Measuring the Performance of the Web-Search Engines by Repeated Query Approach

The modern information retrieval systems become the basic building block of the modern-day web search engines. The factors that affect the performance of the information retrieval system also affect the web search engines performance. Information retrieval systems’ performance depends on the index terms, and their occurrence frequency. Hence, the small variation in the index term’s frequency alters the information retrieval systems’ performance. This work tries to analyze the variation in the performance of information retrieval systems which is due to the changes in the index terms frequency. Based on the occurrence frequency, the index terms are classified as ’Low’, and ’High’ frequency terms; and their performances are also recorded. Low-frequency terms tend to decrease the information retrieval systems’ performance. In contrast, the performance of high-frequency terms is better than its counterpart. High-frequency terms do 10% performance improvement in comparison with the low-frequency terms. By deleting the low-frequency index terms, we can save up to 65% of index terms with the maximum of 26% degradation in IR system’s performance. This trade-off seems to be well paid. Finally, it is confirmed that the system which uses the whole index terms tender better performance than the system which uses either the low or the high-frequency terms.