Supervised Approach to Identify Autism Spectrum Neurological Disorder via Label Distribution Learning

Table 5

Comparative performance evaluation of CSLDSVR and LDL algorithms.

Evaluation metrics	Algorithm	NYU	UM	Leuven	UCLA	KKI
Chebyshev↓	AA-BP	0.223 7 ± 0.035 6	0.218 4 ± 0.045 8	0.248 0 ± 0.044 6	0.250 6 ± 0.053 5	0.254 7 ± 0.052 9
	AA-KNN	0.144 1 ± 0.011 6	0.154 0 ± 0.021 1	0.157 9 ± 0.026 5	0.142 6 ± 0.031 3	0.157 2 ± 0.029 5
	LDSVR	0.150 1 ± 0.024 3	0.140 0 ± 0.012 8	0.162 9 ± 0.034 4	0.169 4 ± 0.053 4	0.160 2 ± 0.057 0
	SA-IIS	0.147 8 ± 0.011 8	0.153 5 ± 0.023 7	0.174 8 ± 0.021 4	0.145 8 ± 0.032 9	0.162 7 ± 0.049 5
	PT-BAYES	0.381 8 ± 0.111 9	0.205 7 ± 0.009 5	0.206 9 ± 0.007 8	0.213 5 ± 0.009 9	0.215 4 ± 0.008 1
	PT-SVM	0.200 5 ± 0.041 2	0.188 5 ± 0.042 3	0.183 1 ± 0.040 1	0.195 8 ± 0.033 0	0.182 2 ± 0.058 9
	CSLDSVR	0.141 3 ± 0.016 2	0.135 2 ± 0.023 6	0.140 2 ± 0.024 4	0.138 6 ± 0.038 4	0.126 7 ± 0.034 9
Cosine↑	AA-BP	0.873 1 ± 0.034 4	0.881 8 ± 0.035 6	0.862 2 ± 0.049 8	0.839 9 ± 0.057 8	0.843 7 ± 0.058 6
	AA-KNN	0.935 4 ± 0.009 6	0.928 6 ± 0.017 3	0.927 4 ± 0.020 8	0.929 7 ± 0.022 4	0.913 0 ± 0.024 4
	LDSVR	0.937 7 ± 0.019 1	0.944 8 ± 0.013 3	0.932 5 ± 0.029 2	0.928 5 ± 0.052 0	0.932 6 ± 0.047 4
	SA-IIS	0.940 7 ± 0.009 3	0.934 4 ± 0.016 7	0.920 5 ± 0.016 0	0.939 5 ± 0.020 3	0.924 6 ± 0.042 5
	PT-BAYES	0.798 5 ± 0.071 3	0.915 6 ± 0.006 2	0.915 1 ± 0.005 3	0.910 4 ± 0.006 9	0.909 2 ± 0.005 7
	PT-SVM	0.898 7 ± 0.038 5	0.904 3 ± 0.042 8	0.914 5 ± 0.030 9	0.897 4 ± 0.036 5	0.906 8 ± 0.045 8
	CSLDSVR	0.940 5 ± 0.012 1	0.947 3 ± 0.018 3	0.923 4 ± 0.025 5	0.942 8 ± 0.036 8	0.936 3 ± 0.029 4
Clark↓	AA-BP	0.468 1 ± 0.064 8	0.461 3 ± 0.099 0	0.517 0 ± 0.083 8	0.537 1 ± 0.110 1	0.542 7 ± 0.104 6
	AA-KNN	0.263 1 ± 0.020 3	0.282 2 ± 0.036 7	0.287 3 ± 0.047 3	0.261 3 ± 0.053 5	0.283 2 ± 0.053 9
	LDSVR	0.272 9 ± 0.036 4	0.255 7 ± 0.021 8	0.287 2 ± 0.062 6	0.295 6 ± 0.092 0	0.281 9 ± 0.100 8
	SA-IIS	0.266 3 ± 0.019 1	0.278 8 ± 0.039 7	0.311 3 ± 0.033 6	0.262 3 ± 0.055 5	0.293 9 ± 0.088 0
	PT-BAYES	0.893 6 ± 0.359 8	0.352 0 ± 0.014 5	0.352 3 ± 0.012 7	0.363 6 ± 0.016 2	0.366 3 ± 0.013 3
	PT-SVM	0.358 0 ± 0.070 2	0.348 1 ± 0.075 8	0.325 3 ± 0.065 5	0.350 5 ± 0.056 1	0.328 7 ± 0.098 1
	CSLDSVR	0.261 6 ± 0.032 1	0.246 3 ± 0.037 6	0.253 9 ± 0.041 8	0.248 4 ± 0.062 6	0.233 4 ± 0.061 8
Intersection↑	AA-BP	0.776 3 ± 0.035 6	0.781 6 ± 0.045 8	0.752 0 ± 0.044 6	0.749 4 ± 0.053 5	0.745 3 ± 0.052 9
	AA-KNN	0.855 9 ± 0.011 6	0.846 0 ± 0.021 1	0.842 1 ± 0.026 5	0.857 4 ± 0.031 3	0.842 8 ± 0.029 5
	LDSVR	0.849 9 ± 0.024 3	0.860 0 ± 0.012 8	0.837 1 ± 0.034 4	0.830 6 ± 0.053 4	0.839 8 ± 0.057 0
	SA-IIS	0.852 2 ± 0.011 8	0.846 5 ± 0.023 7	0.825 2 ± 0.021 4	0.854 2 ± 0.032 9	0.837 3 ± 0.049 5
	PT-BAYES	0.618 2 ± 0.111 9	0.794 3 ± 0.009 5	0.793 1 ± 0.007 8	0.786 5 ± 0.009 9	0.784 6 ± 0.008 1
	PT-SVM	0.799 5 ± 0.041 2	0.811 5 ± 0.042 3	0.816 9 ± 0.040 1	0.804 2 ± 0.033 0	0.817 8 ± 0.058 9
	CSLDSVR	0.858 7 ± 0.041 5	0.864 8 ± 0.023 6	0.859 8 ± 0.024 4	0.861 4 ± 0.038 4	0.873 3 ± 0.034 9
KL↑	AA-BP	0.166 7 ± 0.042 9	0.161 2 ± 0.051 7	0.192 0 ± 0.069 3	0.222 2 ± 0.089 8	0.227 9 ± 0.076 4
	AA-KNN	0.068 5 ± 0.010 1	0.076 0 ± 0.018 4	0.076 6 ± 0.022 1	0.074 6 ± 0.023 2	0.093 2 ± 0.026 6
	LDSVR	0.066 5 ± 0.019 9	0.059 3 ± 0.014 6	0.070 3 ± 0.032 3	0.074 9 ± 0.062 5	0.071 1 ± 0.049 8
	SA-IIS	0.063 9 ± 0.009 3	0.069 8 ± 0.017 8	0.083 7 ± 0.016 6	0.063 9 ± 0.021 0	0.080 0 ± 0.044 1
	PT-BAYES	0.492 9 ± 0.251 0	0.087 9 ± 0.006 7	0.088 0 ± 0.006 1	0.093 5 ± 0.008 0	0.094 8 ± 0.006 6
	PT-SVM	0.108 1 ± 0.041 2	0.105 5 ± 0.047 6	0.090 6 ± 0.032 9	0.110 4 ± 0.040 5	0.100 3 ± 0.048 1
	CSLDSVR	0.060 3 ± 0.041 5	0.056 7 ± 0.019 5	0.069 9 ± 0.024 0	0.060 1 ± 0.046 1	0.068 2 ± 0.030 1