Machine Learning-Based Facial Beauty Prediction and Analysis of Frontal Facial Images Using Facial Landmarks and Traditional Image Descriptors

Table 9

Performance of each model according to correlation for 11–25 features.

S. No.	Number of features	LR	KNN	RF	ANN
1.	11	0.5492	0.625	0.6201	0.5527
2.	12	0.4842	0.5603	0.5535	0.4972
3.	13	0.5076	0.6694	0.3729	0.509
4.	14	0.4859	0.7115	0.4211	0.486
5.	15	0.5019	0.6587	0.542	0.1443
6.	16	0.5792	0.6458	0.3761	0.3275
7.	17	0.5952	0.6067	0.5158	0.5837
8.	18	0.5921	0.5773	0.5424	0.3648
9.	19	0.5811	0.5449	0.4248	0.5838
10.	20	0.5954	0.5321	0.4156	0.5801
11.	21	0.5806	0.5495	0.4187	0.5796
12.	22	0.5854	0.5369	0.4258	0.58
13.	23	0.5897	0.5485	0.4296	0.5732
14.	24	0.5801	0.5331	0.4235	0.5721
15.	25	0.58	0.541	0.4113	0.5821