Research Article

Multi-Label Feature Selection with Conditional Mutual Information

Table 5

Ranking Loss of results after applying multi-label feature selection algorithms on ten datasets.

AlgorithmsCRMILSCLSD2FFIMFPMUAMINMDGFSSLMFS-MCDM

Corel5k0.00010.00110.00820.00800.00790.00070.00080.00060.0010
Delicious00.00260.01380.01550.01270.00490.00210.00190.0027
Flags0.21560.24390.24770.35750.36820.35750.21760.26910.2483
Medical00.01670.01400.01690.01390.32120.01480.01320.0162
Scene0.14840.17810.18430.26250.18690.26030.17920.15610.1907
Enron0.08120.10120.12180.11640.13590.15270.11940.09240.1274
GenBase0.03270.06290.07260.09130.08970.07920.06130.05920.0623
Social0.13290.15830.16720.15970.15910.15470.15170.14280.1581
Yeast0.19230.21740.22780.22640.24010.22390.21040.20770.2065
Emotions0.23830.25360.27520.27930.28110.26740.25480.24790.2507
Average0.07640.09560.10370.128500.12180.16640.09340.09190.1008