Mining for Web User Need
Manojeet Roy1, Ajay Kushwaha2
1Mr. Ajay Kushwaha, Reader CSE Dept..RCET ,Bhilai, M.C.A , MTech(CS),PhD (CSE) pursuing from CSVTU , Chhattisgarh, India.
2Mr. Manojeet Roy, MTech Computer Science Department , CSVTU University/ RCET Organization , City Bhilai, India.
Manuscript received on November 15, 2011. | Revised Manuscript received on December 01, 2011. | Manuscript published on December 30, 2011. | PP: 106-110 | Volume-1 Issue-2, December 2011. | Retrieval Number: B0157121211/2011©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Two fundamental issues regarding the effectiveness of information gathering from the Web: mismatch and overload. Mismatch means some useful and interesting data has been overlooked, whereas overload means some gathered data is not what users want. Classification and clustering has become an increasingly popular method of multivariate analysis over the past two decades, and with it has come a vast amount of published material. Since there is no journal devoted exclusively to cluster analysis as a general topic and since it has been used in many fields of study. Traditional techniques related to information retrieval (IR) have touched upon the fundamental issues [1], [2].However; IR-based systems neither explicitly describe how the systems can act like users nor discover exotic knowledge from very large data sets to answer what users really want. it is challenging to use semantic relations of “kind-of”, “part-of”, and “related-to” and synthesize commonsense and expert knowledge in a single computational model.
Keywords: Web mining, clustering, similarity search.