Search engines play important roles in the success of the Web, search engines helps any Internet user to rapidly find relevant information. But the unsolved problems of current search engines have led to the development of the semantic Web. In the environment of semantic Web, the search engines should be more useful and efficient for searching the relevant Web information. In this paper we have presented the architecture of a semantic search engine, and our work shows how the fundamental elements of the semantic search engine can be used in the fundamental task of information retrieval. And then an improved algorithm based on TFIDF algorithm was proposed to guarantee the retrieve information resources in a more efficient way.