The Study Of The Construction Method Of Software Engineering Knowledge Base Based On Ontology

Software engineering is the core disciplines in the field of information technology (IT), and plays a pivotal role in the development of IT. With the rapidly development of internet technology, software engineering knowledge is continuously updated with a high speed of growth. Due to lack of semantic search capability, traditional software engineering knowledge services cannot meet people’s growing demand of knowledge. Thus, the software engineering knowledge base on semantic build service capacity has become a new research hotspot. Ontology is a formal definition of conceptual model, which is widely used in knowledge engineering, information retrieval and other fields. Because of the superiority in the concept of semantic structure and the ability of stimulating human thinking, it has become a new research direction in field of Knowledge Base Construction Method.Given that the shortcomings of current software engineering domain knowledge base, the present study proposes to build the knowledge base with semantic characteristics using ontology technology. We are aiming to design a knowledge base model with a universal value based on domain knowledge of ontology. This model consists of three parts:the planning and design of knowledge base, the construction of ontology, and knowledge services. First, the planning and design of knowledge base is composed of two parts, the requirements analysis and the design of knowledge architecture; The construction of ontology contains four parts, including knowledge construction, knowledge attribute extraction, the building of knowledge relationship, and the representation of knowledge using ontology technology; Knowledge service is composed by the representation of knowledge, semantic knowledge retrieval and evaluation.The current study presents a new method of ontology construction based on Wikipedia. This new method is very different from the traditional ontology construction method of domain ontology construction. It can extract the Wikipedia knowledge system and knowledge attribute, and realizing the automation of the original domain ontology building. Further, a new path-independent vector variance method was also presented in the current study. It can select the category and entry with a higher belonging degree than the threshold via calculating the belonging degree of each category and entry, and then obtain the entity attribute by excavating the content of each information box.Furthermore, the current study raises a new method of semantic retrieval, which can integrate the ontology reasoning technology and full-text retrieval method. Based on Jena ontology technology and Lucene retrieval technology, this method can propose the inference rules and query strategies of software engineering domain ontology, according to software engineering knowledge, and set up the knowledge retrieval framework based on ontology.Finally, the current study describes the implementation process of ontology knowledge base-based software engineering book content retrieval system, and exhibits the results of ontology-based semantic retrieval.