Table 10 The average recall, precision and \(F_{1}\)-score of CSTH data depending on the noise level (standard deviation in parentheses)

From: Norm ball classifier for one-class classification

Classifier Training Validation Test
Recall Precision \({F}_{1}\)-score Recall Precision \({F}_{1}\)-score Recall Precision \({F}_{1}\)-score
0%
Two norm 0.984 (0.01) 1.000 (0.00) 0.992 (0.00) 0.910 (0.16) 1.000 (0.00) 0.944 (0.10) 0.980 (0.06) 1.000 (0.00) 0.989 (0.03)
Max norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.910 (0.12) 1.000 (0.00) 0.948 (0.07)
One norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.13) 0.990 (0.03) 1.000 (0.00) 0.995 (0.02)
Mix norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.910 (0.12) 1.000 (0.00) 0.948 (0.07)
MoG 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
PDE 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
1-NN 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.920 (0.16) 1.000 (0.00) 0.950 (0.10) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
k-NN 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
SVDD 0.992 (0.02) 1.000 (0.00) 0.996 (0.01) 0.870 (0.30) 0.900 (0.30) 0.884 (0.30) 0.880 (0.21) 0.938 (0.19) 0.906 (0.20)
5%
Two norm 0.985 (0.00) 1.000 (0.00) 0.993 (0.00) 0.860 (0.26) 1.000 (0.00) 0.895 (0.21) 0.900 (0.19) 1.000 (0.00) 0.934 (0.13)
Max norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.900 (0.13) 1.000 (0.00) 0.942 (0.07)
One norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.13) 0.930 (0.18) 1.000 (0.00) 0.952 (0.13)
Mix norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.900 (0.13) 1.000 (0.00) 0.942 (0.07)
MoG 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
PDE 1.000 (0.00) 0.944 (0.01) 0.971 (0.01) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
1-NN 1.000 (0.00) 0.944 (0.01) 0.971 (0.01) 0.890 (0.18) 0.885 (0.15) 0.869 (0.13) 0.990 (0.03) 0.874 (0.16) 0.922 (0.11)
k-NN 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
SVDD 0.948 (0.01) 0.956 (0.01) 0.952 (0.01) 0.950 (0.09) 0.973 (0.04) 0.958 (0.05) 0.930 (0.10) 0.972 (0.04) 0.946 (0.06)
10%
Two norm 0.985 (0.00) 1.000 (0.00) 0.993 (0.00) 0.860 (0.26) 1.000 (0.00) 0.895 (0.21) 0.920 (0.12) 1.000 (0.00) 0.954 (0.07)
Max norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.930 (0.11) 1.000 (0.00) 0.960 (0.06)
One norm 1.000 (0.00) 0.999 (0.00) 0.999 (0.00) 0.910 (0.16) 0.971 (0.09) 0.928 (0.11) 0.990 (0.03) 0.967 (0.10) 0.975 (0.06)
Mix norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.19) 1.000 (0.00) 0.923 (0.13) 0.860 (0.22) 1.000 (0.00) 0.906 (0.15)
MoG 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
PDE 1.000 (0.00) 0.904 (0.01) 0.950 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
1-NN 1.000 (0.00) 0.901 (0.00) 0.948 (0.00) 0.920 (0.16) 0.822 (0.15) 0.851 (0.13) 1.000 (0.00) 0.833 (0.15) 0.900 (0.11)
k-NN 1.000 (0.00) 0.999 (0.00) 0.999 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.991 (0.03) 0.995 (0.01)
SVDD 0.962 (0.02) 0.918 (0.01) 0.939 (0.01) 0.940 (0.09) 0.884 (0.12) 0.905 (0.10) 0.940 (0.10) 0.879 (0.13) 0.899 (0.09)
15%
Two norm 0.985 (0.00) 1.000 (0.00) 0.993 (0.00) 0.860 (0.26) 1.000 (0.00) 0.895 (0.21) 0.920 (0.12) 1.000 (0.00) 0.954 (0.07)
Max norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.930 (0.11) 1.000 (0.00) 0.960 (0.06)
One norm 1.000 (0.00) 0.999 (0.00) 0.999 (0.00) 0.910 (0.16) 1.000 (0.00) 0.944 (0.10) 0.990 (0.03) 0.967 (0.10) 0.975 (0.06)
Mix norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.880 (0.20) 1.000 (0.00) 0.922 (0.14) 0.930 (0.11) 1.000 (0.00) 0.960 (0.06)
MoG 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
PDE 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
1-NN 1.000 (0.00) 0.848 (0.01) 0.917 (0.00) 0.920 (0.16) 0.828 (0.04) 0.865 (0.09) 1.000 (0.00) 0.842 (0.04) 0.914 (0.02)
k-NN 0.995 (0.01) 0.995 (0.01) 0.995 (0.01) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.980 (0.04) 1.000 (0.00) 0.989 (0.02)
SVDD 0.984 (0.02) 0.854 (0.01) 0.914 (0.01) 0.940 (0.09) 0.799 (0.12) 0.857 (0.10) 0.960 (0.07) 0.823 (0.12) 0.880 (0.08)
20%
Two norm 0.985 (0.01) 0.989 (0.03) 0.987 (0.02) 0.840 (0.27) 1.000 (0.00) 0.882 (0.21) 0.920 (0.12) 1.000 (0.00) 0.954 (0.07)
Max norm 1.000 (0.00) 0.925 (0.06) 0.960 (0.03) 0.880 (0.20) 0.991 (0.03) 0.917 (0.13) 0.930 (0.11) 0.971 (0.06) 0.946 (0.07)
One norm 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 0.860 (0.20) 1.000 (0.00) 0.911 (0.13) 0.990 (0.03) 1.000 (0.00) 0.995 (0.02)
Mix norm 1.000 (0.00) 0.925 (0.06) 0.960 (0.03) 0.880 (0.20) 0.991 (0.03) 0.917 (0.13) 0.930 (0.11) 0.971 (0.06) 0.946 (0.07)
MoG 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
PDE 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00) 1.000 (0.00)
1-NN 1.000 (0.00) 0.800 (0.01) 0.889 (0.00) 0.920 (0.16) 0.768 (0.07) 0.831 (0.10) 1.000 (0.00) 0.814 (0.03) 0.897 (0.02)
k-NN 1.000 (0.00) 0.926 (0.06) 0.960 (0.03) 1.000 (0.00) 0.991 (0.03) 0.995 (0.01) 0.970 (0.06) 0.954 (0.07) 0.961 (0.06)
SVDD 0.959 (0.01) 0.815 (0.01) 0.881 (0.01) 0.940 (0.09) 0.820 (0.11) 0.870 (0.08) 0.940 (0.10) 0.812 (0.11) 0.865 (0.09)
  1. The bolded value is the best within the group