CN-121998650-A - Intelligent power customer service method and related equipment

CN121998650ACN 121998650 ACN121998650 ACN 121998650ACN-121998650-A

Abstract

The disclosure provides an intelligent power customer service method and related equipment, the method comprises the steps of obtaining first user input information, conducting semantic analysis on the first user input information to determine first user service information, enabling the first user service information to be used for indicating a service task, enabling the first user service information to be used for being matched with a corresponding service operation flow, enabling the service operation flow to comprise a directed graph formed by a plurality of service operation nodes, enabling node input information corresponding to the service operation nodes to be used for selecting a corresponding path along the directed graph to execute the service task, enabling the node input information to be used for conducting semantic analysis on second user input information of the service operation nodes based on a user, and returning an execution result of the service task to the user.

Inventors

ZHANG QIANFU
XIE QING
XU QIANG
WANG XIAOMIN
GUO LINGHUI
ZHANG PENG
CAO LU
WANG WEI
FANG HONGWANG
HAO QINGLI
YAN FENG
YUE YING
LIU LEI
WANG SHICHAO

Assignees

北京中电普华信息技术有限公司

Dates

Publication Date: 20260508
Application Date: 20251209

Claims (10)

1. A smart power customer service method, comprising: Acquiring first user input information, wherein the first user input information comprises at least one of voice data, text data or video data; Carrying out semantic analysis on the first user input information to determine first user service information, wherein the first user service information is used for indicating a service task; matching corresponding service operation flows based on the first user service information, wherein the service operation flows comprise directed graphs formed by a plurality of service operation nodes; Selecting a corresponding path along the directed graph based on node input information corresponding to the service operation node to execute the service task, wherein the node input information is obtained by semantic analysis based on second user input information of a user aiming at the service operation node; and returning the execution result of the service task to the user.
2. The method of claim 1, wherein the node input information is semantically analyzed based on second user input information of a user for the service operation node, comprising: the node service information is used for prompting the user to execute the information required by the service operation of the service operation node; receiving the second user input information of the user aiming at the node service information; And carrying out semantic analysis on the second user input information to obtain the node input information.
3. The method of claim 1, wherein selecting a corresponding path along the directed graph based on node input information corresponding to the serving operational node comprises: Responsive to detecting that the node input information meets a first condition of the service operation node, selecting a first path associated with the service operation node in the directed graph; Responsive to detecting that the node input information satisfies a second condition of the service operation node, selecting a second path associated with the service operation node in the directed graph; And determining a service abnormality in response to detecting that the node input information does not satisfy any condition of the service operation node.
4. The method of claim 1, further comprising: and calling a corresponding service processor based on the service operation interface corresponding to the service operation node so as to process the node input information and execute the service operation corresponding to the service operation node.
5. The method of claim 1, wherein semantically analyzing the first user input information to determine first user service information comprises: performing data processing on the first user input information to obtain intermediate input information; Extracting semantic features based on the intermediate input information to obtain intermediate input features; determining preliminary user service information based on the intermediate input features; verifying the preliminary user service information based on preset service information; Determining that the preliminary user service information is the first user service information in response to successful verification of the preliminary user service information; and in response to unsuccessful verification of the preliminary user service information, updating the preliminary user service information based on the preset service information to obtain the first user service information.
6. The method of claim 2, wherein outputting node service information corresponding to the service operation node to a user comprises: Determining a user state based on the first user input information and/or the second user input information; updating the content and/or the times of the node service information in response to the user state not meeting the state condition; and/or the number of the groups of groups, And generating the node service information based on the user history interaction information, the context information of the current interaction and the node information template corresponding to the service operation node.
7. The method of claim 1, further comprising at least one of: generating guide information to guide a user to input supplemental information in response to the first user service information being unable to be determined based on the first user input information; Performing semantic analysis based on the user input supplemental information and the first user input information to determine the first user service information; determining a service anomaly in response to the first user service information being unable to be determined based on the user input supplemental information and the first user input information; And executing corresponding exception handling operation based on the exception type of the service exception and the service exception strategy.
8. A smart power customer service device comprising: The system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring first user input information, and the first user input information comprises at least one of voice data, text data or video data; the semantic analysis module is used for carrying out semantic analysis on the first user input information to determine first user service information, wherein the first user service information is used for indicating a service task; The flow matching module is used for matching corresponding service operation flows based on the first user service information, wherein the service operation flows comprise directed graphs formed by a plurality of service operation nodes; the flow execution module is used for selecting a corresponding path along the directed graph based on node input information corresponding to the service operation node so as to execute the service task, wherein the node input information is obtained by semantic analysis based on second user input information of the service operation node; and the result returning module is used for returning the execution result of the service task to the user.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 7 when the program is executed.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 7.

Description

Intelligent power customer service method and related equipment Technical Field The present disclosure relates to the field of power, and in particular, to an intelligent power customer service method and related devices. Background With the continuous penetration of digital transformation in the power industry, customer service is taken as an important bridge for communication between power enterprises and users, and the service quality directly influences the customer satisfaction degree and the market competitiveness of the enterprises. The traditional electric power customer service system mainly depends on a manual agent or a question-answering system based on rules to respond to services, and has the problems of low efficiency, slow response, insufficient professionals and the like. In recent years, with the development of artificial intelligence technology, particularly the breakthrough of large language models (Large Language Models, LLMs) in natural language understanding and generation, new possibilities are provided for upgrading intelligent customer service systems. However, applying large language models to power customer service scenarios still faces a number of challenges. Firstly, the electric power business has high professional and complex performance, relates to a plurality of subdivision fields such as electricity consultation, electricity fee inquiry, fault repair, business handling and the like, and provides higher requirements for accuracy of customer intention recognition. Secondly, customer appeal is various and expression mode is not unified, and traditional single intention recognition method is difficult to satisfy actual demand. In addition, large language models may have a "illusion" phenomenon in generating replies, i.e., generating content deviation facts or business data, leading to misleading users and even introducing legal risks. Therefore, there is a need for an intelligent power customer service method that can accurately identify customer intent, effectively match standardized service flows, and achieve efficient response while ensuring business accuracy. The following technical problems exist in the existing electric power intelligent customer service system: the customer intention recognition is inaccurate, and the existing system is difficult to accurately recognize the actual intention of the customer when facing diversified customer expressions, so that the service path is wrong or information is lost. The service flow is not standard, the traditional customer service system lacks a standardized service chain design, and the service flow cannot be dynamically adjusted according to the intention of a customer. The illusion risk in the large language model application is high, namely when the large language model is directly used for generating replies, contents which are inconsistent with actual service data easily appear, and the credibility and the safety of the service are affected. The context understanding capability is weak, and the lack of an effective context modeling mechanism in multiple rounds of conversations affects the consistency of interaction and user experience. Service data integration is difficult, and an effective docking mechanism with a power service center station is lacking, so that relevant service data of clients cannot be acquired and fused in real time. Disclosure of Invention Accordingly, an object of the present disclosure is to provide an intelligent power customer service method and related devices. In a first aspect of the present disclosure, there is provided an intelligent power customer service method, comprising: Acquiring first user input information, wherein the first user input information comprises at least one of voice data, text data or video data; Carrying out semantic analysis on the first user input information to determine first user service information, wherein the first user service information is used for indicating a service task; matching corresponding service operation flows based on the first user service information, wherein the service operation flows comprise directed graphs formed by a plurality of service operation nodes; Selecting a corresponding path along the directed graph based on node input information corresponding to the service operation node to execute the service task, wherein the node input information is obtained by semantic analysis based on second user input information of a user aiming at the service operation node; and returning the execution result of the service task to the user. In a second aspect of the present disclosure, there is provided an intelligent power customer service device comprising: The system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring first user input information, and the first user input information comprises at least one of voice data, text data or video data; the semantic analysis module