DNA profiling is a forensic technique used to identify individuals based on their unique genetic makeup. By analyzing specific regions of the DNA, this method provides a powerful tool for criminal investigations, paternity testing, and identifying remains, ensuring precise and reliable identification. Our work streamlines DNA analysis post-natural disasters by comparing Short Tandem Repeat regions. Our method combines DNA sequencing, machine learning classification, and gene matching on a diverse real-world dataset. To understand dataset features, we applied correlation matrices and feature selection, utilizing confusion matrices for training and testing. Integration of Convolutional Neural Networks, Naïve Bayes classifier, and other techniques enhances classification and analysis, determining the matching percentage of unknown DNA against a known dataset. Our framework expedites DNA analysis, simplifying disaster victim identification. Automation optimizes matching, saving time and resources. This work leverages advanced technologies to address complex DNA mixtures in disaster scenarios, marking a significant step forward in disaster victim identification. This abstract outline a robust DNA matching process, combining correlation analysis, feature selection, and confusion matrices for dataset insight. Leveraging Convolutional Neural Networks, Naïve Bayes, and various ML methods, it efficiently determines matches between unknown and known DNA. Utilizing the Smith–Waterman algorithm streamlines identification, providing a time-saving solution for DNA matching and profiling for the victims.