Estimating DNA Degradation Levels Using Machine Learning: An Innovative Approach in Forensic Science

Authors- Assistant Professor Dr. Pankaj Malik, Ustat Kaur Khanuja, Priyanshi Laddha, Poorva Jain, Mahima jain

Abstract-DNA evidence plays a crucial role in forensic science, yet its reliability can be compromised due to degradation caused by environmental factors and the passage of time. Traditional methods for assessing DNA degradation often involve labor-intensive and time-consuming techniques, which may not provide accurate results under varied conditions. This study explores the application of machine learning to predict DNA degradation levels from a dataset comprising samples subjected to different environmental stressors. Various algorithms, including Random Forest, Support Vector Machines, and Neural Networks, were evaluated for their effectiveness in estimating degradation levels based on input features such as temperature, humidity, and exposure duration. The results demonstrated that machine learning models can significantly enhance the accuracy of DNA degradation estimation compared to conventional methods. By employing metrics such as precision, recall, and mean squared error, our findings indicate that machine learning not only offers a reliable alternative for DNA analysis but also presents a scalable solution for forensic investigations. This research underscores the potential of integrating advanced computational techniques in forensic science to improve the assessment of critical evidence.

DOI: /10.61463/ijset.vol.12.issue5.283