A method and device for realizing intelligent customer service

Abstract

The present invention discloses a method and device for realizing intelligent customer service, and specifically relates to the technical field of intelligent customer service, including an intelligent customer service realization data acquisition module, an intelligent customer service realization data analysis module, an intelligent customer service realization intelligent evaluation module, an intelligent customer service realization intelligent comprehensive analysis module and an intelligent customer service realization intelligent human-computer interaction module; the intelligent customer service realization data acquisition module is used to collect intelligent performance data of the intelligent customer service, obtain various monitoring parameters for evaluating the intelligent performance, and transmit the various monitoring parameters to the intelligent customer service realization data analysis module; the present invention analyzes the intelligent parsing parameters to obtain an indicator to measure the ability of the intelligent customer service to understand and handle user problems in the realization of intelligence, thereby helping to improve the accuracy of the intelligent customer service system's understanding of user input, processing efficiency and overall service quality.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent customer service, and in particular to a method and device for realizing intelligent customer service. Background Art

[0002] Customer service, or customer service in full, refers to the process by which an enterprise provides customers with support and assistance in all aspects of pre-sales, sales, and after-sales through a series of professional service means and techniques in order to meet customer needs. Intelligent customer service refers to a form of customer service that uses customer service robots to assist humans in conversation, quality inspection, and business processing, thereby reducing labor costs and improving response efficiency, and is widely used in e-commerce, finance, telecommunications and other industries.

[0003] Traditional manual customer service faces problems such as high cost and low efficiency. It usually relies on manual services and makes decisions based on personal experience and judgment, which may be subjective and limited. A customer service staff can only handle a limited number of issues at the same time and may face service pressure during peak hours, resulting in relatively low service efficiency. In addition, due to the limitations of manual operation speed and working hours, the response speed may be slow. With the rapid development of the Internet, the demand for intelligent customer service is growing, and customers have higher and higher requirements for intelligent services, so it is very important to realize the intelligent performance of intelligent customer service.

[0004] Therefore, a method and device for realizing intelligent customer service are provided to solve the above problems, especially to realize intelligent analysis of intelligent customers. Through natural language processing and machine learning technology, the question information input by the user is received, and then semantic analysis is performed. It can more accurately understand the customer's intentions and needs, and communicate with the customer in a more natural and smooth way. It can quickly and accurately answer the user's questions, improve the efficiency and response speed of customer service, reduce customer service costs, reduce dependence on manual labor, and adapt to the ever-changing user needs by continuously learning and updating knowledge graphs and models. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and device for implementing intelligent customer service to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an implementation device of intelligent customer service, including an implementation data collection module of intelligent customer service, an implementation data analysis module of intelligent customer service, an implementation intelligent evaluation module of intelligent customer service, an implementation intelligent comprehensive analysis module of intelligent customer service, and an implementation intelligent human-computer interaction module of intelligent customer service;

[0007] The intelligent customer service realization data collection module is used to collect the intelligent performance data of the intelligent customer service, obtain various monitoring parameters for evaluating the intelligent performance indicators, and transmit the various monitoring parameters to the intelligent customer service realization data analysis module;

[0008] The data analysis module for realizing the intelligent customer service includes an intelligent parsing quality index analysis unit, a knowledge graph intelligent construction quality index analysis unit, an intelligent learning effectiveness index analysis unit, and a cloud computing intelligent response performance index analysis unit. The intelligent parsing quality index analysis unit is used to calculate the intelligent parsing parameters, the knowledge graph intelligent construction quality index analysis unit is used to calculate the knowledge graph intelligent construction parameters, the intelligent learning effectiveness index analysis unit is used to calculate the intelligent learning effectiveness parameters, and the cloud computing intelligent response performance index analysis unit is used to calculate the cloud computing intelligent response parameters, and the corresponding calculation results are output to the intelligent evaluation module for realizing the intelligent customer service;

[0009] The model for obtaining the cloud computing intelligent response performance index in the cloud computing intelligent response performance index analysis unit is: ,in It represents the cloud computing intelligent response performance index. is a constant, , Indicates the normal operating time of the program during the test time. Indicates the system failure time within the test time, Indicates the number of successful requests during the test time. Indicates the number of requests. and Respectively represent the impact factors based on time dimension and request dimension, express Number of sessions resolved quickly within a certain period of time, Indicates the number of successful API calls. Indicates the number of interfaces. Indicates the number of sessions. represents the peak hour response time, represents the standard deviation of the response time during peak hours. Represents the average response time during peak hours;

[0010] The intelligent evaluation module for realizing the intelligent customer service compares the intelligent indicators in the above analysis module with the threshold value, and inputs the comparison result to the intelligent comprehensive analysis module for realizing the intelligent customer service;

[0011] The intelligent comprehensive analysis module for realizing intelligent customer service is used to comprehensively analyze the intelligent evaluation coefficients of the four intelligent indicators calculated by the above evaluation module to obtain the intelligent comprehensive index for realizing intelligent customer service, and output the intelligent comprehensive index to the intelligent human-computer interaction module for realizing intelligent customer service;

[0012] The intelligent human-computer interaction module for realizing the intelligent customer service is used to output the intelligent comprehensive indicators for realizing the intelligent customer service transmitted by the above-mentioned comprehensive analysis module to the software administrator information terminal.

[0013] Preferably, the monitoring parameters in the data collection module for realizing the intelligent customer service include software administrator identity authentication information, intelligent analysis parameters, knowledge graph intelligent construction parameters, intelligent learning efficiency parameters and cloud computing intelligent response parameters. The intelligent analysis parameters are specifically the number of conversations that complete multiple rounds of dialogues, the number of conversations, the number of words correctly segmented in natural language, the number of segmented words, the number of correctly recognized accent types, the number of audios, the number of words replaced by speech recognition, the number of deleted words, the number of additionally inserted words and the total number of words in the original text; the knowledge graph intelligent construction parameters are specifically the number of questions that can be answered by the knowledge base, the number of question types actually raised by users, and the number of verified accurate The parameters of intelligent learning performance include the number of users’ actual clicks on recommended content, the number of times recommended content is displayed to users, the number of correct information extracted, the number of extracted information, the number of correctly detected abnormal sessions, the number of sessions, the accuracy index of the new algorithm and the accuracy index of the old algorithm; the parameters of cloud computing intelligent response include the normal operation time of the program during the test period, the system failure time, the number of successful requests, the number of requests, the number of quickly resolved sessions, the number of sessions, the number of successful API interface calls, the number of interfaces and the response time during peak hours.

[0014] Preferably, the model for obtaining the intelligent parsing quality index in the intelligent parsing quality index analysis unit is: ,in Indicates the quality index of intelligent analysis. Indicates the number of sessions that complete multiple rounds of dialogue. Indicates the number of sessions. Indicates the number of words correctly segmented in natural language, Indicates the number of words segmented. Indicates the number of correctly recognized accent types, Indicates the number of audios. Indicates the number of words replaced by speech recognition, Indicates the number of words deleted by speech recognition. Indicates the number of additional words inserted for speech recognition. Represents the total number of words in the original text, , , as well as They respectively represent the influencing factors of multi-round dialogue completion rate, natural language understanding accuracy, accent type accuracy, and word error rate on the intelligent analysis quality indicators.

[0015] Preferably, the model for the knowledge graph intelligent construction quality index analysis unit to obtain the knowledge graph intelligent construction quality index is: ,in represents the quality index of the intelligent construction of knowledge graph, is a constant, , represents the number of questions that the knowledge base can answer. Indicates the number of question types actually asked by users. Indicates the number of entries that have been verified to be accurate. represents the number of entries in the knowledge base, Indicates the number of newly added knowledge items, represents the number of knowledge items available externally during the same period, represents the efficiency of knowledge intelligent reasoning, , TP represents the number of answers inferred to be correct, and FP represents the number of answers inferred to be wrong. , and They represent the influencing factors of knowledge base coverage, knowledge base accuracy verification rate, and new knowledge absorption rate respectively.

[0016] Preferably, the model for obtaining the intelligent learning effectiveness index in the intelligent learning effectiveness index analysis unit is: ,in represents the intelligent learning effectiveness index, is a constant, , Indicates the number of times users actually click on the recommended content. Indicates the number of times the recommended content is shown to the user. Indicates the number of correct information extracted. Indicates the number of information items extracted. Indicates the number of abnormal sessions that are correctly detected. Indicates the number of sessions. Represents the accuracy index of the new algorithm, Represents the accuracy index of the old algorithm, , , as well as They respectively represent the influencing factors of intelligent recommendation click rate, personalized recommendation accuracy, anomaly detection accuracy and algorithm iteration effect index.

[0017] Preferably, the intelligent evaluation module of the intelligent customer service comprises the following steps:

[0018] Step 1: Obtain the intelligent analysis quality index analysis calculation results from the intelligent analysis quality index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of the intelligent analysis quality index, Indicates the threshold value;

[0019] Step 2: Obtain the calculation results of the quality index analysis of the knowledge graph intelligent construction quality index by using the knowledge graph intelligent construction quality index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of the quality index of the intelligent construction of the knowledge graph, Indicates the threshold value;

[0020] Step 3: Obtain the intelligent learning effectiveness index analysis calculation results from the intelligent learning effectiveness index analysis unit Substituting into the formula: ,in represents the intelligent evaluation coefficient of the intelligent learning effectiveness index, Indicates the threshold value;

[0021] Step 4: Obtain the cloud computing intelligent response performance index analysis calculation results from the cloud computing intelligent response performance index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of cloud computing intelligent response performance indicators, Indicates the threshold value.

[0022] Preferably, the comprehensive analysis model in the intelligent comprehensive analysis module of the intelligent customer service is: ,in It represents the comprehensive index of intelligent customer service. Indicates Intelligent evaluation coefficient, Indicates The weight of each intelligent evaluation coefficient.

[0023] A method for implementing intelligent customer service, characterized in that it includes the following steps:

[0024] Step 1: Collect the intelligent performance data of the intelligent customer service to obtain various monitoring parameters for evaluating the intelligent performance indicators;

[0025] Step 2: Based on the monitoring parameters of the intelligent performance evaluation indicators obtained in step 1, the intelligent analysis parameters, knowledge graph intelligent construction parameters, intelligent learning efficiency parameters and cloud computing intelligent response parameters are calculated to obtain various intelligent indicators;

[0026] Step 3: Compare the calculation result of step 2 with the threshold value, and obtain the intelligent evaluation coefficient of each intelligent indicator according to the comparison result;

[0027] Step 4: Perform a comprehensive analysis on the intelligent evaluation coefficients of various intelligent indicators obtained in step 3 to obtain the intelligent comprehensive indicators for realizing intelligent customer service;

[0028] Step 5: Based on the comprehensive analysis results of step 4, the intelligent comprehensive indicators of intelligent customer service are displayed for human-computer interaction.

[0029] Technical effects and advantages of the present invention:

[0030] 1. The present invention analyzes the intelligent parsing parameters to obtain an indicator to measure the ability of intelligent customer service to understand and process user problems in the realization of intelligence. Natural language understanding is the basis for realizing intelligence. It determines whether the system can effectively interact with users and understand their needs, which helps to improve the accuracy of understanding of user input, processing efficiency and overall service quality of the intelligent customer service system;

[0031] 2. The present invention analyzes the parameters of the intelligent construction of the knowledge graph to obtain an indicator to measure the intelligent customer service system's ability to understand the semantics of user input, its ability to make inferences and judgments, and its ability to continuously learn and evolve by constantly absorbing new knowledge, thereby enabling the intelligent customer service system to handle more complex and changeable problems;

[0032] 3. The present invention analyzes the intelligent learning efficiency parameters to obtain an indicator to measure the ability of the intelligent customer service system to improve its performance and effectiveness through learning and optimization. With the improvement of learning efficiency, it can reduce the number of incorrect answers and misunderstandings of user intentions, which in turn helps the intelligent customer service to understand user intentions more accurately and provide more satisfactory answers, thereby improving user satisfaction and loyalty. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the overall process of the present invention. DETAILED DESCRIPTION

[0034] 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.

[0035] See also Figure 1 As shown, the present invention provides a method and device for realizing intelligent customer service, including a data collection module for realizing intelligent customer service, a data analysis module for realizing intelligent customer service, an intelligent evaluation module for realizing intelligent customer service, an intelligent comprehensive analysis module for realizing intelligent customer service, and an intelligent human-computer interaction module for realizing intelligent customer service.

[0036] The data collection module for realizing intelligent customer service is connected to the data analysis module for realizing intelligent customer service, the intelligent evaluation module for realizing intelligent customer service is respectively connected to the data analysis module for realizing intelligent customer service and the intelligent comprehensive analysis module for realizing intelligent customer service, and the intelligent human-computer interaction module for realizing intelligent customer service is connected to the intelligent comprehensive analysis module for realizing intelligent customer service.

[0037] The intelligent customer service realization data collection module is used to collect the intelligent performance data of the intelligent customer service, obtain various monitoring parameters for evaluating the intelligent performance indicators, and transmit the various monitoring parameters to the intelligent customer service realization data analysis module;

[0038] It should be specifically explained in this embodiment that the various monitoring parameters include software administrator identity authentication information, intelligent analysis parameters, knowledge graph intelligent construction parameters, intelligent learning efficiency parameters and cloud computing intelligent response parameters. The intelligent analysis parameters are specifically the number of conversations that complete multiple rounds of dialogues, the number of conversations, the number of words correctly segmented in natural language, the number of segmented words, the number of correctly recognized accent types, the number of audios, the number of words replaced by speech recognition, the number of deleted words, the number of additionally inserted words and the total number of words in the original text; the knowledge graph intelligent construction parameters are specifically the number of questions that can be answered by the knowledge base, the number of question types actually raised by users, and the number of entries that have been verified to be accurate. , the number of entries in the knowledge base, the number of newly added knowledge entries, the number of knowledge entries available externally during the same period, and the number of incorrect answers; the intelligent learning effectiveness parameters are specifically the number of times users actually click on the recommended content, the number of times the recommended content is displayed to users, the number of correct information extracted, the number of extracted information, the number of correctly detected abnormal sessions, the number of sessions, the new algorithm accuracy index, and the old algorithm accuracy index; the cloud computing intelligent response parameters are specifically the normal program operation time, system failure time, number of successful requests, number of requests, number of quickly resolved sessions, number of sessions, number of successful API interface calls, number of interfaces, and peak period response time during the test time.

[0039] The data analysis module for realizing the intelligent customer service includes an intelligent parsing quality index analysis unit, a knowledge graph intelligent construction quality index analysis unit, an intelligent learning effectiveness index analysis unit, and a cloud computing intelligent response performance index analysis unit. The intelligent parsing quality index analysis unit is used to calculate the intelligent parsing parameters, the knowledge graph intelligent construction quality index analysis unit is used to calculate the knowledge graph intelligent construction parameters, the intelligent learning effectiveness index analysis unit is used to calculate the intelligent learning effectiveness parameters, and the cloud computing intelligent response performance index analysis unit is used to calculate the cloud computing intelligent response parameters, and the corresponding calculation results are output to the intelligent evaluation module for realizing the intelligent customer service. The data analysis includes the following steps;

[0040] Step 1: By analyzing the intelligent parsing parameters, we can measure the ability of intelligent customer service to understand and handle user problems in the process of intelligent realization. We can comprehensively consider the completion rate of multiple rounds of dialogue, the accuracy of natural language understanding, the accuracy of accent type, and the word error rate, which will help improve user experience and service efficiency, and promote technical optimization and iteration. Therefore, the model for obtaining the intelligent parsing quality index in the intelligent parsing quality index analysis unit is: ,in Indicates the quality index of intelligent analysis. Indicates the number of sessions that complete multiple rounds of dialogue. Indicates the number of sessions. Indicates the number of words correctly segmented in natural language, Indicates the number of words segmented. Indicates the number of correctly recognized accent types, Indicates the number of audios. Indicates the number of words replaced by speech recognition, Indicates the number of words deleted by speech recognition. Indicates the number of additional words inserted for speech recognition. Represents the total number of words in the original text, , , as well as They represent the factors affecting the quality index of intelligent parsing, including the completion rate of multi-round dialogue, the accuracy rate of natural language understanding, the accuracy rate of accent type, and the word error rate.

[0041] Step 2: By analyzing the parameters of the intelligent construction of the knowledge graph, we can measure the intelligent customer service system's ability to understand the semantics of user input, its ability to make inferences and judgments, and its ability to continuously learn and evolve by constantly absorbing new knowledge. Taking into account the knowledge base coverage, knowledge base accuracy verification rate, new knowledge absorption rate, and knowledge intelligent recommendation efficiency, the knowledge graph intelligent construction quality index analysis unit obtains the knowledge graph intelligent construction quality index model as follows: ,in represents the quality index of the intelligent construction of knowledge graph, is a constant, , represents the number of questions that the knowledge base can answer. Indicates the number of question types actually asked by users. Indicates the number of entries that have been verified to be accurate. represents the number of entries in the knowledge base, Indicates the number of newly added knowledge items, It indicates the number of knowledge items available externally during the same period. External data sources include the Internet, databases, and other systems. It represents the efficiency of knowledge intelligent reasoning, that is, the efficiency of the intelligent customer service system in finding relevant solutions through logical reasoning and relationship analysis when there is no direct answer in the knowledge base. TP represents the number of answers inferred to be correct, and FP represents the number of answers inferred to be wrong, that is, the number of answers that are actually wrong but inferred to be correct. , and They represent the influencing factors of knowledge base coverage, knowledge base accuracy verification rate, and new knowledge absorption rate respectively;

[0042] Step 3: By analyzing the intelligent learning effectiveness parameters, we can measure the ability of the intelligent customer service system to improve its performance and effectiveness through learning and optimization. With the improvement of learning effectiveness, we can reduce the number of incorrect answers and misunderstandings of user intentions, and improve the stability and reliability of the system. Taking into account the algorithm iteration effect index, intelligent recommendation click rate, personalized recommendation accuracy, and anomaly detection accuracy, the model for obtaining intelligent learning effectiveness indicators in the intelligent learning effectiveness indicator analysis unit is: ,in represents the intelligent learning effectiveness index, is a constant, , Indicates the number of times users actually click on the recommended content. Indicates the number of times the recommended content is shown to the user. Indicates the number of correct information extracted, that is, the number of contents recommended by the system to the user and that the user actually likes or is interested in. Indicates the number of extracted information, that is, the total number of contents recommended to users by the system, regardless of whether the user likes or is interested in them. Indicates the number of abnormal sessions that are correctly detected, that is, the number of abnormal sessions such as insults and spam that are automatically identified and marked by the system. Indicates the number of sessions. Represents the accuracy index of the new algorithm, Represents the accuracy index of the old algorithm, , , as well as They represent the influencing factors of intelligent recommendation click rate, personalized recommendation accuracy, anomaly detection accuracy, and algorithm iteration effect index respectively;

[0043] Step 4: By analyzing the cloud computing intelligent response parameters, we can measure the response speed, processing capacity, stability and reliability of the intelligent customer service system to user requests, taking into account the response time changes during peak hours, the rapid resolution rate, the API interface call success rate and the system availability. Therefore, the model for obtaining the cloud computing intelligent response performance index in the cloud computing intelligent response performance index analysis unit is: ,in It represents the cloud computing intelligent response performance index. is a constant, , Indicates the normal operating time of the program during the test time. Indicates the system failure time within the test time, Indicates the number of successful requests during the test time. Indicates the number of requests. and Respectively represent the impact factors based on time dimension and request dimension, express Number of sessions resolved quickly within a certain period of time, Indicates the number of successful API calls. API is the Application Programming Interface (API). Through API, different systems can share and exchange data, allowing data to flow between multiple applications, improving data utilization and system interoperability. Indicates the number of interfaces. Indicates the number of sessions. Refers to the number of sessions that are quickly resolved within a preset time threshold (such as 10 seconds or 20 seconds). Indicates the peak period response time, STDDEV() indicates the standard deviation of the peak period response time. Represents the average response time during peak hours:

[0044] The intelligent evaluation module for realizing the intelligent customer service compares the intelligent indicators in the above analysis module with the threshold value, and inputs the comparison result to the intelligent comprehensive analysis module for realizing the intelligent customer service, and the intelligent evaluation module includes the following steps;

[0045] Step 1: Obtain the intelligent analysis quality index analysis calculation results from the intelligent analysis quality index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of the intelligent analysis quality index, Indicates the threshold value;

[0046] Step 2: Obtain the calculation results of the quality index analysis of the knowledge graph intelligent construction quality index by using the knowledge graph intelligent construction quality index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of the quality index of the intelligent construction of the knowledge graph, Indicates the threshold value;

[0047] Step 3: Obtain the intelligent learning effectiveness index analysis calculation results from the intelligent learning effectiveness index analysis unit Substituting into the formula: ,in represents the intelligent evaluation coefficient of the intelligent learning effectiveness index, Indicates the threshold value;

[0048] Step 4: Obtain the cloud computing intelligent response performance index analysis calculation results from the cloud computing intelligent response performance index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of cloud computing intelligent response performance indicators, Indicates the threshold value;

[0049] The intelligent comprehensive analysis module for realizing intelligent customer service is used to comprehensively analyze the intelligent evaluation coefficients of the four intelligent indicators calculated by the above evaluation module to obtain the intelligent comprehensive indicators for realizing intelligent customer service, and output the intelligent comprehensive indicators to the intelligent human-computer interaction module for realizing intelligent customer service. The comprehensive analysis model is: ,in It represents the comprehensive index of intelligent customer service. Indicates Intelligent evaluation coefficient, Indicates The weight of each intelligent evaluation coefficient;

[0050] The intelligent human-computer interaction module for realizing the intelligent customer service is used to output the intelligent comprehensive indicators of the intelligent customer service realized by the above-mentioned comprehensive analysis module to the software administrator information terminal. If the intelligent comprehensive indicators belong to the threshold range, it means that the intelligent performance of the intelligent customer service is good. Otherwise, the software administrator is prompted to take timely measures to continuously collect customer questions and feedback, continuously optimize the knowledge graph and algorithm, and improve the performance and service quality of the intelligent customer service. At the same time, personalized intelligent customer service solutions can be customized according to different industries and user needs.

[0051] A method for implementing intelligent customer service, characterized in that it includes the following steps:

[0052] Step 1: Collect the intelligent performance data of the intelligent customer service to obtain various monitoring parameters for evaluating the intelligent performance indicators;

[0053] Step 2: Based on the monitoring parameters of the intelligent performance evaluation indicators obtained in step 1, the intelligent analysis parameters, knowledge graph intelligent construction parameters, intelligent learning efficiency parameters and cloud computing intelligent response parameters are calculated to obtain various intelligent indicators;

[0054] Step 3: Compare the calculation result of step 2 with the threshold value, and obtain the intelligent evaluation coefficient of each intelligent indicator according to the comparison result;

[0055] Step 4: Perform a comprehensive analysis on the intelligent evaluation coefficients of various intelligent indicators obtained in step 3 to obtain the intelligent comprehensive indicators for realizing intelligent customer service;

[0056] Step 5: Based on the comprehensive analysis results of step 4, the intelligent comprehensive indicators of intelligent customer service are displayed for human-computer interaction.

[0057] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0058] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A device for implementing intelligent customer service, characterized in that: It includes a data collection module for realizing intelligent customer service, a data analysis module for realizing intelligent customer service, an intelligent evaluation module for realizing intelligent customer service, an intelligent comprehensive analysis module for realizing intelligent customer service, and an intelligent human-computer interaction module for realizing intelligent customer service; The intelligent customer service realization data collection module is used to collect the intelligent performance data of the intelligent customer service, obtain various monitoring parameters for evaluating the intelligent performance indicators, and transmit the various monitoring parameters to the intelligent customer service realization data analysis module; The data analysis module for realizing the intelligent customer service includes an intelligent parsing quality index analysis unit, a knowledge graph intelligent construction quality index analysis unit, an intelligent learning effectiveness index analysis unit, and a cloud computing intelligent response performance index analysis unit. The intelligent parsing quality index analysis unit is used to calculate the intelligent parsing parameters, the knowledge graph intelligent construction quality index analysis unit is used to calculate the knowledge graph intelligent construction parameters, the intelligent learning effectiveness index analysis unit is used to calculate the intelligent learning effectiveness parameters, and the cloud computing intelligent response performance index analysis unit is used to calculate the cloud computing intelligent response parameters, and the corresponding calculation results are output to the intelligent evaluation module for realizing the intelligent customer service; The model for obtaining the intelligent parsing quality index in the intelligent parsing quality index analysis unit is: ,in Indicates the quality index of intelligent analysis. Indicates the number of sessions that complete multiple rounds of dialogue. Indicates the number of sessions. Indicates the number of words correctly segmented in natural language, Indicates the number of words segmented. Indicates the number of correctly recognized accent types, Indicates the number of audios. Indicates the number of words replaced by speech recognition, Indicates the number of words deleted by speech recognition. Indicates the number of additional words inserted for speech recognition. Represents the total number of words in the original text, , , as well as They represent the factors affecting the quality index of intelligent parsing, including the completion rate of multi-round dialogue, the accuracy rate of natural language understanding, the accuracy rate of accent type, and the word error rate. The model for the knowledge graph intelligent construction quality index analysis unit to obtain the knowledge graph intelligent construction quality index is: ,in represents the quality index of the intelligent construction of knowledge graph, is a constant, , represents the number of questions that the knowledge base can answer. Indicates the number of question types actually asked by users. Indicates the number of entries that have been verified to be accurate. represents the number of entries in the knowledge base, Indicates the number of newly added knowledge items, represents the number of knowledge items available externally during the same period, represents the efficiency of knowledge intelligent reasoning, , TP represents the number of answers inferred to be correct, and FP represents the number of answers inferred to be wrong. , and They represent the influencing factors of knowledge base coverage, knowledge base accuracy verification rate, and new knowledge absorption rate respectively; The model for obtaining the intelligent learning effectiveness index in the intelligent learning effectiveness index analysis unit is: ,in represents the intelligent learning effectiveness index, is a constant, , Indicates the number of times users actually click on the recommended content. Indicates the number of times the recommended content is shown to the user. Indicates the number of correct information extracted. Indicates the number of information items extracted. Indicates the number of abnormal sessions that are correctly detected. Indicates the number of sessions. represents the accuracy index of the new algorithm, Represents the accuracy index of the old algorithm, , , as well as They represent the influencing factors of intelligent recommendation click rate, personalized recommendation accuracy, anomaly detection accuracy, and algorithm iteration effect index respectively; The model for obtaining the cloud computing intelligent response performance index in the cloud computing intelligent response performance index analysis unit is: ,in represents the cloud computing intelligent response performance index, is a constant, , Indicates the normal operating time of the program during the test time. Indicates the system failure time within the test time. Indicates the number of successful requests during the test time. Indicates the number of requests. and Respectively represent the impact factors based on time dimension and request dimension, express Number of sessions resolved quickly within a certain period of time, Indicates the number of successful API calls. Indicates the number of interfaces. Indicates the number of sessions. represents the peak hour response time, represents the standard deviation of the response time during peak hours. Represents the average response time during peak hours; The intelligent evaluation module for realizing the intelligent customer service compares the intelligent indicators in the above analysis module with the threshold value, and inputs the comparison result to the intelligent comprehensive analysis module for realizing the intelligent customer service; The intelligent evaluation module of the intelligent customer service includes the following steps: Step 1: Obtain the intelligent analysis quality index analysis calculation results from the intelligent analysis quality index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of the intelligent analysis quality index, Indicates the threshold value; Step 2: Obtain the calculation results of the quality index analysis of the knowledge graph intelligent construction quality index by using the knowledge graph intelligent construction quality index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of the quality index of the intelligent construction of the knowledge graph, Indicates the threshold value; Step 3: Obtain the intelligent learning effectiveness index analysis calculation results from the intelligent learning effectiveness index analysis unit Substituting into the formula: ,in represents the intelligent evaluation coefficient of the intelligent learning effectiveness index, Indicates the threshold value; Step 4: Obtain the cloud computing intelligent response performance index analysis calculation results from the cloud computing intelligent response performance index analysis unit Substituting into the formula: ,in Represents the intelligent evaluation coefficient of cloud computing intelligent response performance indicators, Indicates the threshold value; The intelligent comprehensive analysis module for realizing intelligent customer service is used to comprehensively analyze the intelligent evaluation coefficients of the four intelligent indicators calculated by the above evaluation module to obtain the intelligent comprehensive index for realizing intelligent customer service, and output the intelligent comprehensive index to the intelligent human-computer interaction module for realizing intelligent customer service; The comprehensive analysis model of the intelligent comprehensive analysis module of the intelligent customer service is: ,in It represents the comprehensive index of intelligent customer service. Indicates Intelligent evaluation coefficient, Indicates The weight of each intelligent evaluation coefficient; The intelligent human-computer interaction module for realizing the intelligent customer service is used to output the intelligent comprehensive indicators for realizing the intelligent customer service transmitted by the above-mentioned comprehensive analysis module to the software administrator information terminal.

Images