Relevance Feedback Based Query Expansion Model Using Borda Count and Semantic Similarity Approach

Table 3

The comparison of different QE term selection methods with Okapi-BM25 in terms of average precision and recall for TREC dataset.

Methods	Top 10 retrieved documents		Top 25 retrieved documents		Top 50 retrieved documents
	Average precision	Average recall	Average precision	Average recall	Average precision	Average recall
Okapi-BM25	0.2378	0.1172	0.2204	0.1911	0.1955	0.3012
CBQE	0.2381	0.1195	0.2216	0.1978	0.1990	0.3160
RSVBQE	0.2435	0.1193	0.2259	0.2096	0.2098	0.3159
IGBQE	0.2510	0.1236	0.2481	0.2147	0.2319	0.3271
KLDBQE	0.2536	0.1304	0.2594	0.2191	0.2413	0.3315