Paper Key : IRJ************968
Author: Duvvala Suchideeksha
Date Published: 03 Apr 2024
Abstract
Identical twins, sharing nearly identical genetic makeup, present a unique challenge for traditional identification methods. This study explores the potential of machine learning (ML) algorithms to predict the presence of identical twins based on various biometric and genetic data. The research leverages a diverse dataset comprising genetic information, facial features, voice patterns, and other physiological attributes collected from a wide range of individuals, including identical twins and nontwin controls. We employ a multifaceted ML approach, including deep learning techniques, to analyse and extract distinctive patterns from the dataset. Feature selection and extraction methods are utilized to identify the most discriminative attributes for twin prediction. In real words twins faces are exists and this twin s can utilize advantages to dupe people in examination or any other organizations. To detect such twins we are applying machine learning algorithm such as Nave Bayes and Random forest which may get trained on possible Real and Twins faces. Additionally, this research investigates the transferability of models across different populations and explores the ethical considerations surrounding privacy and consent when dealing with sensitive genetic and biometric data.