Construction Of Uav Knowledge Base System Based On Microservice Architecture And Knowledge Graph

In recent years,with the rapid development of the UAV technology and the increasing market demand year by year,UAV has been widely used in various fields and plays an indispensable role.With the advent of the “Internet Plus” era and the rapid development of big data technology,the information data from the Internet has shown an exponential growth trend.And how to obtain the effective information needed in the era of information explosion has become an urgent problem to be solved.This paper adopted B/S server architecture,as well as based on the distributed microservice architecture,and based on the distributed and the incremental crawler technology to obtain the massive data related to the UAV industry and other application areas from the Internet;And adopting the algorithms such as natural language processing(NLP),knowledge graph(KG)and machine learning(ML)to extract and mine the implicit and potentially valuable knowledge information from the massive unstructured text data.As well as using the knowledge graph(KG)algorithms and Echarts and other visualization components to perform the visualization analysis,so as to form a customized UAV knowledge base.The main research contents of this paper are as follows:(1)Knowledge acquisition and storage technology.This system used Scrapy as the crawler framework,and using the distributed combined with the incremental crawler technology to efficiently obtain the massive data related to the UAV from the Internet.And using the Redis cache database to efficiently the crawled URLs,as well as combined the My SQL relational database steadily stores the structured data which was finally extracted.(2)System construction process and architecture design.Using the cloud computing and big data management platform,based on the distributed server and big data technology to build the system.And the knowledge base is constructed with a bottom-up approach combined with a top-down approach.Due to the increase in the amount of the system crawling tasks and the demand for the distributed servers,this system architecture has also been optimized and upgraded to the Spring Cloud distributed microservices architecture from the initial SMM monolithic architecture,so as to improve the scalability and portability of this system.(3)Data extraction and knowledge extraction technology.This system used the machine-like learning mechanism to semi-automatically extract the unstructured text data.By building the custom rule base to train the sample data,so as to extract the labels to match the extracted structured data;as well as using the natural language processin(NLP)algorithms to extract the key sentences to form the article’s summary information from the obtained text data.By using the knowledge graph(KG)and other algorithms to realize the knowledge extraction of the obtained text data,and finally extracting the structured knowledge elements to add to the knowledge base,so as to realize the knowledge discovery and and the knowledge mining functions of the knowledge base.(4)Knowledge representation and data visualization analysis.Adopting the knowledge tree form for achieving knowledge representation,and then using the visualization technologies such as Echarts and knowledge graphs to display and analyze the obtained and processed data in various visual forms such as the statistical charts,keyword cloud graphs and institutional co-occurrences and author co-occurrence diagrams,so as to visually and clearly display the research status and the development trends of UAV and other industries.