The surge in connected devices has led to an increase in distributed denial-of-service (DDoS) attacks, posing a threat to proper functionality. Machine learning (ML) techniques have shown promise in detecting and mitigating DDoS attacks by analyzing traffic patterns for anomalies. This paper presents an ML-based approach for detecting DDoS attacks, utilizing feature extraction to isolate relevant characteristics from network traffic data. The algorithm incorporates random forest, decision tree and k-nearest neighbors to distinguish between normal and anomalous traffic. Our proposed method stands as a reliable and effective means to identify DDoS attacks, enhancing overall security and preventing misuse for malicious activities. By showcasing the potential of machine learning, this approach addresses the pressing issue of cyber-attacks. In this paper, we suggest for more accurate and efficient mitigation that guarantees continuous network services, a proactive machine learning solution is essential.