Optimization of query processing over internet services

Abstract

Information at World Wide Web is retrieved through queries that are submitted by the users to the search engine For a given query various techniques are followed to access the WWW to process it and retrieve the most related documents All techniques retrieve more or less same documents but vary in time that it consumes to generate such results newlineTherefore the best method is needed to reduce the time required for information retrieval This requirement is focused by enormous researchers to discover and invent many techniques Though the invented techniques retrieved the documents quickly they did not concentrate on the relevancy of retrieved documents Hence an essential technique that depletes less time and retrieves more accurate document is necessary To address this issue a framework named Semantic and Feature Aggregated Information Retrieval newline is proposed in this dissertation The proposed framework comprises the following components namely 1 clustering 2 indexing 3 information retrieval and 4 ranking Each component is described in four different phases At the first phase the documents are indexed through a technique named multilevel hashing that helps to find the location of information easily his indexing technique properly maintains the web documents in two levels a database named BulidKeyDB is built as the first and keywords are indexed in the second level In second phase the documents are clustered which is newlineanother way for increasing the speed of the searching operation newline newline