Aggregating Content and Connectivity based Techniques for Measure of Web Search Quality

Abstract

Web Searching is very much popular among the Internet users. A large number of public search engines are available for this purpose. So, there must be some procedure to measure the quality of web search. In this paper, classical content-based retrieval techniques such as Vector Space Model and Boolean Similarity Measures are combined with connectivity-based techniques such as PageRank for measure of web search quality. These techniques are combined using Modified Shimura technique of Rank aggregation. The aggregated ranking obtained in the process is compared with the original ranking given by the search engine. The correlation coefficient thus obtained is averaged for a set of queries. We show our experimental results pertaining to seven public search engines and fifteen queries.