Table 9. Performance comparison of different methods on Z498 dataset.

Method	Prediction accuracy (%)
	All-α	All-β	α/β	α+β	Overall
Rough sets by Cao et al.(2006) [20]	87.9	91.3	97.1	86.0	90.8
SVM fusion by Chen et al. (2006) [39]	99.1	96.0	80.9	91.5	91.4
Markov-SVM by Qin et al. (2012) [27]	91.6	94.4	96.3	91.5	93.6
NN-CDM by Liu et al. (2010) [40]	96.3	93.7	95.6	89.9	93.8
Information-theoretical approach by Zheng et al. (2010) [22]	95.3	93.7	97.8	88.3	93.8
IGA-SVM by Li et al. (2008) [24]	96.3	93.6	97.8	89.2	94.2
LogitBoost by Feng et al. (2005) [21]	92.6	96.0	97.1	93.0	94.8
CWT-PCA-SVM by Li et al. (2009) [25]	94.4	96.8	97.0	92.3	95.2
AAC-PSSM-AC by Liu et al. (2012) [23]	94.4	96.8	97.8	93.8	95.8
Our method	100.00	99.15	100.00	100.00	99.79