TY  - JOUR
T1  - Adaptive Selection of Top-m Retrieval Strategies for Data Fusion in Information Retrieval
AU - , N.P. Gopalan AU - , K. Batri 
JO  - International Journal of Soft Computing
VL  - 2
IS  - 1
SP  - 11
EP  - 16
PY  - 2007
DA  - 2001/08/19
SN  - 1816-9503
DO  - ijscomp.2007.11.16
UR  - https://makhillpublications.co/view-article.php?doi=ijscomp.2007.11.16
KW  - Information retrieval
KW  -data fusion
KW  -adaptive selection
KW  -genetic algorithm
KW  -student-t test
AB  - Data fusion for Information Retrieval (IR) usually combines various strategies (schemes) to enhance the performance. The process of selecting the top `m` strategies with appropriate weights has been presented in this study. Genetic Algorithm (GA) with relevant judgments obtained for the user specified queries as training data is employed for the adaptive selection of the suitable strategies. The significance of experimental results obtained with three benchmark test collections is examined using Student-t test. The present adaptive method has been observed to perform consistently well in comparison with the individual ones participating in the fusion.
ER  -