Information Retrieval and the semantic web

Information Retrieval towards the semantic web has been one of the motivations of semantic web since it was introduce by Berners-Lee. In this paper, we proposed a semantic web search model to enhance efficiency and accuracy of information retrieval for unstructured and semi-structured documents. In order to increase system´s scalability, we employ RDF knowledge based to store metadata in our systems. In addition, we introduce a Ranking Evaluator to measure the similarity between documents with semantic information for rapid and correct information retrieval. More importantly, the system gives precise answers to precise question with the introduction of Ranking Evaluator. Compared to previous works, another important idea we proposed is that we use a Search Arbiter to judge whether the query is answered by Keyword-based Search Engine or Ontology Search Engine, which is based on whether there is not enough ontology knowledge or not. Also, key techniques are discussed in our paper. In a word, we believe our system can well be applied to semantic web for information retrieval.