestions for improvement and advice based on the results of the response quality evaluation.

[1556] "Means of supporting skill development" refers to the function of providing training and educational programs to improve operators' response skills.

[1557] "Means of providing psychological reassurance" refers to functions that reduce the stress and anxiety that operators experience while handling complaints, and provide them with a sense of security.

[1558] "Means for achieving standardization of response skills" refers to functions that reduce variations in response skills among operators and maintain a consistent level of quality.

[1559] "Means to support the resolution of employment retention challenges" refer to functions that reduce the turnover rate of operators and support long-term employment retention.

[1560] This invention is a support system for counter staff and operators to appropriately handle customer complaints. The aim of this system is to improve the quality of complaint handling, reduce response times, and support the skill development and motivation maintenance of operators.

[1561] Hardware and software to be used

[1562] Hardware: Servers, terminals (PCs, tablets, smartphones)

[1563] Software: Generative AI models (e.g., GPT-4), sentiment analysis engines (e.g., Affectiva), database management systems (e.g., MySQL)

[1564] Specific operation of the system

[1565] Receiving a complaint

[1566] User: Customer service representatives and operators receive customer complaints.

[1567] Terminal: The customer service representative or operator enters the details of the complaint into the terminal. For example, if a customer complains that "the product has not arrived," they would enter the details in text format.

[1568] Emotion analysis

[1569] Terminal: Sends the entered claim details to the server.

[1570] Server: Receives the complaint details and sends them to the sentiment analysis engine.

[1571] Emotion analysis engine: Analyzes the content of a complaint to identify the user's emotional state. For example, it might be analyzed as "anger."

[1572] Server: Receives analysis results from the emotion analysis engine and proceeds to the next processing step.

[1573] Response generation

[1574] Server: Based on the sentiment analysis results, it sends prompt messages to the generative AI model. For example, it might send a prompt message such as, "The customer is angry because their product hasn't arrived. Generate an appropriate response."

[1575] Generative AI model: Generates appropriate responses based on prompt text. For example, it can generate a response such as, "We are very sorry, customer. We will investigate immediately and address the issue as soon as possible."

[1576] Server: Receives the generated response and proceeds to the next processing step.

[1577] Response quality evaluation

[1578] Server: Executes algorithms to evaluate the quality of the generated response. For example, it evaluates the appropriateness of the response and its impact on customer satisfaction.

[1579] Server: Saves evaluation results to a database and generates feedback.

[1580] Feedback and skill development support

[1581] Server: Based on the evaluation results, it generates feedback for the operator. For example, it might generate feedback such as, "The response was appropriate this time, but it would be even better if you could provide a more specific solution."

[1582] Terminal: Operators receive feedback and participate in training programs to improve their skills. For example, they may conduct simulation training to improve the quality of their responses.

[1583] Examples of specific cases and prompt statements

[1584] Specific example: When a customer service representative receives a complaint from a customer that "the product has not arrived."

[1585] Terminal: The counter staff member enters the details of the complaint.

[1586] Server: Sends the complaint details to the sentiment analysis engine.

[1587] Emotion analysis engine: Analyzes the customer's emotional state as "anger".

[1588] Server: Sends a prompt message to the AI ​​model stating, "The customer is angry because the product has not arrived."

[1589] Generative AI model: Generates the response, "We are very sorry, customer. We will investigate immediately and take immediate action."

[1590] Server: Evaluates the quality of the response and provides feedback.

[1591] Example of a prompt:

[1592] "The customer is angry because their product hasn't arrived. Please generate an appropriate response."

[1593] "The customer is dissatisfied with the quality of service. Please generate an appropriate response."

[1594] This system helps improve operators' skills and maintain their motivation, thus resolving the challenge of continued employment. The flow of a specific process in Example 3 will be explained using Figure 21.

[1595] Step 1: Receiving the complaint

[1596] User: A customer service representative or operator receives a complaint from a customer. For example, a customer complains that "the product hasn't arrived."

[1597] Terminal: The customer service representative or operator enters the details of the complaint into the terminal. The entered complaint details are saved in text format.

[1598] Input: Customer complaint details (e.g., "The product did not arrive").

[1599] Output: Claims data in text format.

[1600] Step 2: Emotion Analysis

[1601] Terminal: Sends the entered claim details to the server.

[1602] Server: Receives the complaint details and sends them to the sentiment analysis engine.

[1603] Emotion analysis engine: Analyzes the content of a complaint to identify the user's emotional state. For example, it might be analyzed as "anger."

[1604] Input: Claim data in text format.

[1605] Output: Emotion analysis results (e.g., "anger").

[1606] Step 3: Response Generation

[1607] Server: Based on the sentiment analysis results, it sends prompt messages to the generative AI model. For example, it might send a prompt message such as, "The customer is angry because their product hasn't arrived. Generate an appropriate response."

[1608] Generative AI model: Generates appropriate responses based on prompt text. For example, it can generate a response such as, "We are very sorry, customer. We will investigate immediately and address the issue as soon as possible."

[1609] Input: Sentiment analysis results and prompt text.

[1610] Output: The generated response.

[1611] Step 4: Evaluate the quality of the response

[1612] Server: Executes algorithms to evaluate the quality of the generated response. For example, it evaluates the appropriateness of the response and its impact on customer satisfaction.

[1613] Server: Saves evaluation results to a database and generates feedback.

[1614] Input: The generated response.

[1615] Output: Response quality evaluation results and feedback.

[1616] Step 5: Feedback and Skill Development Support

[1617] Server: Based on the evaluation results, it generates feedback for the operator. For example, it might generate feedback such as, "The response was appropriate this time, but it would be even better if you could provide a more specific solution."

[1618] Terminal: Operators receive feedback and participate in training programs to improve their skills. For example, they may conduct simulation training to improve the quality of their responses.

[1619] Input: Response quality evaluation results and feedback.

[1620] Output: Operator skill development and implementation of training programs.

[1621] (Application Example 3)

[1622] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[1623] Traditional customer complaint handling systems often lacked sufficient support for call center staff and operators to properly address complaints, resulting in inconsistent quality of service. Furthermore, the lack of functionality to accurately analyze user emotions and generate responses accordingly led to inappropriate complaint handling, sometimes resulting in escalating customer complaints. Additionally, the systems placed a significant mental burden on operators, posing challenges to their continued employment. To address these issues, a system is needed that analyzes user emotions and generates appropriate responses.

[1624] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[1625] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response times, means for analyzing user emotions, means for generating appropriate responses based on the analysis results, means for evaluating the quality of the generated responses, means for providing feedback on the evaluation results, means for providing a sense of security, means for standardizing response skills, and means for supporting the resolution of employment retention issues. As a result, customer service staff and operators can provide appropriate responses in accordance with user emotions, improving the quality of complaint handling, preventing escalation, reducing mental burden, and promoting employment retention.

[1626] A "customer service representative" is an employee whose role is to directly receive and handle inquiries and complaints from customers and users.

[1627] An "operator" is an employee whose role is to operate systems or machines and perform specific tasks.

[1628] A "complaint handling support system" is a system designed to assist in appropriately responding to complaints from customers and users.

[1629] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints.

[1630] "Means to prevent online backlash" refer to functions designed to prevent problems from escalating due to inappropriate handling of complaints.

[1631] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1632] "Means for analyzing user emotions" refers to functions that analyze user emotions from their statements and actions.

[1633] "Means for generating appropriate responses" refers to the function of generating appropriate responses based on analyzed emotions.

[1634] "Means for evaluating the quality of the generated response" refers to a function for evaluating the quality of the generated response.

[1635] "Means for providing feedback on evaluation results" refers to a function for providing feedback to the operator on the quality of the evaluated response.

[1636] "Means of providing psychological reassurance" refers to functions that provide psychological reassurance to operators.

[1637] "Means for achieving standardization of response skills" refers to functions for standardizing the response skills of operators.

[1638] "Means to support the resolution of employment retention challenges" refers to functions that support operators' skill development and motivation maintenance, thereby resolving employment retention challenges.

[1639] In order to implement this invention, it is necessary to build a claims handling support system. This system will help customer service staff and operators to respond to claims appropriately and will use the following hardware and software.

[1640] hardware

[1641] Server: Plays a central role in the customer complaint handling support system. It processes and analyzes data and generates responses.

[1642] Terminal: A smartphone or computer used by a customer service representative or operator. It receives user input and displays responses from the server.

[1643] software

[1644] OpenAI API: Provides generative AI models used for response generation.

[1645] EmotionEngine: A library for analyzing user emotions.

[1646] Database: A system for storing claim data and response data.

[1647] Processing flow

[1648] 1. Receiving user input

[1649] Users enter complaints through their devices. For example, they might say, "My security alarm keeps going off and it's causing me problems."

[1650] 2. Emotion analysis

[1651] The server uses EmotionEngine to analyze emotions from user input. For example, it might recognize the emotion "confusion" from user input.

[1652] 3. Response generation

[1653] The server uses the OpenAI API to generate an appropriate response based on the analyzed emotions. An example of a prompt is as follows:

[1654] User's emotion: Confusion

[1655] User input: I'm having trouble with a security alarm that keeps going off.

[1656] Please generate an appropriate response.

[1657] 4. Response Quality Evaluation

[1658] The server evaluates the quality of the generated response. For example, it checks whether it contains keywords such as "thank you" or "gratitude."

[1659] 5. Provide feedback

[1660] The server provides feedback on the evaluation results to the operator, allowing them to improve their response skills.

[1661] Specific example

[1662] If a user enters "I'm having trouble because my security alarm keeps going off," the server uses EmotionEngine to analyze the emotion "confusion." Then, it uses the OpenAI API to generate a response like the following:

[1663] "It seems you're having trouble. We'll address the security alarm issue immediately. Please wait a moment."

[1664] By evaluating the quality of the generated responses and providing feedback to the operators, it is possible to improve the operators' response skills.

[1665] In this way, customer service representatives and operators can respond appropriately to users' emotions, improving the quality of complaint handling, preventing escalations, reducing mental stress, and promoting continued employment.

[1666] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[1667] Step 1:

[1668] The user enters a complaint via their device. The entered complaint might be something like, "I'm having trouble because the security alarm keeps going off." This input data is sent to the server.

[1669] Step 2:

[1670] The server passes the user's input data to the EmotionEngine, which analyzes the user's emotions. The EmotionEngine analyzes the input data and recognizes, for example, an emotion such as "confusion." This analysis result is then returned to the server.

[1671] Step 3:

[1672] Based on the analyzed emotions, the server uses the OpenAI API to generate an appropriate response. The server generates a prompt message similar to the following and sends it to the OpenAI API.

[1673] User's emotion: Confusion

[1674] User input: I'm having trouble with a security alarm that keeps going off.

[1675] Please generate an appropriate response.

[1676] The OpenAI API generates a response based on this prompt and returns it to the server.

[1677] Step 4:

[1678] The server evaluates the quality of the generated response. For example, it checks whether the response contains keywords such as "thank you" or "gratitude." This evaluation result is stored within the server.

[1679] Step 5:

[1680] The server provides feedback on the evaluation results to the operator. This feedback is displayed to the operator via a terminal. This allows the operator to improve their response skills.

[1681] Step 6:

[1682] The operator will respond to the user appropriately based on the feedback provided by the server. For example, they might respond with, "It seems you're having trouble. We'll address the security alarm issue immediately. Please wait a moment."

[1683] In this way, customer service representatives and operators can respond appropriately to users' emotions, improving the quality of complaint handling, preventing escalations, reducing mental stress, and promoting continued employment.

[1684] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1685] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1686] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[1687] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1688] [Third Embodiment]

[1689] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1690] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1691] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1692] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1693] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1694] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1695] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1696] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1697] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1698] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1699] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1700] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[1701] "Example of form 1"

[1702] One embodiment of the present invention is a complaint handling support system for counter staff and operators. This system incorporates an AI (artificial intelligence)-based automatic response generation module as a means of providing appropriate responses to complaints. This module analyzes the content of the complaint and generates an appropriate response. Furthermore, it includes a function to predict complaint escalation as a means of preventing further escalation. In addition, it employs an optimization algorithm to speed up response generation as a means of shortening response time. "Embodiment Example 2"

[1703] As an example of the present invention, there is a system that further includes means for providing a sense of security and means for achieving standardization of response skills. As means for providing a sense of security, the system has a function to display a confidence score for the response generated by the AI. This allows the operator to confirm the reliability of the response proposed by the AI ​​and gain a sense of security. Furthermore, as means for achieving standardization of response skills, the system allows operators to refer to the responses generated by the AI, thereby reducing the variation in response skills among operators.

[1704] "Example of form 3"

[1705] As a third embodiment of the present invention, there is a system that further includes means to support the resolution of the challenge of continued employment. Specifically, it has a function to evaluate the quality of responses generated by AI and to support the improvement of operators' skills. This helps to improve operators' skills and maintain their motivation, thereby supporting the resolution of the challenge of continued employment.

[1706] The following describes the processing flow for each example of the form.

[1707] "Example of form 1"

[1708] Step 1: The customer service representative or operator receives the complaint.

[1709] Step 2: The system analyzes the content of the claim.

[1710] Step 3: An AI (artificial intelligence) automated response generation module generates an appropriate response.

[1711] Step 4: The system predicts the escalation of a claim.

[1712] Step 5: An optimization algorithm is activated to speed up response generation and reduce response time.

[1713] "Example of form 2"

[1714] Step 1: The customer service representative or operator reviews the response generated by the AI.

[1715] Step 2: The system displays a confidence score for the AI-generated response.

[1716] Step 3: The customer service representative or operator decides on the final response based on the confidence score.

[1717] Step 4: Customer service staff and operators use AI-generated responses as a reference to standardize their response skills.

[1718] "Example of form 3"

[1719] Step 1: The customer service representative or operator uses the AI-generated response.

[1720] Step 2: The system evaluates the quality of the response generated by the AI.

[1721] Step 3: The system provides feedback to help operators improve their skills.

[1722] Step 4: Customer service staff and operators use feedback to improve their skills and maintain their motivation.

[1723] (Example 1)

[1724] Next, we will describe Embodiment 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1725] Traditional complaint handling systems often lacked sufficient support for customer service representatives and operators to properly address complaints, resulting in inconsistencies in the quality and speed of responses. Furthermore, there was no way to predict the risk of complaint escalation, increasing the likelihood of escalation due to delays. Moreover, generating quick and appropriate responses was difficult amidst the pressure to shorten response times. A system is needed to address these challenges.

[1726] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1727] In this invention, the server includes means for providing appropriate responses to claims, means for preventing further escalation, means for shortening response times, means for analyzing the content of claims, means for generating automated responses, means for predicting escalation, and means for accelerating response generation. This improves the quality and speed of claim handling, reduces the risk of escalation, and enables the provision of quick and appropriate responses.

[1728] "Means of providing appropriate responses to complaints" refers to a function that generates and provides appropriate responses to complaints from users.

[1729] "Means to prevent further escalation of the issue" refers to functions designed to prevent complaints from escalating and the problem from spreading.

[1730] "Means for shortening response time" refers to functions that shorten response time by quickly generating and providing responses to complaints.

[1731] "Means for analyzing complaint content" refers to a function that analyzes the content of user complaints to identify the type of problem and the emotions involved.

[1732] "Means for generating automatic responses" refers to a function that automatically generates an appropriate response based on the analysis results.

[1733] "Means for predicting escalation" refers to functions for predicting the likelihood of a claim escalating and for assessing the associated risks.

[1734] "Means for speeding up response generation" refers to functions that optimize the response generation process and provide responses quickly.

[1735] This invention is a support system for customer service representatives and operators to appropriately handle customer complaints. The system includes multiple means for providing appropriate responses to complaints, preventing escalation, and shortening response times.

[1736] System Configuration

[1737] Hardware and software

[1738] Server: Analyzes claim details, generates automated responses, predicts escalations, and speeds up response generation.

[1739] Terminal: A device (such as a PC or smartphone) used by the user to enter a complaint and receive a response from the server.

[1740] Natural Language Processing (NLP) techniques: Used to analyze the content of the claims. Specifically, Google Cloud Natural Language API and IBM Watson Natural Language Understanding will be used.

[1741] Generative AI model: Used to generate automated responses. Specifically, it utilizes OpenAI's GPT-3.

[1742] Machine learning algorithms: Used to perform escalation prediction. Specifically, scikit-learn and TensorFlow are used.

[1743] Cache technology and parallel processing: Used to speed up response generation.

[1744] System operation

[1745] 1. Entering a complaint

[1746] The user enters the details of the complaint from their device.

[1747] Example: The user enters "The product quality is poor."

[1748] 2. Analysis of the complaint content

[1749] The server receives the complaint details sent by the user.

[1750] The server uses the Google Cloud Natural Language API to parse the claim content.

[1751] Specifically, the text of the complaint is sent to the API for sentiment analysis and keyword extraction.

[1752] Example: Extract the keyword "quality problem" from the text "The product quality is poor" and determine that the emotion is "negative".

[1753] 3. Generating automated responses

[1754] The server uses a generative AI model (GPT-3) based on the analysis results to generate an appropriate response.

[1755] Specifically, the analysis results are input as prompts into the AI ​​model, which then generates response sentences.

[1756] Example: Enter the prompt "The user has submitted a complaint that the product quality is poor. Please generate an appropriate response." and retrieve the generated response "We apologize. Could you please tell us more about the product quality?".

[1757] 4. Escalation prediction

[1758] The server uses machine learning algorithms to predict the likelihood of a claim escalating.

[1759] Specifically, we will use scikit-learn to build a model that assesses the risk of escalation based on past claim data, and then calculate a risk score for the current claim.

[1760] Example: A high risk score is assigned to a complaint that states, "I have contacted them multiple times, but the issue remains unresolved."

[1761] 5. Providing a response

[1762] The server provides the generated response to the user.

[1763] Specifically, the generated response message is sent to the terminal and displayed on the user's screen.

[1764] Example: The user's device will display the response, "We apologize. Could you please tell us more about the quality of the product?"

[1765] 6. Speeding up response generation

[1766] The server uses caching technology and parallel processing to speed up response generation.

[1767] Specifically, past responses to the same type of claim are stored in a cache and reused. Furthermore, processing speed is improved by executing multiple claim processing tasks in parallel.

[1768] Example: If a response to a complaint such as "The product quality is poor" is stored in the cache, retrieve the response from the cache and provide it quickly.

[1769] Specific example

[1770] A user submits a complaint stating that "the product quality is poor."

[1771] The server sends the complaint details to the Google Cloud Natural Language API, which extracts the keywords "quality issues" and the sentiment "negative."

[1772] The server inputs the prompt "The user has entered a complaint that the product quality is poor. Please generate an appropriate response." into the generating AI model (GPT-3) and generates the response "We apologize. Could you please tell us more about the product quality?"

[1773] The server uses scikit-learn to calculate the escalation risk score and assigns a high risk score to the case.

[1774] The server sends the generated response to the user's terminal and displays it on the user's screen.

[1775] The server stores responses to the same type of claim in a cache to speed up response generation in subsequent instances.

[1776] The above describes the embodiments for carrying out this invention.

[1777] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1778] Step 1:

[1779] Entering a complaint

[1780] The user enters the details of the complaint from their device.

[1781] Input: The content of the complaint entered by the user (e.g., "The product quality is poor").

[1782] Output: The claim details are sent to the server.

[1783] Specific action: The user enters the complaint details into the input form on the device and presses the submit button.

[1784] Step 2:

[1785] Received the details of the complaint

[1786] The server receives the complaint details sent by the user.

[1787] Input: The content of the complaint submitted by the user.

[1788] Output: The claim details are saved on the server.

[1789] Specific operation: The server receives an HTTP request and saves the claim details to the database.

[1790] Step 3:

[1791] Analysis of complaint content

[1792] The server uses the Google Cloud Natural Language API to parse the claim content.

[1793] Input: Saved claim details.

[1794] Output: Analysis results (keywords, sentiment analysis results, etc.).

[1795] Specific operation: The server sends the claim details to the API and retrieves the analysis results returned by the API.

[1796] Step 4:

[1797] Generating an automated response

[1798] The server uses a generative AI model (GPT-3) based on the analysis results to generate an appropriate response.

[1799] Input: Analysis results.

[1800] Output: The generated response.

[1801] Specific operation: The server inputs the analysis results as prompts into the AI ​​model and retrieves the generated response.

[1802] Example: Enter the prompt "The user has submitted a complaint that the product quality is poor. Please generate an appropriate response." and generate the response "We apologize. Could you please tell us more about the product quality?"

[1803] Step 5:

[1804] Escalation prediction

[1805] The server uses machine learning algorithms to predict the likelihood of a claim escalating.

[1806] Input: Claim details and analysis results.

[1807] Output: Escalation risk score.

[1808] Specific operation: The server uses a model built on past claim data to calculate a risk score for the current claim.

[1809] Example: A high risk score is assigned to a complaint that states, "I have contacted them multiple times, but the issue remains unresolved."

[1810] Step 6:

[1811] Providing a response

[1812] The server provides the generated response to the user.

[1813] Input: The generated response.

[1814] Output: The response message is displayed on the user's terminal.

[1815] Specific operation: The server sends the generated response message to the user's terminal and displays it on the user's screen.

[1816] Example: The user's device will display the response, "We apologize. Could you please tell us more about the quality of the product?"

[1817] Step 7:

[1818] Speeding up response generation

[1819] The server uses caching technology and parallel processing to speed up response generation.

[1820] Input: Past complaint details and responses.

[1821] Output: Cached response statement.

[1822] Specific operation: The server stores past responses to the same type of claim in a cache and reuses them. It also improves processing speed by executing multiple claim processing in parallel.

[1823] Example: If a response to a complaint such as "The product quality is poor" is stored in the cache, retrieve the response from the cache and provide it quickly.

[1824] (Application Example 1)

[1825] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1826] Traditional customer complaint handling systems had several problems: long response times and decreased customer satisfaction because customer service representatives and operators handled complaints manually. Furthermore, it was difficult to predict complaint escalation, making it impossible to notify superiors at the appropriate time, increasing the risk of complaints escalating. In addition, inconsistencies in handling skills and significant mental stress among staff also created challenges in retaining employees.

[1827] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1828] In this invention, the server includes means for providing appropriate responses to complaints, means for preventing further escalation, means for shortening response times, means for analyzing complaint content and generating appropriate responses, means for predicting complaint escalation, means for automatically notifying senior personnel, means for generating responses using a generation AI model, and means for generating responses using prompt sentences. This enables more efficient and faster complaint handling, leading to improved customer satisfaction and reduced risk of complaint escalation. Furthermore, by standardizing response skills and providing a sense of security, it can also solve the challenge of retaining employees.

[1829] A "customer service representative" is someone whose job is to directly handle customer inquiries and complaints.

[1830] An "operator" is someone whose job is to operate systems or machines and perform specific tasks.

[1831] A "customer complaint handling support system" is a system designed to assist in appropriately responding to customer complaints.

[1832] "Means of providing appropriate responses" refers to the function of generating and providing appropriate responses to complaints.

[1833] "Means to prevent online firestorms" are functions designed to prevent complaints from escalating and problems from becoming serious.

[1834] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1835] "Means for analyzing the content of a claim and generating an appropriate response" refers to a function that analyzes the content of a claim and generates an appropriate response based on that analysis.

[1836] "Methods for predicting complaint escalation" refer to functions that predict the likelihood of a complaint escalating to a higher-level employee.

[1837] The "automatic notification mechanism for senior personnel" is a function that automatically notifies senior personnel when a complaint escalates.

[1838] "Response generation means using a generative AI model" refers to a function that uses an artificial intelligence model to generate a response to a claim.

[1839] A "response generation means using a prompt statement" is a function for generating a response to a claim using a specific input statement (prompt statement).

[1840] As an example of how to implement this invention, we will describe a customer support application for an e-commerce site. This system can automatically generate appropriate responses to customer complaints and inquiries using AI and respond quickly. It also has a function to predict complaint escalation and automatically notify higher-level personnel as needed.

[1841] System Configuration

[1842] This system consists of the following main components:

[1843] 1. Claim Analysis Module: Analyzes the claim content and generates an appropriate response.

[1844] 2. Escalation prediction module: Predicts the escalation of claims.

[1845] 3. Notification Module: Automatically notifies senior personnel when an escalation is anticipated.

[1846] 4. Generative AI Model: An artificial intelligence model for generating responses to complaints.

[1847] 5. Prompt statement generation module: Generates a response using a specific input statement (prompt statement).

[1848] Hardware and software

[1849] This system uses internet-connected smartphones or PCs as hardware. The software utilizes Python and the OpenAI API.

[1850] Processing flow

[1851] 1. Entering the complaint details: The user enters the details of the complaint.

[1852] 2. Response Generation: The claim analysis module analyzes the claim content and generates an appropriate response using a generation AI model.

[1853] 3. Escalation Prediction: The escalation prediction module predicts the likelihood of a claim escalating.

[1854] 4. Notifications: If an escalation is anticipated, the notification module will automatically notify the higher-level responsible party.

[1855] Specific example

[1856] For example, if a customer enters a complaint stating, "I did not receive the product, so I would like a refund," the system will operate as follows:

[1857] 1. Entering the complaint details: The user enters, "I did not receive the product, so I would like a refund."

[1858] 2. Response Generation: The complaint analysis module analyzes this content and uses a generation AI model to generate a response such as, "We apologize. We will immediately investigate the issue of the product not being delivered and proceed with the refund process."

[1859] 3. Escalation Prediction: The escalation prediction module predicts that this claim has a "high" likelihood of escalation.

[1860] 4. Notification: The notification module automatically notifies the senior person in charge, who then takes over the responsibility.

[1861] Example of a prompt

[1862] Please enter your complaint details: I did not receive the item and would like a refund.

[1863] In this way, customer support for e-commerce sites can be streamlined, and customer satisfaction can be improved.

[1864] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1865] Step 1:

[1866] The user enters the details of their complaint.

[1867] Input: The user enters, "I did not receive the product, so I would like a refund."

[1868] Specific action: The user enters the details of their complaint in text format into an input form on their smartphone or PC.

[1869] Step 2:

[1870] The server receives the claim details and sends them to the claim analysis module.

[1871] Input: The complaint details entered by the user.

[1872] Output: Claim details sent to the claim analysis module.

[1873] Specific operation: The server receives input from the user and sends its contents to the claims analysis module.

[1874] Step 3:

[1875] The claim analysis module analyzes the claim content and generates an appropriate response using a generative AI model.

[1876] Input: Claim details sent to the claim analysis module.

[1877] Output: The generated response text.

[1878] Specific operation: The claim analysis module analyzes the claim content using natural language processing technology and sends it as a prompt to the generative AI model. The generative AI model generates an appropriate response and returns it to the claim analysis module.

[1879] Step 4:

[1880] The server sends the generated response to the user.

[1881] Input: The generated response text.

[1882] Output: The response text displayed to the user.

[1883] Specific operation: The server receives the generated response and sends it to the user's terminal. The user's terminal displays the response.

[1884] Step 5:

[1885] The escalation prediction module predicts the likelihood of a claim escalating.

[1886] Input: Complaint details.

[1887] Output: Escalation prediction result (high / low).

[1888] Specific operation: The escalation prediction module analyzes the complaint content and uses a generative AI model to predict the likelihood of escalation. The prediction result is returned to the server.

[1889] Step 6:

[1890] The server receives the escalation prediction results and automatically notifies higher-level personnel as needed.

[1891] Input: Escalation prediction result.

[1892] Output: Notification to the superior in charge.

[1893] Specific operation: The server receives the escalation prediction result, and if the prediction result is "high," it sends a notification to the higher-level person in charge. The notification is sent via email or a dedicated application.

[1894] In this way, the system can streamline and expedite complaint handling, thereby improving customer satisfaction and reducing the risk of complaints escalating.

[1895] (Example 2)

[1896] Next, we will describe Example 2 of the morphological example. 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."

[1897] Conventional complaint handling support systems lack sufficient support for operators to provide appropriate responses, and they lack means to provide a sense of security and standardize response skills. Furthermore, they lack effective means to shorten response times and prevent issues from escalating. As a result, operators face increased burdens and inconsistent response quality, which is a significant challenge.

[1898] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1899] In this invention, the server includes means for receiving inquiries, means for generating responses using a generative AI model, means for calculating a confidence score for the generated responses, means for displaying the responses and confidence scores, means for providing appropriate responses to complaints, means for preventing further escalation of issues, and means for shortening response times. This allows operators to verify the reliability of AI-generated responses and gain peace of mind, while also reducing variability in response skills, shortening response times, and preventing escalations of issues.

[1900] "Means for receiving inquiries" refers to the function that allows a server to receive inquiries from users.

[1901] "Means for generating responses using a generative AI model" refers to a function that uses a generative AI model to generate an appropriate response to an incoming inquiry.

[1902] "Means for calculating a confidence score for a generated response" refers to a function for evaluating the reliability of a generated response and calculating a confidence score.

[1903] "Means for displaying responses and confidence scores" refers to a function that visually displays the generated responses and their confidence scores to the user.

[1904] "Means of providing appropriate responses to complaints" refers to a function that provides responses in order to address complaints appropriately.

[1905] "Means to prevent further escalation of the issue" refers to a function that provides preventative measures to prevent problems from escalating during complaint handling.

[1906] "Means for shortening response time" refers to functions designed to improve efficiency in order to reduce the time spent handling complaints.

[1907] "Means of providing psychological reassurance" refers to a function that allows operators to gain psychological reassurance by verifying the reliability of the AI's responses.

[1908] "Means for achieving standardization of response skills" refers to functions that reduce variations in response skills among operators and achieve uniform responses.

[1909] "Means to support the resolution of employment continuity issues" refers to functions that provide support for resolving issues related to the continued employment of operators.

[1910] This invention provides operators with peace of mind and standardizes their response skills by generating responses using a generative AI model and displaying a confidence score in a claims handling support system. A specific embodiment of this system is described below.

[1911] Server Processing

[1912] The server receives a query from the user. The query is a prompt, such as "Please tell me the progress of the new project." The server inputs this query into a generative AI model (e.g., GPT-4) and generates an appropriate response. For the generated response, the server calculates a confidence score. This confidence score is calculated using the AI ​​model's internal evaluation mechanism and additional evaluation algorithms. The server sends the generated response and the confidence score to the terminal.

[1913] Terminal processing

[1914] The terminal receives the response and confidence score sent from the server. The received response and confidence score are displayed on the terminal's screen. For example, it might say, "Current progress is on track. The next milestone is next week. (Confidence score: 85%)." The terminal uses visual aids to make it easy to understand, such as displaying a high confidence score in green and a low score in red.

[1915] User processing

[1916] The user checks the response and confidence score displayed on the device. They evaluate the reliability of the response based on the confidence score. For example, if the confidence score is 85%, the user can trust and accept the response. If the confidence score is low, the user can make another inquiry. When making a re-inquiry, they can add more specific questions based on the previous response.

[1917] Specific example

[1918] Specific example 1: A means of providing a sense of mental security

[1919] A user sends an inquiry asking, "Please tell me the progress of the new project." The server uses a generative AI model (e.g., GPT-4) to generate a response saying, "The current progress is on track. The next milestone is next week," and calculates a confidence score of 85%. The terminal displays this response and the confidence score, and the user feels reassured after checking the confidence score.

[1920] Specific example 2: A means to achieve standardization of response skills

[1921] Operator A receives an inquiry asking, "Could you please explain the procedure for handling customer complaints?" The server uses a generative AI model to generate a response saying, "First, listen carefully to the customer's story, then confirm the details of the problem, and finally, propose a solution." Operator A uses this response as a reference, and Operator B can use the same procedure as a reference for similar inquiries, thereby reducing variability in their response skills.

[1922] Example of a prompt

[1923] "Please tell me about the progress of the new project."

[1924] "Please explain the procedure for handling customer complaints."

[1925] By inputting these prompt sentences into a generating AI model, appropriate responses and confidence scores can be obtained.

[1926] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1927] Step 1:

[1928] The server receives inquiries from users. When a user enters a prompt message such as "Please tell me the progress of the new project" through their terminal and presses the send button, the server receives the inquiry. The input is the user's inquiry, and the output is the inquiry data stored on the server.

[1929] Step 2:

[1930] The server inputs the received query into the generative AI model. Specifically, the server sends the query content to the generative AI model (e.g., GPT-4) via an API. The input is the query data, and the output is the prompt message sent to the generative AI model.

[1931] Step 3:

[1932] The generative AI model generates a response based on a prompt. The generative AI model uses an internal natural language processing algorithm to generate an appropriate response. The input is the prompt, and the output is the generated response text.

[1933] Step 4:

[1934] The server calculates a confidence score for the generated response. The server uses the AI ​​model's internal evaluation mechanism and additional evaluation algorithms to assess the reliability of the response and calculate the confidence score. The input is the generated response text, and the output is the confidence score.

[1935] Step 5:

[1936] The server sends the generated response and confidence score to the terminal. The server combines the response text and confidence score into a single data packet and sends it to the terminal. The input is the response text and confidence score, and the output is the data packet sent to the terminal.

[1937] Step 6:

[1938] The terminal receives the response and confidence score sent from the server. The terminal analyzes the received data packets and extracts the response text and confidence score. The input is the data packet sent from the server, and the output is the analyzed response text and confidence score.

[1939] Step 7:

[1940] The terminal displays the received response and confidence score to the user. The terminal screen displays "Current progress is on track. The next milestone is next week. (Confidence score: 85%)." The input is the parsed response text and confidence score, and the output is the information displayed to the user.

[1941] Step 8:

[1942] The user checks the response and confidence score displayed on the device. The user then evaluates the reliability of the response based on the confidence score. The input is the displayed response text and confidence score, and the output is the user's evaluation result.

[1943] Step 9:

[1944] The user either accepts the response or makes a new inquiry. If the confidence score is high, the user trusts and accepts the response. If the confidence score is low, the user makes a new inquiry with more specific questions. The input is the user's evaluation result, and the output is the new inquiry or the acceptance of the response.

[1945] (Application Example 2)

[1946] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1947] Traditional complaint handling support systems often resulted in significant mental strain on operators when providing appropriate responses to complaints, leading to inconsistencies in their response skills. Furthermore, security operators lacked a reliable means to verify the reliability of AI-generated suggestions, making quick and appropriate responses difficult. This resulted in prolonged response times and an increased risk of the issue escalating.

[1948] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1949] This invention includes a server that provides appropriate responses to complaints, prevents further escalation of the issue, shortens response times, displays a confidence score for AI-generated responses, reduces variations in response skills by allowing operators to refer to AI-generated responses, and enables security operators to receive AI suggestions and respond while checking their confidence scores. This allows operators to respond quickly and appropriately while gaining peace of mind.

[1950] A "customer service representative" is a person whose job is to directly handle inquiries and complaints from customers and users.

[1951] An "operator" is a person whose job is to operate a system or machine and perform a specific task.

[1952] A "complaint handling support system" is a system designed to assist in appropriately responding to complaints from customers and users.

[1953] "Means of providing appropriate responses" refers to the function of generating accurate and effective responses to complaints and providing them to operators.

[1954] "Means to prevent online backlash" refers to the function of taking measures to prevent problems from escalating in handling complaints.

[1955] "Means to shorten response time" refers to functions that reduce the time spent handling complaints.

[1956] "A means of displaying a confidence score for AI-generated responses" refers to a function that quantifies the reliability of AI-generated responses and displays it to the operator.

[1957] "A means of reducing variations in response skills by having operators refer to responses generated by AI" refers to a function that reduces differences in response skills among operators by having operators refer to responses generated by AI.

[1958] A "security operator" is a person whose job is to handle security-related tasks and ensure the safety of systems and facilities.

[1959] A "confidence score" is a numerical value used to evaluate the reliability of a response generated by AI.

[1960] As an embodiment of this invention, a claims handling support system used by security operators will be specifically described.

[1961] System Configuration

[1962] This system consists of the following main components:

[1963] 1. Server: Hosts the AI ​​model and generates responses to queries.

[1964] 2. Terminal: A device such as a smartphone or tablet used by a security operator.

[1965] 3. User Interface: An application for operators to input queries and view AI responses and confidence scores.

[1966] Hardware and software to be used

[1967] Hardware: Smartphones, tablets, servers

[1968] Software: Flask (Python web framework), requests (library for sending HTTP requests), generative AI model

[1969] Data processing and data calculation

[1970] The server receives queries from security operators and sends prompts to the generative AI model. The generative AI model generates responses to the queries and returns those responses to the server. The server calculates a confidence score for the responses and sends it to the operator's terminal.

[1971] Processing flow

[1972] 1. Query Input: The security operator inputs the query through the terminal's user interface.

[1973] 2. Sending the query: The terminal sends the query to the server.

[1974] 3. Response generation: The server sends a query to the generating AI model and generates a response.

[1975] 4. Calculation of confidence score: The server calculates a confidence score for the generated response.

[1976] 5. Display of response and score: The server sends the response and confidence score to the terminal for the operator to review.

[1977] Specific example

[1978] For example, a security operator might enter a query such as, "What should I do if an intruder enters the building?" This query is sent to a server, and a generative AI model generates an appropriate response. The server calculates a confidence score for the generated response and displays the response and score on the operator's terminal.

[1979] Example of a prompt

[1980] "Please tell me how to respond if a suspicious person enters a building."

[1981] In this way, security operators can take quick and appropriate action by referring to AI suggestions. This allows them to gain peace of mind while reducing variability in response skills.

[1982] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1983] Step 1:

[1984] The security operator enters the query through the terminal's user interface.

[1985] Input: Query entered by the security operator (e.g., "Please tell me how to respond if an intruder enters the building.")

[1986] Output: The query is entered into the terminal.

[1987] Specific actions: The security operator opens the application on their smartphone or tablet, enters the query in the text box, and presses the "Submit" button.

[1988] Step 2:

[1989] The terminal sends a query to the server.

[1990] Input: Query entered on the terminal

[1991] Output: The query is sent to the server.

[1992] Specific operation: The application on the terminal sends the entered query to the server as an HTTP request.

[1993] Step 3:

[1994] The server sends queries to the generated AI model and generates responses.

[1995] Input: Query sent to the server

[1996] Output: Response from the generative AI model

[1997] Specific operation: The server sends the received query as a prompt to the generating AI model, and the generating AI model generates a response to the query and returns it to the server.

[1998] Step 4:

[1999] The server calculates a confidence score for the response it generates.

[2000] Input: Response from a generative AI model

[2001] Output: Confidence score

[2002] Specific operation: The server analyzes the content of the generated response and applies an algorithm to calculate a confidence score.

[2003] Step 5:

[2004] The server sends the response and confidence score to the terminal.

[2005] Input: Responses and confidence scores from the generated AI model

[2006] Output: Response and confidence score sent to the terminal

[2007] Specific operation: The server sends the response and confidence score together as an HTTP response to the terminal.

[2008] Step 6:

[2009] The terminal displays its response and confidence score to the security operator.

[2010] Input: Response sent from the server and confidence score

[2011] Output: Response and confidence score displayed to the security operator

[2012] Specific o...

Claims

[Claim 1] A first means of receiving inquiries from users, A second means for generating a first prompt statement for an operator to output a first response to the user based on an inquiry received from the user, A third means for inputting the generated first prompt sentence into a generation AI model to generate the first response, A fourth means for calculating a confidence score for the generated first response using the internal evaluation mechanism of the generating AI model and / or an additional evaluation algorithm, A fifth means for displaying the generated first response and the calculated confidence score on the terminal of the operator responding to the user inquiry, A sixth means for extracting and identifying multiple emotional states, including anger, dissatisfaction, and confusion, from the text of the user's inquiry, using an emotion identification model that functions as an emotion engine. A seventh means that, based on the user's inquiry and the user's emotional state, a model has been constructed to evaluate the risk of escalation based on past complaint data, and a model using a machine learning algorithm calculates a risk score, and if the calculated risk score is high, it automatically notifies the senior person in charge of handling the user's inquiry, Includes, The second means generates a second prompt statement for the operator to output a second response corresponding to the emotional state for responding to the user's inquiry, based on the user's inquiry and the user's emotional state. The third means inputs the generated second prompt sentence into the generating AI model to generate the second response, The fourth means also calculates the confidence score for the generated second response, The fifth means displays the generated second response and the calculated confidence score on the terminal of the operator who responds to the user inquiry. system.

Images