Egyptian Journal of Forensic Sciences

Table 7 Multiple linear regression equations and ANN models with respect to males, females and results combined

From: Sex and stature estimation from anthropometric measurements of the foot: linear analyses and neural network approach on a Turkish sample

Regression Results^*
RFH + RFL + RFB						LFH + LFL + LFB
Linear Regression	R	R²	Adj. R²	MAE (cm)	RMSE (cm)	Linear Regression	R	R²	Adj. R²	MAE (cm)	RMSE (cm)
Male	0.670	0.449	0.424	3.31	4.27	Male	0.674	0.454	0.429	3.32	4.25
Female	0.760	0.577	0.556	3.45	4.22	Female	0.782	0.612	0.593	3.38	4.05
Overall	0.893	0.798	0.793	3.38	4.25	Overall	0.898	0.807	0.802	3.36	4.15
Neural Network	R	R²	Adj. R²	MAE (cm)	RMSE (cm)	Neural Network	R	R²	Adj. R²	MAE (cm)	RMSE (cm)
Male	0.686	0.471	0.447	3.22	4.18	Male	0.689	0.475	0.451	3.21	4.16
Female	0.774	0.600	0.580	3.36	4.11	Female	0.793	0.628	0.610	3.22	3.96
Overall	0.898	0.807	0.803	3.29	4.15	Overall	0.902	0.814	0.810	3.21	4.07

R Pearson’s correlation coefficient, R² Coefficient of determination, MAE Mean absolute error, RMSE Root mean squared error, RFH Right foot stature, RFL Right foot length, RFB Right foot breadth, LFH Left foot stature, LFL Left foot length, LFB Left foot breadth
^*All analyses were performed with tenfold cross-validation

Back to article page