User power consumption business question and answer method and system based on knowledge graph

Abstract

The invention provides a user power consumption business question and answer method and system based on a knowledge graph, and belongs to the technical field of power business handling. According to the method, entity extraction and association are carried out on the historical customer work orders, the historical customer work orders are converted into triples, the triples are stored in the neo4j database to construct the knowledge graph, and business data are effectively integrated to form a structured knowledge system. By means of the constructed common question template and the VSM algorithm, the user question can be quickly and accurately recognized, a Cypher query statement is utilized to efficiently search for question reasons and solving measures in the knowledge graph, finally, an answer text is generated based on an answer text template to be fed back to the user, a work order is sent, efficient circulation from question consultation to service processing is achieved, and the user experience is improved. The quality and efficiency of the power consumption business question and answer service are comprehensively improved, the service process is optimized, and more professional and convenient service experience is provided for the user.

Description

Technical Field

[0001] The present invention belongs to the technical field of power business handling, and specifically relates to a user power business question and answer method and system based on a knowledge graph. Background Art

[0002] As a graph database based on "entity-relationship-entity" triples, knowledge graph stores knowledge information and their relationships in symbolic form. Entities contain types and attributes, and entities are related to each other through relationships, forming a network-like knowledge structure. This structure makes knowledge graphs have broad application prospects in fields such as information retrieval and question-answering systems. Question-answering systems based on knowledge graphs can retrieve relevant information and provide answers from knowledge graphs by matching questions in natural language. They have been applied in many fields such as medical question-answering, food safety question-answering, book knowledge question-answering, and fault location.

[0003] In recent years, language big model technology has acquired the ability to understand the deep structure and semantics of language through self-supervised learning of large-scale unlabeled text data, and has been widely used in the field of question answering. These technologies can handle complex natural language queries and provide more accurate and intelligent answers. However, in the field of electricity business, information services still mainly rely on manual services, and there are problems such as inaccurate business positioning, poor connection between departments and customers, and difficulty in determining the type of guarantee, which leads to low business processing efficiency, unclear processes, and high requirements for the professional and technical level of seat service personnel, which affects the resolution of customer electricity problems and the improvement of service quality.

[0004] At present, consulting services in the field of electricity consumption are still in the manual service stage. There are problems such as low efficiency, difficulty in business handling, and unclear processes in business positioning, department-customer docking, and guarantee type determination. In addition, there are also high requirements for the professional and technical level of seat service personnel, which affects solving customer electricity consumption problems and improving the quality of electricity consumption services. These problems have seriously affected the efficiency and quality of electricity consumption services, and an intelligent solution is urgently needed to improve service levels and customer satisfaction. Summary of the invention

[0005] In view of this, the present invention provides a user electricity business question-and-answer method and system based on knowledge graph, aiming to construct an electricity business knowledge graph and develop a user electricity business question-and-answer system by combining existing technical means and a long-term customer service question library. The system aims to solve the problems existing in the existing electricity business information service, such as inaccurate business positioning, poor connection between departments and customers, difficulty in determining the type of guarantee, unclear processes, and low efficiency of manual services.

[0006] In order to achieve the above object, the technical solution provided by the present invention is as follows:

[0007] In a first aspect, the present invention provides a user electricity business question-answering method based on a knowledge graph, comprising the following steps:

[0008] Process historical customer work orders, extract entities from the user's requirement statements and the corresponding solution statements given by the staff, associate them, and convert them into entity-relationship-entity triples;

[0009] Store the transformed triples into the neo4j database to build a knowledge graph;

[0010] Build common question templates based on electricity consumption business;

[0011] Using VSM algorithm, the questions raised by users are matched with corresponding question templates;

[0012] After matching the template of the corresponding problem, use Cypher query statements to correspond to the cause and solution in the knowledge graph;

[0013] Generate an answer text based on the answer text template, feedback the corresponding reasons and solutions to the user through the answer text, and send a work order to the relevant business personnel based on user needs.

[0014] Furthermore, in the step of extracting and associating entities in the user's requirement content statement in the work order and the corresponding solution statement given by the staff, it also includes:

[0015] The extracted entities are classified, including business types, business subcategories, business needs, solutions and business personnel. Business types include fault reporting, electricity business, and consulting inquiries. Business subcategories include power outages in a certain area, capacity expansion, meter box failures, and charging pile business. Business needs include power outage information and business processes. Solutions include contacting personnel in relevant departments, filling out platform reports, and seeking professional personnel to handle business outside of responsibilities. Business personnel are specific personnel names.

[0016] Furthermore, in the step of classifying the extracted entities, it also includes: establishing a standard entity table and a standard entity relationship table, the standard entity table records the classified entities and their examples, the standard entity relationship table records the entity relationship types and their examples, and the entity relationship types include subordination, corresponding requirements, corresponding solutions, and centralized responsibility.

[0017] Furthermore, in the step of constructing a common question template based on the electricity consumption business, it includes: selecting typical examples from historical customer work orders, vectorizing user sentences in typical examples using the word frequency method, setting corresponding dictionaries for different business types and business subclasses, and establishing a vector representation based on the word frequency of each sentence in the triple, thereby constructing a common question template.

[0018] Furthermore, the step of matching the question raised by the user with the corresponding question template by using the VSM algorithm includes: vectorizing the question actually raised by the user by using the word frequency method, and calculating the cosine distance to achieve matching with the pre-built question template.

[0019] In a second aspect, the present invention provides a user electricity business question-answering system based on a knowledge graph, comprising:

[0020] The data processing unit is used to process historical customer work orders, extract entities from the user's demand content statements in the work order and the corresponding solution statements given by the staff, associate them, and convert them into entity-relationship-entity triples;

[0021] The knowledge graph generation unit is used to store the transformed triples into the neo4j database and construct the knowledge graph;

[0022] A question and answer template building unit is used to build common question templates based on electricity consumption business;

[0023] A question matching unit, used to match the question raised by the user with the corresponding question template using a VSM algorithm;

[0024] The query unit is used to match the template of the corresponding problem and use Cypher query statements to correspond to the cause and solution in the knowledge graph;

[0025] The answer generation and feedback unit is used to generate answer text based on the answer text template, feedback the corresponding reasons and solutions to the user, and send work orders to relevant business personnel according to the needs.

[0026] Furthermore, the data processing unit further includes:

[0027] The extracted entities are classified, including business types, business subcategories, business needs, solutions and business personnel. Business types include fault reporting, electricity business, and consulting inquiries. Business subcategories include power outages in a certain area, capacity expansion, meter box failures, and charging pile business. Business needs include power outage information and business processes. Solutions include contacting personnel in relevant departments, filling out platform reports, and seeking professional personnel to handle business outside of responsibilities. Business personnel are specific personnel names.

[0028] Furthermore, in classifying the extracted entities, it also includes: establishing a standard entity table and a standard entity relationship table, the standard entity table records the classified entities and their examples, the standard entity relationship table records the entity relationship types and their examples, and the entity relationship types include subordination, corresponding requirements, corresponding solutions, and centralized responsibility.

[0029] Furthermore, in the question and answer template establishment unit, it includes: selecting typical examples from historical customer work orders, vectorizing user sentences in typical examples using the word frequency method, setting corresponding dictionaries for different business types and business subclasses, and establishing vector representations according to the word frequency of each sentence in the triple, thereby constructing commonly used question templates.

[0030] Furthermore, in the question matching unit, it includes: vectorizing the question actually asked by the user using the word frequency method, and calculating the cosine distance to achieve matching with the pre-built question template.

[0031] In summary, the present invention provides a method and system for user electricity business question and answer based on knowledge graph, which extracts entities from historical customer work orders, associates and converts them into triples and stores them in neo4j database to build knowledge graph, effectively integrating business data to form a structured knowledge system. With the help of the constructed common question templates and VSM algorithm, user problems can be quickly and accurately identified, and the cause and solution of the problem can be efficiently found in the knowledge graph using Cypher query statements. Finally, the answer text is generated based on the answer text template to feedback the user and send the work order, realizing the efficient flow from problem consultation to business processing, comprehensively improving the quality and efficiency of electricity business question and answer services, optimizing the service process, and providing users with a more professional and convenient service experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 A flowchart of a user electricity business question-and-answer method based on a knowledge graph provided in an embodiment of the present invention;

[0034] Figure 2 A knowledge graph relationship diagram provided for an embodiment of the present invention;

[0035] Figure 3 A flow chart of a user electricity business question and answer system based on a knowledge graph provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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.

[0037] See also Figure 1 The embodiment of the present invention provides a user electricity business question-answering method based on a knowledge graph, comprising the following steps:

[0038] S1: Process historical customer work orders, extract entities from the user's requirement content statements in the work order and the corresponding solution statements given by the staff, associate them, and convert them into entity-relationship-entity triples.

[0039] It should be noted that natural language processing (NLP) related technologies, such as named entity recognition (NER) algorithms, can be used to identify key entities (such as names of people, organization names, business terms, etc.) from user needs and solution statements, and then associate them into "entity-relationship-entity" triples based on the semantic relationship between the entities.

[0040] Through the above conversion process, the unstructured data in historical work orders can be converted into structured data, laying the foundation for the subsequent construction of the knowledge graph and facilitating the system to store, query and reason about business knowledge.

[0041] S2: Store the transformed triples into the neo4j database to build a knowledge graph.

[0042] It should be noted that neo4j is a graph database that stores data in the form of nodes and relationships. The triple data generated in step S1 is stored in neo4j according to the storage format of the graph database, where each entity is a node and the relationship between entities is an edge, thereby constructing a knowledge graph.

[0043] S3: Build common question templates based on electricity consumption business.

[0044] It should be noted that the commonly used question templates are a series of standardized question templates that are constructed by collecting, organizing and analyzing common problems in the electricity business field, extracting common features, keywords and sentence structures in the questions.

[0045] S4: Use the VSM algorithm to match the questions raised by users with the corresponding question templates.

[0046] It should be noted that the VSM (Vector Space Model) algorithm is an information retrieval model that represents text in the form of vectors and measures the similarity between texts by calculating the similarity between vectors.

[0047] In this step, the user's question and the constructed question template are first converted into vector form, usually by mapping the words in the text into coordinates in the vector space through methods such as the bag-of-words model and TF-IDF (term frequency-inverse document frequency). Then, the similarity between the user's question vector and each question template vector is calculated using a similarity calculation method (such as cosine similarity), and the template with the highest similarity is the matching result.

[0048] S5: After matching the template of the corresponding problem, use Cypher query statements to correspond to the cause and solution in the knowledge graph.

[0049] It should be noted that the Cypher query statement is the query language of the neo4j graph database, which is similar to SQL in relational databases and is used to query and retrieve data in the knowledge graph.

[0050] Cypher query statements can perform precise or fuzzy queries in the knowledge graph based on the attributes, labels, and other conditions of nodes and relationships. When the corresponding question template is matched, the corresponding Cypher query statement is constructed based on the question type and keywords corresponding to the template to find the cause and solution nodes related to the problem in the knowledge graph, thereby accurately obtaining the answer information related to the user's question in the knowledge graph.

[0051] S6: Generate an answer text based on the answer text template, feed back the corresponding reasons and solutions to the user through the answer text, and send a work order to the relevant business personnel based on the user's needs.

[0052] It should be noted that the answer text template is a pre-designed text template with a fixed format and content framework, which is used to generate text to reply to users.

[0053] The information such as the cause and solution obtained in step S5 is filled in according to the format and requirements of the answer text template to generate a complete answer text. The answer text is then fed back to the user, and the relevant work order is sent to the corresponding business personnel for further processing according to the user's needs.

[0054] This embodiment provides a user electricity business question and answer method based on knowledge graph. By introducing knowledge graph technology into the user electricity business question and answer field, business knowledge is integrated and managed in a structured manner, which changes the way of knowledge storage and retrieval in traditional question and answer methods, and improves the efficiency of knowledge utilization and the accuracy of question and answer. It also combines the entity extraction and VSM algorithm of natural language processing to realize an efficient processing flow from historical work order data to user problem matching, which can quickly and accurately understand user problems and locate relevant knowledge, and improve the performance of the question and answer system. Finally, through the linkage between the question and answer system and the business work order, the seamless connection between problem consultation and business processing is realized, the service process of electricity business is optimized, the service quality and efficiency are improved, and a more convenient service experience is provided to users.

[0055] In one embodiment, the step of extracting and associating entities in the user's requirement content statement in the work order and the corresponding solution statement given by the staff also includes:

[0056] The extracted entities are classified, including business types, business subcategories, business needs, solutions and business personnel. Business types include fault reporting, electricity business, and consulting inquiries. Business subcategories include power outages in a certain area, capacity expansion, meter box failures, and charging pile business. Business needs include power outage information and business processes. Solutions include contacting personnel in relevant departments, filling out platform reports, and seeking professional personnel to handle business outside of responsibilities. Business personnel are specific personnel names.

[0057] This embodiment provides a basis for building a more sophisticated and accurate knowledge graph by classifying entities. Entities of different categories are stored in the knowledge graph in a structured manner, and the relationships between entities are clearer because of the classification, which facilitates efficient query and reasoning from the knowledge graph later. At the same time, when users ask questions, they can more accurately match the entities and relationships of the corresponding categories according to the questions, so as to obtain more accurate answers from the knowledge graph and improve the accuracy and professionalism of the responses of the question and answer system. In addition, clear entity classification helps to sort out and optimize business processes. During the work order processing, it is possible to quickly arrange appropriate business personnel to handle the work according to different business types, subcategories, needs and solutions, thereby improving the efficiency of business processing and ensuring the smooth progress of business processes. It is also convenient for statistical analysis and management decision-making of the business. The knowledge graph relationship constructed based on the above-mentioned classified entities is as follows: Figure 2 shown.

[0058] In one embodiment, the step of classifying the extracted entities also includes: establishing a standard entity table and a standard entity relationship table, the standard entity table records the classified entities and their examples, the standard entity relationship table records the entity relationship types and their examples, and the entity relationship types include subordination, corresponding requirements, corresponding solutions, and centralized responsibility.

[0059] The standard entity table and standard entity relationship table are shown in Table 1 and Table 2 respectively:

[0060]

[0061]

[0062] By establishing the above table, business knowledge can be made clearer and easier to understand. In the process of processing work orders and business processes, the relationship between different entities can be quickly clarified, which facilitates the flow and collaboration of business. For example, business personnel can clearly know the solution corresponding to a certain business requirement and the person in charge of the solution, thereby improving the efficiency and coordination of business processing, and also helping new employees to quickly familiarize themselves with the business knowledge system.

[0063] In one embodiment, the step of constructing a commonly used question template based on the electricity consumption business includes: selecting typical examples from historical customer work orders, vectorizing user sentences in typical examples using a word frequency method, setting corresponding dictionaries for different business types and business subclasses, and establishing a vector representation based on the word frequency of each sentence in a triple, thereby constructing a commonly used question template.

[0064] This embodiment selects representative examples that can reflect common electricity business problems from a large number of historical customer work orders. These typical examples cover various business scenarios and user expressions, and they are the basic materials for building question templates. For example, in the historical work orders of electricity business, common questions such as "What should I do if there is a sudden power outage at home" and "How to apply for capacity expansion services" may be screened out as typical examples. By selecting these typical examples, it can be ensured that the constructed question template can cover most of the questions asked by users in actual applications.

[0065] The word frequency method is a method of converting text into vector form. By counting the frequency of each word in the user's sentence, the natural language text is mapped to the vector space. Based on the business types (such as fault repair, electricity business, consultation inquiry) and business subcategories (such as power outage in a belt, capacity expansion, meter box failure, etc.) obtained by the previous classification, the corresponding dictionaries are constructed respectively. The dictionary contains common words and terms in the business field. Combined with the triples (entity-relationship-entity) generated previously, the frequency of words in each typical example sentence in the triple is counted. For example, in a sentence, the word "power outage" appears frequently in the triples related to "power outage in a belt", so the word is given a higher weight in the vector representation. The vector representation established in this way not only takes into account the word frequency of the sentence itself, but also combines the association of business knowledge in the knowledge graph, and more comprehensively reflects the connection between the sentence and business knowledge, thereby constructing a more targeted and practical common problem template.

[0066] In one embodiment, the step of matching the question raised by the user with the corresponding question template using the VSM algorithm includes: vectorizing the question actually raised by the user using the word frequency method, and calculating the cosine distance to achieve matching with the pre-built question template.

[0067] The cosine distance is based on the vector space model (VSM) and is used to measure the similarity between two vectors. In this embodiment, the user question is vectorized by the word frequency method to obtain a vector, and each pre-built question template also exists in the form of a vector.

[0068] When calculating the cosine distance between two vectors, the vector dot product and the vector modulus are used. The formula is as follows:

[0069]

[0070] in, and are two vectors, is their vector dot product, and are their modulus lengths respectively. The value range of cosine distance is between ([-1, 1]). The closer the value is to 1, the smaller the angle between the two vectors is, that is, the higher the text similarity is; the closer the value is to -1, the larger the angle between the two vectors is, the lower the similarity is. When it is actually applied to text matching, we usually focus on the case of high similarity, that is, the case where the cosine distance is close to 1.

[0071] In the electricity business scenario, users ask questions in a variety of ways. Through word frequency normalization and cosine distance calculation, various user expressions can be scientifically matched with pre-built question templates. For example, users may use different word orders and words to describe meter failure problems. This method can ignore the surface differences in language, grasp the core semantics, and accurately find the most matching question template, thereby improving matching accuracy.

[0072] Based on the above embodiments, the user electricity business question-answering method based on knowledge graph provided by the present invention has the following advantages:

[0073] (1) It effectively solves the problem of being unable to quickly locate a specific business when user needs are unclear. Through the structured knowledge system of the knowledge graph, it is possible to deeply analyze user problems, accurately grasp user intentions, quickly match corresponding electricity consumption businesses, and improve business processing efficiency.

[0074] (2) Compared with traditional manual services, this method does not have the problem of users having difficulty in handling business due to the uneven technical levels of service agents. It is based on established algorithms and knowledge graphs, and can provide stable and consistent service quality to ensure the smoothness of user business handling.

[0075] (3) Most large language models require a large amount of data for training, and are prone to problems such as confusing sentences and logical contradictions. However, this method is based on the knowledge graph, does not require complex deep learning training, and has low requirements on machine performance. At the same time, with the clear logical structure of the knowledge graph, it can effectively avoid logical errors and ensure the accuracy and reliability of the answer.

[0076] Based on the same inventive concept, the embodiment of the present application also provides a user electricity business question and answer system based on a knowledge graph for implementing the user electricity business question and answer method based on a knowledge graph. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiment of the user electricity business question and answer system based on a knowledge graph provided below can be referred to the limitations of the user electricity business question and answer method based on a knowledge graph above, and will not be repeated here.

[0077] The embodiment of the present invention further provides a user electricity business question-answering system based on a knowledge graph, comprising:

[0078] The data processing unit is used to process historical customer work orders, extract entities from the user's demand content statements in the work order and the corresponding solution statements given by the staff, associate them, and convert them into entity-relationship-entity triples;

[0079] The knowledge graph generation unit is used to store the transformed triples into the neo4j database and construct the knowledge graph;

[0080] A question and answer template building unit is used to build common question templates based on electricity consumption business;

[0081] A question matching unit, used to match the question raised by the user with the corresponding question template using a VSM algorithm;

[0082] The query unit is used to match the template of the corresponding problem and use Cypher query statements to correspond to the cause and solution in the knowledge graph;

[0083] The answer generation and feedback unit is used to generate answer text based on the answer text template, feedback the corresponding reasons and solutions to the user, and send work orders to relevant business personnel according to the needs.

[0084] Furthermore, the data processing unit further includes:

[0085] The extracted entities are classified, including business types, business subcategories, business needs, solutions and business personnel. Business types include fault reporting, electricity business, and consulting inquiries. Business subcategories include power outages in a certain area, capacity expansion, meter box failures, and charging pile business. Business needs include power outage information and business processes. Solutions include contacting personnel in relevant departments, filling out platform reports, and seeking professional personnel to handle business outside of responsibilities. Business personnel are specific personnel names.

[0086] Furthermore, in classifying the extracted entities, it also includes: establishing a standard entity table and a standard entity relationship table, the standard entity table records the classified entities and their examples, the standard entity relationship table records the entity relationship types and their examples, and the entity relationship types include subordination, corresponding requirements, corresponding solutions, and centralized responsibility.

[0087] Furthermore, in the question and answer template establishment unit, it includes: selecting typical examples from historical customer work orders, vectorizing user sentences in typical examples using the word frequency method, setting corresponding dictionaries for different business types and business subclasses, and establishing vector representations according to the word frequency of each sentence in the triple, thereby constructing commonly used question templates.

[0088] Furthermore, in the question matching unit, it includes: vectorizing the question actually asked by the user using the word frequency method, and calculating the cosine distance to achieve matching with the pre-built question template.

[0089] Figure 3This is a flow chart of the user electricity business question and answer system based on the knowledge graph. The figure shows that after the user asks a question, the system processes the question data and extracts keywords; then, the system constructs a graph database with triples converted from the data in the historical question library, and then stores it in neo4j to form a knowledge graph; then, Cypher statement queries are used, combined with question templates and dictionaries, to find answers in the knowledge graph, and finally the answers to the questions are fed back to the user.

[0090] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0091] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0092] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0093] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0094] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A user electricity business question-answering method based on knowledge graph, characterized in that: The steps include: Process historical customer work orders, extract entities from the user's requirement statements and the corresponding solution statements given by the staff, associate them, and convert them into entity-relationship-entity triples; The transformed triples are stored in the neo4j database to construct a knowledge graph; Build common question templates based on electricity consumption business; Using VSM algorithm, the questions raised by users are matched with corresponding question templates; After matching the template of the corresponding problem, a Cypher query statement is used to correspond to the cause and solution in the knowledge graph; An answer text is generated based on the answer text template, the corresponding reasons and solutions are fed back to the user through the answer text, and a work order is sent to relevant business personnel based on user needs.

2. The user electricity business question-answering method based on knowledge graph according to claim 1 is characterized in that: The step of extracting entities from the user's requirement content statement in the work order and the corresponding solution statement given by the staff and associating them also includes: The extracted entities are classified, including business types, business subcategories, business needs, solutions and business personnel, where the business types include fault reporting, electricity usage services, and consulting inquiries; the business subcategories include power outages in a certain area, capacity expansion, meter box failures, and charging pile services; the business needs include power outage information and business processes; the solutions include contacting personnel in relevant departments, filling out platform reports, and seeking professional personnel to handle business outside of responsibilities; the business personnel are specific personnel names.

3. The user electricity business question-answering method based on knowledge graph according to claim 2 is characterized in that: The step of classifying the extracted entities also includes: establishing a standard entity table and a standard entity relationship table, wherein the standard entity table records the classified entities and their examples, and the standard entity relationship table records the entity relationship types and their examples, wherein the entity relationship types include subordination, corresponding requirements, corresponding solutions, and centralized responsibility.

4. The user electricity business question-answering method based on knowledge graph according to claim 1 is characterized in that: The step of constructing a commonly used question template based on the electricity consumption business includes: selecting typical examples from the historical customer work orders, vectorizing the user sentences in the typical examples using the word frequency method, setting corresponding dictionaries for different business types and business subclasses, and establishing a vector representation based on the word frequency of each sentence in the triple, thereby constructing a commonly used question template.

5. The user electricity business question-answering method based on knowledge graph according to claim 4 is characterized in that: The step of matching the question raised by the user with the corresponding question template by using the VSM algorithm includes: vectorizing the question actually raised by the user by using the word frequency method, and calculating the cosine distance to achieve matching with the pre-built question template.

6. A user electricity business question-answering system based on knowledge graph, characterized in that: include: The data processing unit is used to process historical customer work orders, extract entities from the user's demand content statements in the work order and the corresponding solution statements given by the staff, associate them, and convert them into entity-relationship-entity triples; A knowledge graph generation unit, used for storing the converted triples into a neo4j database to construct a knowledge graph; A question and answer template building unit is used to build common question templates based on electricity consumption business; A question matching unit, used to match the question raised by the user with the corresponding question template using a VSM algorithm; A query unit, used to match the template corresponding to the problem, and then use Cypher query statements to find the cause and solution in the knowledge graph; The answer generation and feedback unit is used to generate answer text based on the answer text template, feedback the corresponding reasons and solutions to the user, and send work orders to relevant business personnel according to the needs.

7. The user electricity business question and answer system based on knowledge graph according to claim 6 is characterized in that: The data processing unit further includes: The extracted entities are classified, including business types, business subcategories, business needs, solutions and business personnel, where the business types include fault reporting, electricity usage services, and consulting inquiries; the business subcategories include power outages in a certain area, capacity expansion, meter box failures, and charging pile services; the business needs include power outage information and business processes; the solutions include contacting personnel in relevant departments, filling out platform reports, and seeking professional personnel to handle business outside of responsibilities; the business personnel are specific personnel names.

8. The user electricity business question and answer system based on knowledge graph according to claim 7 is characterized in that: In classifying the extracted entities, it also includes: establishing a standard entity table and a standard entity relationship table, wherein the standard entity table records the classified entities and their examples, and the standard entity relationship table records the entity relationship types and their examples, and the entity relationship types include subordination, corresponding requirements, corresponding solutions, and centralized responsibility.

9. The user electricity business question and answer system based on knowledge graph according to claim 6 is characterized in that: The question and answer template establishment unit includes: selecting typical examples from the historical customer work orders, vectorizing the user sentences in the typical examples using the word frequency method, setting corresponding dictionaries for different business types and business subclasses, and establishing a vector representation based on the word frequency of each sentence in the triple, thereby constructing a commonly used question template.

10. The user electricity business question and answer system based on knowledge graph according to claim 9 is characterized in that: In the question matching unit, it includes: using the word frequency method to vectorize the question actually raised by the user, and calculating the cosine distance to achieve matching with the pre-built question template.

Images