Method and system for quickly responding to business problem in combination with natural language processing model

Abstract

The invention belongs to the technical field of artificial intelligence and computer application, and discloses a method for quickly responding to a service problem in combination with a natural language processing model. The method comprises the following specific steps: S1, question submission and preliminary identification; S2, evaluation of an answer effect and a manual intervention request; S3, manual processing and communication; S4, auxiliary prompting and question answering; the system is enabled through a natural language processing model, the problem solving efficiency and adaptability are remarkably improved, the deep understanding ability enables the system to cross semantic gaps, adapt to various scenes, rapidly respond to requirements and achieve efficient interaction with low time consumption and low cost, meanwhile, the model also endows the system with the ability of rapid answering and self-learning, and the system has good application prospects. Rapid changes of business rules and personnel structures are flexibly handled, and development and operation burdens are effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence and computer application technology, and specifically relates to a method and system for quickly answering business questions by combining a natural language processing model. Background Art

[0002] In today's ever-changing technology, artificial intelligence technologies such as ChatGPT are leading natural language processing to a new level and are widely used in intelligent customer service, chatbots and other fields to achieve smooth dialogue and answer multiple questions. With the rapid expansion of Internet business, the business environment is becoming increasingly complex and the turnover of personnel is accelerating. Frontline employees often encounter problems and need to quickly seek help from business managers. At this time, an efficient problem submission and identification mechanism is particularly important to ensure that problems are quickly communicated and solved to ensure smooth business operations. The current internal communication group feedback mechanism of enterprises faces challenges: delayed response to problems affects customer experience, the person in charge is prone to miss key information due to information overload, and the message is not systematically archived, making it difficult to pass on knowledge. In addition, the system lacks flexibility and is difficult to automatically adapt to business and personnel changes. Uncovered scenarios still need additional solutions. To this end, it is imperative to build a business problem ticket system. The system allows users to submit tickets by business classification and intelligently assign them to the corresponding person in charge to ensure that problems are paid attention to and handled in a timely manner. At the same time, information is archived in an orderly manner, promoting knowledge precipitation and reuse, and providing solid support for enterprise operations. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for quickly answering business questions in combination with a natural language processing model, so as to solve the problems raised in the above-mentioned background technology.

[0004] In order to achieve the above object, the present invention provides the following technical solution: a method for quickly answering business questions by combining a natural language processing model, the specific steps of the method are as follows:

[0005] S1: Question submission and preliminary identification: Frontline employees ask questions through the system, and NLP technology quickly analyzes the intention. The system makes a preliminary match based on the knowledge base and model and automatically answers;

[0006] S2: Evaluate the answer effect and request manual intervention: Evaluate the automatic answer effect. If you are not satisfied, you can request manual intervention. After passing, the system will automatically establish a communication channel and notify you;

[0007] S3: Manual processing and communication: The person in charge will connect with the communication and discuss the details of the problem in detail to ensure accurate understanding and answer;

[0008] S4: Auxiliary prompts and question answers: During the communication process, the system provides auxiliary prompts and assistance, and the manual customer service staff provides detailed answers based on the prompts;

[0009] S5: Recording and archiving of the answering process: The system records the entire answering process and deposits it into the knowledge base in a structured manner to accelerate the answering of similar questions in the future;

[0010] S6: Model training and optimization: Accumulate knowledge training models, iterate and optimize to form a closed loop, and improve service experience and quality.

[0011] Preferably, the specific steps of question submission and preliminary identification in S1 are as follows:

[0012] Step 1: Frontline staff ask questions

[0013] Question input: Frontline employees raise questions through the system interface chat tool. Employees enter text, select options, upload pictures or files to describe their problems or needs;

[0014] Step 2: Natural Language Processing and Understanding

[0015] Parsing questions: After receiving questions from frontline employees, the system uses natural language processing technology to parse the questions, including word segmentation, part-of-speech tagging, and syntactic analysis, to understand the basic structure and composition of the questions;

[0016] Identify intent: Based on the analysis of the question, the system further uses intent recognition technology to infer the employee's intention to ask the question, usually relying on pre-trained models or rule libraries to identify the core requirements or inquiry points of the question;

[0017] Step 3: Preliminary matching and automatic answering

[0018] Knowledge base retrieval: The system searches the existing knowledge base based on the parsed question content and identified intent. The knowledge base contains common questions and their answers, domain knowledge, and business rules.

[0019] Initial matching: The system tries to match the question with entries in the knowledge base to find the most relevant or best matching answer;

[0020] Automatic answers: Once a match is found, the system will automatically generate an answer based on the matching results and display it to front-line employees. If no exact matching answer is found in the knowledge base, the system will give some relevant suggestions or guide employees to further describe the problem.

[0021] Preferably, the specific steps of evaluating the answer effect and requesting manual intervention in S2 are as follows:

[0022] Step 1: Evaluation of answer effectiveness: The system first provides automatic answers to frontline employees, and then evaluates whether the answers meet their needs through a built-in evaluation mechanism or direct feedback from the questioner. The evaluation is based on the relevance, accuracy, and completeness of the answers.

[0023] Step 2: Request human intervention: If frontline employees are dissatisfied with the automated answer or believe the answer is inaccurate, the system provides a clear option or button to allow them to request human intervention. Employees can initiate the request with a simple click and can optionally provide additional context or explanation.

[0024] Step 3: Establish a communication channel: Once the application for manual intervention is approved, the system automatically establishes a direct communication channel between the front-line employees and the manual customer service or the person in charge of the relevant field, which is an instant chat window, email conversation or phone call, and will immediately send notifications to both parties to ensure that both parties are aware of and participate in the problem-solving process in a timely manner.

[0025] Preferably, the manual processing and communication in S3 means that after the system detects that the front-line employee is dissatisfied with the automatic answer and requests manual intervention, the relevant person in charge will immediately receive a notification sent by the system. After receiving the notification, the person in charge will quickly access the established communication channel, instant chat window or telephone line. After access, the manual customer service or person in charge will have an in-depth exchange with the questioner to further explore the details and background information of the problem.

[0026] Preferably, the auxiliary prompts and problem solving in S4 refer to the intelligent intervention of the system in the process of in-depth communication between the manual customer service or person in charge and the front-line employees on the problem, providing real-time auxiliary prompts according to the type of the current problem and the existing knowledge base content, suggesting information that needs to be completed, and more accurately locating the problem; or directly citing relevant material links to help quickly obtain authoritative answers. The manual customer service or person in charge makes full use of these system auxiliary prompts, combined with their own professional knowledge and experience, to provide detailed and accurate answers to the front-line employees, thereby efficiently solving the problem.

[0027] Preferably, the specific solution process of S5 is recorded and archived as follows:

[0028] Step 1: Recording the answering process: During the process of the manual customer service or person in charge answering questions with the front-line staff, the system automatically captures and records all relevant communication content, including the original question asked by the employee, the answer given by the manual customer service or person in charge, the auxiliary prompts provided by the system, and any additional clarification or supplementary information;

[0029] Step 2: Structural processing: The system performs structural processing on the recorded answering process in order to better organize and manage the data. This involves classifying, marking and indexing the text information to form a structured data format, classifying the questions, answers and auxiliary prompts respectively, and assigning unique identifiers or tags. Through structural processing, the system can retrieve and use this information more efficiently.

[0030] Step 3: Sedimentation into the knowledge base: The structured solution process is systematically deposited into the knowledge base. The knowledge base is a system that stores and organizes information and knowledge to support future decision-making and problem solving. In this scenario, the knowledge base will contain a large number of problem-solving cases, each of which has been structured and contains a complete solution process.

[0031] Preferably, the model training and optimization in S6 refers to the system continuously using newly accumulated knowledge and answer cases to conduct in-depth training and optimization of the automatic answering and auxiliary prompting mechanisms. This process forms a closed-loop process from question reception, answer provision, answer archiving to model retraining, ensuring that the system performance continues to improve. Through continuous iteration, it is committed to bringing users a more efficient and accurate service experience, and continuously optimizing the overall service quality.

[0032] Preferably, based on the above method, the system consists of a question submission module, a natural language processing module, a knowledge base and preliminary matching module, an evaluation and manual intervention request module, a communication channel establishment module, a manual processing and communication module, a manual processing and communication module, a solution process recording and archiving module, a model training and optimization module, and a user interface and user experience module.

[0033] The beneficial effects of the present invention are as follows:

[0034] The present invention empowers the system through the natural language processing model, which significantly improves the problem-solving efficiency and adaptability. Its deep understanding ability allows the system to cross the semantic gap, adapt to various scenarios, respond to needs quickly, and achieve efficient interaction with low time consumption and low cost. At the same time, the model also gives the system the ability to quickly answer and self-learn, and flexibly respond to rapid changes in business rules and personnel structure, effectively reducing the burden of development and operation. What is more worth mentioning is that the data generated in the process can be automatically stored in a structured manner and precipitated into a valuable solution library, which not only enhances the accuracy of problem handling, but also points out the direction for frequent problem identification and business optimization, and provides solid data support. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0036] Figure 2 A schematic diagram of a framework for processing business issues of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] like Figure 1 to Figure 2 As shown, the embodiment of the present invention provides a method for quickly answering business questions by combining a natural language processing model. The specific steps of the method are as follows:

[0039] S1: Question submission and preliminary identification: Frontline employees ask questions through the system, and NLP technology quickly analyzes the intention. The system makes a preliminary match based on the knowledge base and model and automatically answers;

[0040] S2: Evaluate the answer effect and request manual intervention: Evaluate the automatic answer effect. If you are not satisfied, you can request manual intervention. After passing, the system will automatically establish a communication channel and notify you;

[0041] S3: Manual processing and communication: The person in charge will connect with the communication and discuss the details of the problem in detail to ensure accurate understanding and answer;

[0042] S4: Auxiliary prompts and problem solving: During the communication process, the system provides auxiliary prompts and assistance, and the human customer service staff gives detailed answers based on the prompts.

[0043] S5: Recording and archiving of the answering process: The system records the entire answering process and deposits it into the knowledge base in a structured manner to accelerate the answering of similar questions in the future;

[0044] S6: Model training and optimization: Accumulate knowledge training models, iterate and optimize to form a closed loop, and improve service experience and quality.

[0045] Model training: Use newly accumulated knowledge and solution cases to train and optimize the system's automatic answering and auxiliary prompt capabilities.

[0046] Closed loop formation: Through continuous iterative training, the system performance is improved, forming a closed loop from question raising to answering, archiving, and retraining, continuously optimizing user experience and service quality.

[0047] The specific steps of question submission and preliminary identification in S1 are as follows:

[0048] Step 1: Frontline staff ask questions

[0049] Question input: Frontline employees ask questions through system interfaces (such as web forms, application interfaces) or chat tools (such as instant messaging software, chat robots). Employees describe their questions or needs by entering text, selecting options, uploading pictures or files, etc.

[0050] Step 2: Natural Language Processing and Understanding

[0051] Parsing questions: After receiving questions from frontline employees, the system uses natural language processing (NLP) technology to parse the questions, including word segmentation, part-of-speech tagging, and syntactic analysis, to understand the basic structure and composition of the questions;

[0052] Identify intent: Based on the analysis of the question, the system further uses intent recognition technology to infer the employee's intention to ask the question, usually relying on pre-trained models or rule libraries to identify the core requirements or inquiry points of the question;

[0053] Step 3: Preliminary matching and automatic answering

[0054] Knowledge base retrieval: The system searches the existing knowledge base based on the parsed question content and identified intent. The knowledge base contains common questions and their answers, domain knowledge, business rules, etc.

[0055] Preliminary matching: The system attempts to match the question with entries in the knowledge base to find the most relevant or best matching answer. This step may involve algorithms such as similarity calculation and keyword matching.

[0056] Automatic answers: Once a match is found, the system will automatically generate an answer based on the matching results and display it to front-line employees. If no exact match is found in the knowledge base, the system may give some relevant suggestions or guide employees to further describe the problem.

[0057] The process of question submission and preliminary identification simplifies problem handling and improves efficiency. Frontline employees can ask questions conveniently, and NLP technology can accurately analyze and identify intentions to ensure that questions are accurately understood. The knowledge base can quickly search and match, and automatically answer questions to reduce waiting and improve user experience. Even if there is no completely matching answer, it can provide suggestions and guidance to promote effective problem solving.

[0058] The specific steps of evaluating the answer effect and requesting manual intervention in S2 are as follows:

[0059] Step 1: Evaluation of answer effectiveness: The system first provides automatic answers to front-line employees, and then evaluates whether the answers meet their needs through a built-in evaluation mechanism or direct feedback from the questioner. The evaluation may be based on factors such as the relevance, accuracy, and completeness of the answers.

[0060] Step 2: Request human intervention: If frontline employees are dissatisfied with the automated answer or believe the answer is inaccurate, the system provides a clear option or button to allow them to request human intervention. Employees can initiate the request with a simple click and can optionally provide additional context or explanation.

[0061] Step 3: Establish a communication channel: Once the application for manual intervention is approved (either immediately or after a certain review process), the system automatically establishes a direct communication channel between the front-line staff and the manual customer service or the person in charge of the relevant field. This may be an instant chat window, email conversation or phone call, and notifications will be sent to both parties immediately to ensure that both parties are aware of and participate in the problem-solving process in a timely manner.

[0062] The advantages of evaluating the effectiveness of answers and manually intervening in the request process are significant: it ensures quality monitoring of automatic answers and improves accuracy through multi-dimensional evaluation; at the same time, the convenient mechanism for applying for manual intervention allows front-line employees to quickly obtain professional support when needed; the instantly established communication channel ensures seamless connection of information between the two parties, accelerates the problem-solving process, and improves overall service efficiency and user experience.

[0063] Among them, the manual processing and communication in S3 refers to that after the system detects that the front-line employees are dissatisfied with the automatic answers and requests manual intervention, the relevant person in charge will immediately receive a notification sent by the system. After receiving the notification, the person in charge will quickly access the established communication channel, such as an instant chat window or a telephone line. After access, the manual customer service or person in charge will have an in-depth exchange with the questioner to further explore the details and background information of the question. Through meticulous inquiries and clarifications, they will ensure a comprehensive and accurate understanding of the problem, laying a solid foundation for providing more accurate and effective answers in the future.

[0064] Manual processing and communication ensure the accuracy and depth of problem answers. When automatic answers do not meet the needs, manual intervention responds quickly, and through instant communication, the details of the problem are deeply discussed to fully and accurately understand the needs. This process not only improves the user experience, but also enhances the pertinence and effectiveness of problem solving, laying a solid foundation for providing more accurate and effective answers.

[0065] Among them, the auxiliary prompts and problem answers in S4 refer to the intelligent intervention of the system in the process of in-depth communication between the manual customer service or person in charge and the front-line employees on the problem, and providing real-time auxiliary prompts according to the type of current problem and the existing knowledge base content, suggesting information that needs to be completed, and more accurately locating the problem; or directly citing relevant material links to help quickly obtain authoritative answers. The manual customer service or person in charge makes full use of these system auxiliary prompts, combined with their own professional knowledge and experience, to provide detailed and accurate answers to the front-line employees, thereby efficiently solving the problem.

[0066] The combination of auxiliary prompts and problem solving significantly improves the efficiency and quality of problem solving. The system intelligently intervenes to provide real-time prompts and data links, helping to accurately locate problems and quickly obtain authoritative answers. Human customer service combines professional knowledge with system assistance to provide detailed and accurate answers to front-line employees, efficiently solve problems, enhance user experience, and promote knowledge sharing and reuse.

[0067] The specific solution process of S5 is recorded and archived as follows:

[0068] Step 1: Recording the answering process: During the process of the manual customer service or manager and the front-line staff answering questions, the system automatically captures and records all relevant communication content. This includes the employee's original question, the manual customer service or manager's answer, the auxiliary prompts provided by the system, and any additional clarification or supplementary information. The content of the record should be detailed and accurate to preserve the full context of the answering process.

[0069] Step 2: Structuring: The system structures the recorded answering process to better organize and manage the data. This step involves classifying, labeling, and indexing the text information to form a structured data format. For example, questions, answers, and auxiliary prompts can be categorized separately and assigned unique identifiers or tags. Through structuring, the system can retrieve and use this information more efficiently.

[0070] Step 3: Sedimentation into the knowledge base: The structured solution process is deposited into the knowledge base by the system. The knowledge base is a system that stores and organizes information and knowledge to support future decision-making and problem solving. In this scenario, the knowledge base will contain a large number of problem-solving cases, each of which is structured and contains a complete solution process. When encountering similar problems in the future, the system can quickly retrieve relevant cases in the knowledge base to provide fast and accurate solutions.

[0071] The recording of the specific answering process, structured processing and knowledge base sedimentation have brought significant benefits: it ensures the integrity and accuracy of the information, facilitating subsequent review and audit; structured processing improves data retrieval efficiency and accelerates information utilization; and the establishment of a knowledge base has realized the accumulation and reuse of knowledge, and can quickly provide accurate answers to similar problems, significantly improving service efficiency and quality, while reducing labor costs and promoting the continuous appreciation of organizational knowledge assets.

[0072] Among them, the model training and optimization in S6 refers to the system's continuous use of newly accumulated knowledge and answer cases to conduct in-depth training and optimization of automatic answering and auxiliary prompt mechanisms. This process forms a closed-loop process from question reception, answer provision, answer archiving to model retraining, ensuring that system performance continues to improve. Through continuous iteration, it is committed to bringing users a more efficient and accurate service experience, and continuously optimizing the overall service quality.

[0073] Model training and optimization form a closed-loop process, continuously using new knowledge and cases to train the system and improve the accuracy of its automatic answers and auxiliary prompts. This process ensures the continuous improvement of system performance and brings users a more efficient and accurate service experience. Through iterative optimization, the overall service quality has been steadily improved, enhancing the competitiveness of the system and user satisfaction.

[0074] Among them, based on the above method, the system consists of a question submission module, a natural language processing (NLP) module, a knowledge base and preliminary matching module, an evaluation and manual intervention request module, a communication channel establishment module, a manual processing and communication module, a manual processing and communication module, a solution process recording and archiving module, a model training and optimization module, a user interface (UI) and a user experience (UX) module.

[0075] The system efficiently and automatically handles business problems. It uses NLP to quickly understand and preliminarily match the knowledge base, significantly improving processing efficiency and speed. It provides accurate answers, reduces misunderstandings, and ensures accurate information transmission. The intelligent evaluation mechanism guides manual intervention in a timely manner to ensure that problems are properly resolved. The seamless communication experience ensures instant information transmission and feedback. The system continuously learns and optimizes, and uses recorded data for iterative training to continuously improve performance. UI / UX design optimizes the interactive experience and improves user satisfaction. Automated processing significantly reduces labor costs, especially for repetitive problems. At the same time, 24 / 7 uninterrupted service enhances business continuity, ensures that problems are resolved at any time, and provides solid support for enterprise operations.

[0076] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0077] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for quickly answering business questions in combination with a natural language processing model, characterized in that: The specific steps of this method are as follows: S1: Question submission and preliminary identification: Frontline employees ask questions through the system, and NLP technology quickly analyzes the intention. The system makes a preliminary match based on the knowledge base and model and automatically answers; S2: Evaluate the answer effect and request manual intervention: Evaluate the automatic answer effect. If you are not satisfied, you can request manual intervention. After passing, the system will automatically establish a communication channel and notify you; S3: Manual processing and communication: The person in charge will connect with the communication and discuss the details of the problem in detail to ensure accurate understanding and answer; S4: Auxiliary prompts and question answers: During the communication process, the system provides auxiliary prompts and assistance, and the manual customer service staff provides detailed answers based on the prompts; S5: Recording and archiving of the answering process: The system records the entire answering process and deposits it into the knowledge base in a structured manner to accelerate the answering of similar questions in the future; S6: Model training and optimization: Accumulate knowledge training models, iterate and optimize to form a closed loop, and improve service experience and quality.

2. According to claim 1, a method for quickly answering business questions in combination with a natural language processing model is characterized by: The specific steps of problem submission and preliminary identification in S1 are as follows: Step 1: Frontline staff ask questions Question input: Frontline employees raise questions through the system interface chat tool. Employees enter text, select options, upload pictures or files to describe their problems or needs; Step 2: Natural Language Processing and Understanding Parsing questions: After receiving questions from frontline employees, the system uses natural language processing technology to parse the questions, including word segmentation, part-of-speech tagging, and syntactic analysis, to understand the basic structure and composition of the questions; Identify intent: Based on the analysis of the question, the system further uses intent recognition technology to infer the employee's intention to ask the question, usually relying on pre-trained models or rule libraries to identify the core requirements or inquiry points of the question; Step 3: Preliminary matching and automatic answering Knowledge base retrieval: The system searches the existing knowledge base based on the parsed question content and identified intent. The knowledge base contains common questions and their answers, domain knowledge, and business rules. Initial matching: The system tries to match the question with entries in the knowledge base to find the most relevant or best matching answer; Automatic answers: Once a match is found, the system will automatically generate an answer based on the matching results and display it to front-line employees. If no exact matching answer is found in the knowledge base, the system will give some relevant suggestions or guide employees to further describe the problem.

3. According to claim 1, a method for quickly answering business questions in combination with a natural language processing model is characterized by: The specific steps of evaluating the answer effect and requesting manual intervention in S2 are as follows: Step 1: Evaluation of answer effectiveness: The system first provides automatic answers to frontline employees, and then uses a built-in evaluation mechanism or direct feedback from the questioner to evaluate whether the answer meets their needs. The evaluation is based on the relevance, accuracy, and completeness of the answer. Step 2: Request human intervention: If frontline employees are dissatisfied with the automated answer or believe the answer is inaccurate, the system provides a clear option or button to allow them to request human intervention. Employees can initiate the request with a simple click and can optionally provide additional context or explanation. Step 3: Establish a communication channel: Once the application for manual intervention is approved, the system automatically establishes a direct communication channel between the front-line employees and the manual customer service or the person in charge of the relevant field, which is an instant chat window, email conversation or phone call, and will immediately send notifications to both parties to ensure that both parties are aware of and participate in the problem-solving process in a timely manner.

4. The method for quickly answering business questions in combination with a natural language processing model according to claim 1, characterized in that: The manual processing and communication in S3 refers to that after the system detects that the front-line staff is dissatisfied with the automatic answer and requests manual intervention, the relevant person in charge will immediately receive a notification sent by the system. After receiving the notification, the person in charge will quickly access the established communication channel, instant chat window or telephone line. After access, the manual customer service or person in charge will have an in-depth communication with the questioner to further explore the details and background information of the problem.

5. The method for quickly answering business questions in combination with a natural language processing model according to claim 1, characterized in that: The auxiliary prompts and problem solving in S4 refer to the intelligent intervention of the system in the process of in-depth communication between the manual customer service or person in charge and the front-line employees on the problem. According to the type of the current problem and the existing knowledge base content, the system provides real-time auxiliary prompts, suggestions for information that needs to be completed, and more accurately locates the problem; or directly references relevant material links to help quickly obtain authoritative answers. The manual customer service or person in charge makes full use of these system auxiliary prompts, combined with their own professional knowledge and experience, to provide detailed and accurate answers to the front-line employees, thereby efficiently solving the problem.

6. The method for quickly answering business questions in combination with a natural language processing model according to claim 1, characterized in that: The specific solution process of S5 is recorded and archived as follows: Step 1: Recording the answering process: During the process of the manual customer service or person in charge answering questions with the front-line staff, the system automatically captures and records all relevant communication content, including the original question asked by the employee, the answer given by the manual customer service or person in charge, the auxiliary prompts provided by the system, and any additional clarification or supplementary information; Step 2: Structural processing: The system performs structural processing on the recorded answering process in order to better organize and manage the data. This involves classifying, marking and indexing the text information to form a structured data format, classifying the questions, answers and auxiliary prompts respectively, and assigning unique identifiers or tags. Through structural processing, the system can retrieve and use this information more efficiently. Step 3: Sedimentation into the knowledge base: The structured solution process is systematically deposited into the knowledge base. The knowledge base is a system that stores and organizes information and knowledge to support future decision-making and problem solving. In this scenario, the knowledge base will contain a large number of problem-solving cases, each of which has been structured and contains a complete solution process.

7. The method of quickly answering business questions in combination with a natural language processing model according to claim 1, characterized in that: The model training and optimization in S6 refers to the system's continuous use of newly accumulated knowledge and answer cases to conduct in-depth training and optimization of the automatic answer and auxiliary prompt mechanisms. This process forms a closed-loop process from question reception, answer provision, answer archiving to model retraining, ensuring that the system performance continues to improve. Through continuous iteration, it is committed to bringing users a more efficient and accurate service experience, and continuously optimizing the overall service quality.

8. A system for quickly answering business questions in combination with a natural language processing model, characterized in that: Based on the above method, the system consists of a question submission module, a natural language processing module, a knowledge base and preliminary matching module, an evaluation and manual intervention request module, a communication channel establishment module, a manual processing and communication module, a manual processing and communication module, a solution process recording and archiving module, a model training and optimization module, and a user interface and user experience module.

Images