Knowledge Graph-based Knowledge Acquisition System In Open Space

Most of the information on the Internet is stored as web pages,linking documents on the web through hyperlinks.People can understand the information in the document,but the computer cannot fully understand.The traditional knowledge base mainly helps users understand intuitively,but cannot provide semantic information which can be understood by computers.The traditional knowledge base limits logical reasoning in the process of knowledge service and organization,and limits smart associations.Based on the knowledge graph,the complex semantic processing can annotate the entities information semantically according to the ontology model,generate knowledge network about the biology subject,and achieve the organization of knowledge semantics.Based on the knowledge graph,we developed this open space knowledge acquisition system,Provide external data support for the Q & A solution system.Under this premise,The main contents of this paper are as follows:(1)Introduction of technology involved.Technology include Resource Description Framework(RDF),graph database,MVC pattern,and web crawler.(2)Analysis of the system's requirements based on fully research.The requirements analysis mainly includes the user requirements,functional requirements,performance requirements,and security requirements.(3)Detailed design of the system based on the requirement analysis.This part mainly includes the overall system design,system-specific module design,system database and class design.The design of each module includes six submodules: words processing module,data crawling module,data processing module,data storage module,interface module and user management module.Database design includes database conceptual design and database physical design.(4)System Development Based on MVC mode.Select JQuery,Bootstrap and other framework to achieve page display.The basic function of the system including adding,deleting,inquiring and modifing achieved by BeetlSQL framework.JavaScript helped achieve page interaction.We selected Apache Shiro framework guarantees secure login and authority management of the system.We selected NEO4 J as the system's storage database,and web crawler as a system of collection tools.(5)Testing of developed systems based on software testing methods.System testing includes user interface testing and functional module testing.Test results show that the system interface is concise and clear,and the functions meet the requirements.