In this paper, we proposed a Query Expansion (QE) approach on a Turkish Text collection based on word-context matrix with a sliding fixed sized window and Singular Value Decomposition (SVD) method. Our query expansion approach uses semantic relationship of terms to improve the existing query expansion methods available in the literature, namely Bo1 (Bose-Einstein 1), Bo2 (Bose-Einstein 2) and KL (Kullback-Leibler). We evaluated our approach on Milliyet collection, which is a Turkish IR test bed containing more than 400K documents and 72 queries. The experimentation shows that our approach clearly improves the all three QE methods in terms of major Information Retrieval (IR) performance measures such as MAP, R-precision and P@10.
Author and article information
Department of Computer Engineering
Dokuz Eylul University, Izmir, TURKEY
Faculty of Technology
Celal Bayar University, Manisa, TURKEY