Intelligent question and answer method, device and equipment for optical module digital manufacturing scene based on knowledge graph and medium

Abstract

The invention discloses an intelligent question and answer method, device, equipment and medium for an optical module digital manufacturing scene based on a knowledge graph, and belongs to the technical field of intelligent manufacturing. Forming a knowledge graph containing equipment information, process flow and time node information; the method comprises the following steps: analyzing a natural language problem input by a user based on a preset domain feature word library of an ultra-high-speed scene, and classifying the natural language problem into equipment query, process query or time query; according to the classification result of the natural language questions, the natural language questions are converted into graph database query statements, and retrieval is carried out in the knowledge graph; and converting a query result into a natural language form understood by the user, dynamically responding and returning to the user. The information acquisition efficiency and the decision support level in the manufacturing scene are improved, and the optimization of the production process and the improvement of the production efficiency are ensured.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent manufacturing technology, and more specifically, relates to an intelligent question-answering method, device, equipment and medium for ultra-high-speed optical module digital manufacturing scenarios based on a knowledge graph. Background Art

[0002] In recent years, ultra-high-speed optical module manufacturing technology has developed rapidly, especially driven by the demand in 5G communications, data centers, high-performance computing and other fields. my country's optical module manufacturing industry has gradually occupied an important position in ultra-high-speed transmission scenarios. Ultra-high speed usually refers to optical communication modules with a transmission rate of 400Gbps or more. Compared with traditional modules, it has higher requirements for accuracy, efficiency and complexity in the design, manufacturing and testing links.

[0003] In this context, the intelligent question-answering method based on knowledge graph plays an important role. This method can not only quickly and accurately answer questions related to equipment, processes, time, etc. in the manufacturing process, but also improve the level of automation and intelligence in the production process, which is beneficial to improving production efficiency, optimizing resource allocation and enhancing decision-making support capabilities, and promoting the high-quality development of my country's ultra-high-speed optical module manufacturing industry.

[0004] Knowledge graph technology has powerful data association and organization capabilities. By integrating and modeling multi-source heterogeneous data, it can achieve structured representation of information and mining of complex relationships. In addition, knowledge graphs also have semantic understanding and reasoning capabilities, which can quickly discover potential associations in massive data and provide more accurate queries and intelligent decision support. It not only improves the efficiency of information retrieval, but also provides a solid foundation for data-driven analysis and optimization, and is widely used in intelligent question answering, recommendation systems and other fields. However, this technology still faces many challenges in practical applications. On the one hand, the degree of automation of knowledge acquisition and updating is not enough, and the diversity and complexity of data increase the difficulty of graph construction; on the other hand, the accuracy of knowledge reasoning and understanding still needs to be improved, especially when dealing with complex relationships and uncertain information. In addition, existing knowledge graph applications also have bottlenecks in scale expansion and real-time response, lack of sufficient dynamic update and adaptive capabilities, which restricts its comprehensive application in more complex scenarios. Summary of the invention

[0005] In response to the above defects or improvement needs of the prior art, the present invention proposes an intelligent question-answering method, device, equipment and medium for digital manufacturing scenarios of ultra-high-speed optical modules based on knowledge graphs, which improves the information acquisition efficiency and decision-making support level in manufacturing scenarios, and ensures the optimization of production processes and the improvement of production efficiency.

[0006] To achieve the above objectives, according to one aspect of the present invention, a method for intelligent question answering in ultra-high-speed optical module digital manufacturing scenarios based on knowledge graph is provided, comprising:

[0007] Model the entities including equipment, processes and time involved in the ultra-high-speed optical module manufacturing scenario to form a knowledge graph containing equipment information, process flow and time node information;

[0008] Based on the preset domain feature word library of ultra-high-speed scenarios, it parses the natural language questions input by users, extracts the entities and feature words of the natural language questions, and then classifies the natural language questions into equipment queries, process queries or time queries;

[0009] According to the classification results of natural language questions, the natural language questions are converted into graph database query statements, and data retrieval related to ultra-high-speed optical module manufacturing is performed in the knowledge graph to extract the corresponding equipment information, process flow or time node information;

[0010] The query results are converted into natural language form that users can understand, and the dynamic response is returned to the user.

[0011] In some optional implementation schemes, the entities including equipment, processes and time involved in the ultra-high-speed optical module manufacturing scenario are modeled, and a knowledge graph including equipment information, process flow and time node information is formed, including:

[0012] Identify and abstractly model entities including equipment information, process flow, and time node information in ultra-high-speed optical module manufacturing scenarios, and clarify the attributes and characteristics of each entity;

[0013] Preprocess the multi-source heterogeneous data by integrating them from different data sources;

[0014] Determine the association relationship between entities according to the attributes and characteristics of each entity, and establish a relationship graph between entities based on the association relationship between entities;

[0015] Integrate preprocessed multi-source heterogeneous data to form a knowledge graph containing multi-source information.

[0016] In some optional implementation schemes, the domain feature word library based on ultra-high-speed scenarios parses the natural language questions input by users, extracts entities and feature words of the natural language questions, and further classifies the natural language questions into equipment queries, process queries or time queries, including:

[0017] Through natural language processing technology, the natural language questions input by users are segmented and parsed, entities and characteristic words in the natural language questions are identified, and the target information of the natural language questions is determined according to the entities and characteristic words in the natural language questions;

[0018] Combined with the preset domain feature word library for ultra-high-speed scenarios, the Aho-Corasick automaton algorithm is used to match the target information in natural language questions and classify them according to the target information;

[0019] Transform the categorized questions into executable query format.

[0020] In some optional implementation schemes, according to the classification result of the natural language question, the natural language question is converted into a graph database query statement, and data retrieval related to ultra-high-speed optical module manufacturing is performed in the knowledge graph to extract corresponding equipment information, process flow or time node information, including:

[0021] According to the classification results of the natural language questions, entities and relationships between entities related to the natural language questions are identified;

[0022] Combined with the structure in the knowledge graph, the identified entities and relationships between them are mapped to the grammatical rules of graph database queries, and query statements that conform to Cypher syntax are dynamically generated;

[0023] Execute the generated Cypher query statement, retrieve the corresponding information from the knowledge graph, and return structured data.

[0024] In some optional implementation schemes, converting the query results into a natural language form that the user understands, dynamically responding and returning to the user, includes:

[0025] Analyze entities, attributes, and relationships related to natural language questions based on the structured data returned by Cypher query statements;

[0026] Use template generation or natural language generation technology to organize entities, attributes and relationships related to natural language questions into coherent natural language expressions;

[0027] Based on the context and expectations of the user's question, an answer is generated and fed back to the user.

[0028] In some optional embodiments, the method further comprises:

[0029] Combine user feedback and historical query data to optimize and update the knowledge graph.

[0030] In some optional implementation schemes, the combining of user feedback and historical query data to optimize the knowledge graph and update the knowledge graph includes:

[0031] Identify common questions, query patterns, and deficiencies in intelligent question-answering systems based on user feedback, interaction data, and historical query data;

[0032] Use the identified common problems, query patterns, and existing deficiencies to adjust and optimize the knowledge graph, update the knowledge graph content, and improve the accuracy of question parsing and answer generation;

[0033] Through the self-learning mechanism, new data is continuously accumulated and analyzed.

[0034] According to another aspect of the present invention, there is provided an intelligent question-answering device for ultra-high-speed optical module digital manufacturing scenarios based on a knowledge graph, comprising:

[0035] A knowledge graph construction module is used to model entities including equipment, processes, and time involved in the ultra-high-speed optical module manufacturing scenario to form a knowledge graph containing equipment information, process flow, and time node information;

[0036] The natural language parsing module is used to parse the natural language questions input by users based on the preset domain feature word library of ultra-high-speed scenarios, extract the entities and feature words of the natural language questions, and then classify the natural language questions into equipment queries, process queries or time queries;

[0037] A query statement generation module is used to convert natural language questions into graph database query statements based on the classification results of natural language questions, perform data retrieval related to ultra-high-speed optical module manufacturing in the knowledge graph, and extract corresponding equipment information, process flow or time node information;

[0038] The answer generation module is used to convert the query results into a natural language form that the user can understand, dynamically respond and return it to the user.

[0039] According to another aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0040] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0041] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0042] The present invention provides an intelligent question-answering method for ultra-high-speed optical module digital manufacturing scenarios based on knowledge graph technology. First, by constructing a knowledge graph of multi-dimensional information such as equipment, process, time, etc., data integration and structured management in the manufacturing process are realized. Then, natural language processing technology is used to parse user questions, and key entities are quickly extracted and efficient Cypher query statements are generated in combination with domain feature vocabulary. Finally, natural language answers are dynamically generated based on the query results to respond to user needs in real time. Aiming at the problem of low efficiency of complex information retrieval and decision support in ultra-high-speed optical module manufacturing scenarios, the present invention proposes an intelligent question-answering solution based on knowledge graph, which significantly improves the accuracy and response speed of information query, is beneficial to optimize production processes, improve production efficiency, and provide strong decision support for intelligent manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the overall architecture of a method provided by an embodiment of the present invention;

[0044] Figure 2 It is a schematic diagram of a process of constructing a knowledge graph provided by an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of a problem analysis process provided by an embodiment of the present invention;

[0046] Figure 4 It is a schematic diagram of a query statement and answer generation process provided by an embodiment of the present invention;

[0047] Figure 5 It is a schematic diagram of an intelligent question-answering system optimization and self-learning process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0049] like Figure 1 As shown, an embodiment of the present invention provides an intelligent question-answering method for ultra-high-speed optical module digital manufacturing scenarios based on a knowledge graph, including:

[0050] Step S1: Construct a knowledge graph. Using graph database technology, key entities such as equipment, processes, and time involved in the ultra-high-speed optical module manufacturing scenario are modeled, and a knowledge graph covering high-precision equipment information, highly complex process flows, and strict time management nodes is formed through structured processing.

[0051] Step S2: Parsing natural language questions. Natural language processing technology is used to parse the natural language questions input by the user in combination with the vocabulary and feature definitions for the specific fields of ultra-high-speed scenarios. Key entities and feature words are extracted through the Aho-Corasick automaton algorithm, and the questions are classified into ultra-high-speed module manufacturing specific question types such as equipment query, process query, and time query.

[0052] Step S3: Generate Cypher query statements. According to the question classification results, the user question is converted into a Cypher query statement of the graph database. Perform data retrieval related to ultra-high-speed optical module manufacturing to extract information such as high-precision equipment status, complex process parameters, or real-time production plans.

[0053] Step S4: Dynamically generate answers. Based on the query results, dynamically generate structured natural language answers. Combining the template method and machine learning model, the search results are converted into a natural language form that is easy for users to understand, and the core elements of ultra-high-speed module production are explained and displayed to ensure the accuracy and pertinence of the answers.

[0054] Step S5: System optimization and self-learning. Combine user feedback and historical query data to continuously optimize the knowledge graph and question-answering system. Regularly update the knowledge base content related to ultra-high-speed optical module manufacturing, improve the ability to analyze high-precision, high-complexity and high-real-time problems, and enhance the intelligence level of the system in ultra-high-speed scenarios.

[0055] Furthermore, if Figure 2 As shown in the figure, building a knowledge graph is a complex and systematic process involving multiple steps such as data collection, processing, modeling, and relationship mapping. First, it is necessary to identify and abstractly model key entities such as equipment, processes, and time in the optical module manufacturing scenario, and clarify the attributes and characteristics of each entity. Then, by integrating multi-source heterogeneous data from different data sources, the data is cleaned and structured to ensure the integrity and consistency of the data. Finally, based on the association relationship between entities (such as the connection between equipment and processes, and the dependency between processes and time nodes), a relationship graph between entities is established to form a knowledge graph containing multi-dimensional information to support intelligent question-answering and data analysis applications.

[0056] Furthermore, if Figure 3As shown in the figure, parsing natural language questions is a key step in the intelligent question-answering system, which involves semantic understanding of user input, extraction of key information, and classification of questions. First, the system uses natural language processing technology to segment and parse the text input by the user, and identifies entities and feature words in the sentence. The identified entities and feature words are used to determine the core information of the question, and are classified according to this information to generate the corresponding query format to support the retrieval and answer generation of the knowledge graph. Then, combined with the customized domain feature word library, the Aho-Corasick automaton algorithm is used to extract and match the key information in the question, and the questions are preliminarily classified according to categories such as equipment, process, and time. Finally, the classified questions are converted into executable query formats to provide a basis for subsequent knowledge graph queries and answer generation.

[0057] Furthermore, if Figure 4 As shown in the figure, generating Cypher query statements is a key step to achieve efficient retrieval of knowledge graphs. First, based on the parsed natural language questions and their classification results, the system further extracts relevant entities and the relationships between them, and converts them into query statements that conform to Cypher syntax for retrieving accurate information from the knowledge graph. Then, combined with the structure in the knowledge graph, the identified entities and relationships are mapped to the specific grammatical rules of graph database queries, and query statements that conform to Cypher syntax are dynamically generated. Finally, the system executes the generated Cypher query statements to retrieve the corresponding information from the knowledge graph, providing accurate data support for intelligent question and answer.

[0058] Furthermore, dynamically generating answers is a key step in the intelligent question-answering system to convert search results into user-understandable information. First, the system analyzes the entities, attributes, and relationships related to the question based on the structured data returned by the Cypher query statement. Then, template generation or natural language generation technology is used to organize this data into coherent natural language expressions to ensure that the information is accurate and easy to understand. Finally, the system generates personalized and structured answers based on the context and expectations of the user's question, and feeds back to the user in a clear and concise manner.

[0059] Furthermore, if Figure 5 As shown in the figure, system optimization and self-learning are key processes to ensure the continuous improvement of the performance and adaptiveness of the intelligent question-answering system. First, the system identifies common problems, query patterns, and possible deficiencies in the system based on user feedback and interaction data. Then, the knowledge graph and question-answering system are adjusted and optimized using this data, the graph content is updated, and the accuracy of question parsing and answer generation is improved. Finally, the system continuously accumulates and analyzes new data through a self-learning mechanism, gradually improving its ability to deal with complex problems, and achieving continuous optimization of the system and improvement of its intelligence level.

[0060] In particular, system optimization and self-learning not only rely on user feedback, but also proactively identify knowledge blind spots and system performance bottlenecks by analyzing historical query data and user behavior patterns. First, the system pays special attention to high-frequency and complex questions, optimizing the structure of the knowledge graph and question-answering logic for these problems. Then, combined with machine learning algorithms, the system gradually improves the accuracy of question parsing and the naturalness of answer generation through automated training and fine-tuning. Finally, the system achieves adaptive evolution in the process of continuous updating, and has the ability to cope with diverse needs in dynamic scenarios, thereby further improving the reliability and practicality of intelligent question-answering.

[0061] Through the above technical solution, the present invention fully considers the complexity and dynamics of multi-source heterogeneous data in the manufacturing process, and improves the ability to accurately analyze and reason about multi-dimensional information such as equipment, processes, and time by building and updating high-precision knowledge graphs, combining domain feature lexicons and natural language processing technology. By adopting an adaptive graph database query optimization strategy, efficient data retrieval and intelligent question answering are achieved, which significantly improves the efficiency of information acquisition and decision support in manufacturing scenarios, ensuring the optimization of production processes and the improvement of production efficiency.

[0062] Another embodiment of the present invention further provides an intelligent question-answering device for ultra-high-speed optical module digital manufacturing scenarios based on a knowledge graph, comprising:

[0063] A knowledge graph construction module is used to model entities including equipment, processes, and time involved in the ultra-high-speed optical module manufacturing scenario to form a knowledge graph containing equipment information, process flow, and time node information;

[0064] The natural language parsing module is used to parse the natural language questions input by users based on the preset domain feature word library of ultra-high-speed scenarios, extract the entities and feature words of the natural language questions, and then classify the natural language questions into equipment queries, process queries or time queries;

[0065] A query statement generation module is used to convert natural language questions into graph database query statements based on the classification results of natural language questions, perform data retrieval related to ultra-high-speed optical module manufacturing in the knowledge graph, and extract corresponding equipment information, process flow or time node information;

[0066] The answer generation module is used to convert the query results into a natural language form that the user can understand, dynamically respond and return it to the user.

[0067] The specific implementation of each module can refer to the description of the above method embodiment, and the embodiment of the present invention will not be repeated.

[0068] Another embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods when executing the computer program.

[0069] Another embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0070] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, and two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0071] It will be easily understood by those skilled in the art that 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 and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent question-answering method for ultra-high-speed optical module digital manufacturing scenarios based on knowledge graph, characterized in that: include: Model the entities including equipment, processes and time involved in the ultra-high-speed optical module manufacturing scenario to form a knowledge graph containing equipment information, process flow and time node information; Based on the preset domain feature word library of ultra-high-speed scenarios, it parses the natural language questions input by users, extracts the entities and feature words of the natural language questions, and then classifies the natural language questions into equipment queries, process queries or time queries; According to the classification results of natural language questions, the natural language questions are converted into graph database query statements, and data retrieval related to ultra-high-speed optical module manufacturing is performed in the knowledge graph to extract the corresponding equipment information, process flow or time node information; The query results are converted into natural language form that users can understand, and the dynamic response is returned to the user.

2. The method according to claim 1, characterized in that The entities including the equipment, process and time involved in the ultra-high-speed optical module manufacturing scenario are modeled, and a knowledge graph containing equipment information, process flow and time node information is formed, including: Identify and abstractly model entities including equipment information, process flow, and time node information in ultra-high-speed optical module manufacturing scenarios, and clarify the attributes and characteristics of each entity; Preprocess the multi-source heterogeneous data by integrating them from different data sources; Determine the association relationship between entities according to the attributes and characteristics of each entity, and establish a relationship graph between entities based on the association relationship between entities; Integrate preprocessed multi-source heterogeneous data to form a knowledge graph containing multi-source information.

3. The method according to claim 2, characterized in that The domain feature word library based on the ultra-high-speed scenario parses the natural language questions input by the user, extracts the entities and feature words of the natural language questions, and then classifies the natural language questions into equipment queries, process queries or time queries, including: Through natural language processing technology, the natural language questions input by users are segmented and parsed, entities and characteristic words in the natural language questions are identified, and the target information of the natural language questions is determined according to the entities and characteristic words in the natural language questions; Combined with the preset domain feature word library for ultra-high-speed scenarios, the Aho-Corasick automaton algorithm is used to match the target information in natural language questions and classify them according to the target information; Transform the categorized questions into executable query format.

4. The method according to claim 3, characterized in that According to the classification result of the natural language question, the natural language question is converted into a graph database query statement, and data retrieval related to ultra-high-speed optical module manufacturing is performed in the knowledge graph to extract corresponding equipment information, process flow or time node information, including: According to the classification results of the natural language questions, entities and relationships between entities related to the natural language questions are identified; Combined with the structure in the knowledge graph, the identified entities and relationships between them are mapped to the grammatical rules of graph database queries, and query statements that conform to Cypher syntax are dynamically generated; Execute the generated Cypher query statement, retrieve the corresponding information from the knowledge graph, and return structured data.

5. The method according to claim 4, characterized in that The step of converting the query result into a natural language form that the user understands, dynamically responding and returning the result to the user includes: Analyze entities, attributes, and relationships related to natural language questions based on the structured data returned by Cypher query statements; Use template generation or natural language generation technology to organize entities, attributes and relationships related to natural language questions into coherent natural language expressions; Based on the context and expectations of the user's question, an answer is generated and fed back to the user.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: Combine user feedback and historical query data to optimize and update the knowledge graph.

7. The method according to claim 6, characterized in that The method of combining user feedback and historical query data to optimize the knowledge graph and update the knowledge graph includes: Identify common questions, query patterns, and deficiencies in intelligent question-answering systems based on user feedback, interaction data, and historical query data; Use the identified common problems, query patterns, and existing deficiencies to adjust and optimize the knowledge graph, update the knowledge graph content, and improve the accuracy of question parsing and answer generation; Through the self-learning mechanism, new data is continuously accumulated and analyzed.

8. An intelligent question-answering device for ultra-high-speed optical module digital manufacturing scenarios based on knowledge graph, characterized in that: include: A knowledge graph construction module is used to model entities including equipment, processes, and time involved in the ultra-high-speed optical module manufacturing scenario to form a knowledge graph containing equipment information, process flow, and time node information; The natural language parsing module is used to parse the natural language questions input by users based on the preset domain feature word library of ultra-high-speed scenarios, extract the entities and feature words of the natural language questions, and then classify the natural language questions into equipment queries, process queries or time queries; A query statement generation module is used to convert natural language questions into graph database query statements based on the classification results of natural language questions, perform data retrieval related to ultra-high-speed optical module manufacturing in the knowledge graph, and extract corresponding equipment information, process flow or time node information; The answer generation module is used to convert the query results into a natural language form that the user can understand, dynamically respond and return it to the user.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Images