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

Table 3

The Comparison Results between MOFSRank and Feature Selection Baselines for Ranking on LETOR Data Sets, Averaged on Five Folds.

	N@1	N@2	N@3	N@4	N@5	N@6	N@7	N@8	N@9	N@10	MAP
NP2004
FenchelRank	0.5600	0.7400	0.7636	0.7728	0.7808	0.7962	0.801	0.8054	0.8096	0.8157	0.6830
FSMRank	0.5467	0.7800	0.7784	0.7943	0.8000	0.8071	0.8190	0.8279	0.8279	0.8279	0.6837
	0.5867	0.7533	0.7686	0.7711	0.7848	0.7899	0.7899	0.8055	0.8097	0.8137	0.6963
MOFSRank	0.5547	0.8000	0.8218	0.8411	0.8466	0.8505	0.8514	0.8518	0.8543	0.8543	0.7064
HP2004
FenchelRank	0.6667	0.7667	0.7961	0.8095	0.8181	0.8232	0.8232	0.8232	0.8274	0.8274	0.7447
FSMRank	0.6133	0.7933	0.8070	0.8187	0.8187	0.8255	0.8302	0.8383	0.8383	0.8383	0.7205
	0.6267	0.7867	0.8035	0.8135	0.8135	0.8135	0.8182	0.8182	0.8224	0.8265	0.7242
MOFSRank	0.6720	0.8080	0.8343	0.8406	0.8477	0.8508	0.8541	0.8585	0.8594	0.8622	0.7614
TD2004
FenchelRank	0.3600	0.3933	0.3725	0.3606	0.3462	0.3363	0.3329	0.3263	0.3220	0.3202	0.2368
FSMRank	0.3600	0.3400	0.3384	0.3260	0.3151	0.3080	0.3092	0.3128	0.3138	0.3133	0.2267
	0.3733	0.3933	0.3630	0.3462	0.3324	0.3279	0.3262	0.3261	0.3213	0.3205	0.2314
MOFSRank	0.4533	0.4467	0.4105	0.3901	0.3795	0.3711	0.3616	0.3579	0.3551	0.3560	0.2427
MQ2008
FenchelRank	0.3762	0.4164	0.4402	0.4598	0.4790	0.4933	0.4989	0.4619	0.2277	0.2317	0.4785
FSMRank	0.3750	0.4211	0.4404	0.4611	0.4809	0.4938	0.4986	0.4624	0.2283	0.2321	0.4760
	0.3720	0.4167	0.4390	0.4614	0.4805	0.4934	0.4995	0.4625	0.2268	0.2307	0.4800
MOFSRank	0.3957	0.4300	0.4521	0.4718	0.4888	0.5021	0.5064	0.4700	0.2360	0.2406	0.4867
OHSUMED
FenchelRank	0.5456	0.5390	0.5166	0.4989	0.4826	0.4742	0.4721	0.4680	0.4652	0.4637	0.4486
FSMRank	0.5397	0.5124	0.5070	0.4938	0.4808	0.4725	0.4692	0.4637	0.4567	0.4534	0.4455
	0.5489	0.5337	0.5032	0.4812	0.4765	0.4688	0.4631	0.4591	0.4594	0.4610	0.4477
MOFSRank	0.5707	0.5353	0.5144	0.4970	0.4872	0.4774	0.4719	0.4676	0.4646	0.4641	0.4489