Classification of gender using neural network and Support Vector Machine from bone morphology properties of monkey

Forensic anthropologists are often asked to evaluate partial skeletal remains or severely damaged. The purpose of this research is comparing the performance of Artificial Neural Network (ANN) and Support Vector Machine (SVM) for gender classification. The performance of both models in classifying the gender data is compared and validated using the Ryan and Shaw Dataset (RSD) in terms of accuracy, sensitivity and specificity percentage. These measurements were taken on 226 femurs (126 females and 100 males) and 216 humerus (126 females and 90 males) of old world monkey primates. The comparison result shows that the ANN classifier outperforms SVM classifier for correctly classifying the bone morphology properties in gender data in term of accuracy and specificity percentage value; 71% and 56% respectively. The conclusion of Artificial Neural Network (ANN) is a powerful classification model that improves the accuracy rate of gender classification models for skeletal remains.