For improving the retrieval performance of large-scale component repositories, a novel approach of components retrieval, Automatic Tags Extraction(ATE) retrieval, is proposed in this paper. In this method, component tags are extracted automatically from application domain terms, high-frequency terms, high-weight terms and facet terms in description document of component at first, and then the improved VSM (Vector Space Mode) similarity algorithm is used to retrieve on the tags. Our experiments show that, compared with some common retrieval methods, the ATE retrieval is more feasible and efficient.
Discussion(0)
No comments yet. Be the first to comment.