Table 5 Prediction accuracy with converted and not converted input features.

Dataset	SVM-RBF^a		SVM-linear^b		SVM-poly^c		SVM-sig^d		RF		ANNs		DT
Dataset	Con.	No con.	Con.	No con.	Con.	No con.	Con.	No con.	Con.	No con.	Con.	No con.	Con.	No con.
1	0.985	0.985	0.990	0.605	1.00	0.56	0.990	0.540	1.00	0.865	1.00	0.975	0.995	0.895
2	0.970	0.905	0.975	0.600	0.985	0.640	0.995	0.455	0.960	0.795	0.990	0.930	0.965	0.785
3	0.985	0.860	0.975	0.465	0.980	0.575	0.975	0.500	0.860	0.780	0.995	0.910	0.900	0.705
4	0.960	0.810	0.925	0.515	0.985	0.400	0.980	0.420	0.850	0.655	0.985	0.825	0.865	0.695
5	0.970	0.790	0.910	0.535	0.965	0.550	0.980	0.460	0.810	0.615	0.995	0.780	0.840	0.600
6	0.945	0.815	0.860	0.500	0.985	0.475	0980	0.485	0.770	0.620	0.990	0.770	0.795	0.615
7	0.940	0.715	0.905	0.530	0.980	0.500	0.980	0.535	0.865	0.610	0.985	0.670	0.795	0.585
8	0.970	0.675	0.955	0.410	0.970	0.455	0.955	0.455	0.760	0.545	0.995	0.695	0.760	0.610
9	0.955	0.660	0.885	0.515	0.960	0.460	0.955	0.435	0.790	0.510	0.990	0.665	0.770	0.580
10	0.955	0.655	0.860	0.480	0.955	0.525	0.975	0.525	0.735	0.520	0.960	0.600	0.750	0.625

Here, a: SVM with radial basis function (RBF) kernel; b: SVM with linear kernel; c: SVM with polynomial kernel; d: SVM with sigmoid kernel. RF: Random Forest; ANNs: artificial neuron network; DT: Decision Tree; Con: the converted input features; No con: the not converted input features.

Quick links

Search