MOFSRank: A Multiobjective Evolutionary Algorithm for Feature Selection in Learning to Rank

Table 4

The number of selected features between MOFSRank and feature selection baselines for ranking on LETOR data sets, averaged on five-folds.

	NP2004	HP2004	TD2004	MQ2008	OHSUMED
FenchelRank	18.60	12.00	32.40	15.80	13.00
FSMRANK	32.00	13.80	28.20	13.20	18.00
	14.60	7.00	17.20	2.40	9.60
MOFSRank	5.72	5.12	5.88	4.80	5.40