A Global Fusion Re-Ranking for Effective Data Search with Correlation Model

A. Sudha

Abstract: Web service plays an important role in e-business and e-commerce applications. The web service applications are interoperable and it will work on any platform; large scale distributed systems can be developed easily. Surfing on the internet is becoming more prominent in day today life. In searching there are major issues like noisy data and unwanted data. The existing system provides additional results rather than appropriate results and uses PageRank algorithm. PageRank uses link analysis algorithm to measure the page relevance in a hyperlinked set of documents. In order to improve the existing co?diffusion of keywords and ranking; the system introduces the result remerging and re-ranking concepts. Basically ranking will be performed by the popularity; key term and its frequency count. In the proposed system an enhanced ranking concept is used which improves the performance of the co-diffusion ranking system. The ranking is done locally for two or more search engines and a global re-ranking are done at the end considering the page structure which includes pre-link; post link and the popularity as well. All these are performed by using the newly proposed Semantic Dual Correlation Algorithm which makes searching effective.

Keywords: eb mining; Expert search; performance; data mining; Expert Search Information Retrieval