Performance Evaluation of Image Search Engines Based on the Number of Query Words

Abstract

In this article, the impact of the number of query words on information retrieval effectiveness of image search engines was investigated. First of all, three popular search engines, namely, Google, Yahoo and MSN, were selected. Then, 40 queries were extracted from the list of Wordtracker and categorized in four groups as one-, two-, three- and four-word queries. After every query was run on the selected image search engines and binary human relevance judgments were done, the performance evaluation of image search engines was carried out on two cut-off points (first 10 and first 20 retrievals) in terms of precision and normalized recall. In general, the information retrieval effectiveness of image search engines decreases as the number of query words increases. Retrieval performance of image search engines needs to be improved.