Research On Component Retrieval Technologies Based On Faceted Classification Schema

Software reuse is one of the practical and feasible approaches that resolve software crisis at present. The development of component and component library is the premise of software reuse. In practice, the number of components increases with years, the components also are updated with the development of technologies. Therefore, how to retrieve a mass of components effectively in the component library becomes the emphasis of the research on the component library and the software reuse.The subject represented in this thesis stems from National 863 Project Subject "Research on software support system based on high performance wide-band communication network"(No.2004AA116010).The thesis mainly researches on component retrieval technologies based on the faceted classification schema. By studying related background, current research work and theories, problems of the current component retrieval method based on faceted classification schema are pointed out. A new method of component retrieval method based on faceted classification schema is put forward which combines the features of faceted classification schema and the tree matching model. This method can give dual attention to the recall and precision of component retrieval and has higher retrieval efficiency. The design and realization of a component retrieval tool based on the new retrieving method is also represented in this thesis. And it is used in the Dalian component library. The efficiency and feasibility is proved by the experiments. At the end of this thesis, our main work is summarized and our future work is discussed.The goal of our subject is to present a more efficient method for component retrieval by the research on the key technologies of the component retrieval. Our work will be beneficial to improve the capability of the component library, and to reduce the cost of software reuse. The application of our research work can help the enterprises doing software reuse with high efficiency, high quality and low costs, as well as improving the quality of software products and the enterprise competitive capability.