Retrieval and Evaluation of Target Component Based on Ontology Knowledge

Software reuse most focuses on component based software development (CBSD). However, it's not so accurate and efficient in the process, to solve this problem, this paper proposes an intelligent knowledge-driven method of target component retrieval and evaluation. This method is based on Ontology component description. With the help of the knowledge graph, a semantic mapping is formed between the component to be queried and the component description library, so the entity component is located. In order to measure the component matching performance, it gives multi-angle evaluation of queried candidate components by an evaluation index system. Based on component query information, it makes the searching target clearer, extends the semantic scope of components to be queried. In order to assembly components, a multi agent system (MAS) is also established. The result shows that this method not only makes the component retrieval process higher recall and precision, but also makes the component retrieval process more intelligent by meeting the assembly requirement.