Performance comparison of internet search engines

Abstract

This dissertation reports an investigation into the performance and searching technique of seven internet search engines. It provides a general overview of the search engines as well as a quick reference for the searching techniques and search tools. The main focus of this dissertation is in its evaluation of seven internet search engines using two evaluations methodologies. The first evaluation is using Signal Detection Analysis to test the sensitivity (d') and decision bias (beta). The sensitivity of the search engine is important for finding the useful information in its index. The larger the value of sensitivity, the better the finding of useful information. The decision bias is to measure how conservative or risky the search engine is in reporting useful links. The second evaluation is to measure the search engines accuracy in retrieving information and the indexing mechanisms of each search engine using Title, Meta-tag and keywords. The evaluation is carried out for 20 queries for the testing and evaluation of each search engine