Cybersecurity in the face of DDoS attacks is crucial for safeguarding digital assets and ensuring uninterrupted services. DDoS attacks typically utilize a large volume of packets generated from multiple compromised systems (zombie systems), causing significant disruption and inconvenience to targeted users. This study presents use of cyber forensics approach towards mitigating DDOS attack over the networks employing machine learning methods such as random forest, support vector machine, integrating the intrusion detection architecture and network framework by effectively analyzing the traffic dataset. In this paper two datasets CIC-DDoS2019 and DDoS attack in SDN dataset are being used and performance metrics such as accuracy, recall and precision are utilized to optimize detection accuracy and model performance, contributing to a robust detection framework. This work finds that Light GBM outperformed other algorithms due to its ability to handle high-dimensional data effectively and reduce overfitting through ensemble learning, with the accuracy rate of 98.88% with CICDDoS2019 data and 99.99% with SDN data. This comprehensive approach offers insights for defense mechanisms emphasizing the importance of proactive cybersecurity measures in combating evolving cyber threats.