Web page indexing through page ranking for effective semantic search

The Semantic Web realization is based on the availability of a significant mass of metadata for the web content, associated with the appropriate information about the world. Searching the vast internet for small particular information leads to many anomalies in the results of present searching techniques such as imprecision, large search result, unable to interpret the sense of user's query and so on. With the purpose of overcoming the pitfalls of the existing approaches, we have proposed architecture of semantic information retrieval to enhance the relevancy of search results. An algorithm is proposed to compute the rank of the documents and then semantic indexing is being performed by these ranked web pages. In our algorithm, we are considering the two factors for calculating the page rank; one is the frequency of the keyword occurred in the web page and another is associative factor of the same keyword with the meaningful interrogative words which are generally ignored by the existing search schemes. We have tested our algorithm for various documents and hence the proposed approach provides the most relevant results on the top of the result set.