A MACHINE LEARNING MODEL FOR DETERMINATION OF GENDER UTILIZING HYBRID CLASSIFIERS

Abstract

This study addresses the challenge of identifying gender in forensic anthropology when faced with incomplete, burned, or damaged skeletal remains. The objective is to utilize the pelvis and femur, established as reliable gender indicators, for accurate identification. Employing measurements of the subpubic angle of the pelvis and femur angles, a principal component analysis (PCA) method is applied to create two attributes for machine learning model input. The study incorporates a hybrid machine learning system, combining an Artificial Neural Network (ANN) and Support Vector Machine (SVM) design. Testing with acquired data yields an 83.33% accuracy, classified as "good" based on the Area Under the Curve (AUC) from the confusion matrix. The implications of this research extend to both theoretical and practical domains, providing a method for accurate gender identification in challenging forensic scenarios. The contribution lies in offering a reliable and innovative approach that can be applied to enhance gender determination practices in forensic anthropology, thereby advancing both theory and real-world applications.).