Distributed Denial of Service (DDoS) attacks represent a significant threat to the security and availability of Internet of Things (IoT) networks, where vast numbers of interconnected devices with limited security controls are often exploited to launch large-scale, coordinated attacks. This research focuses on the detection and mitigation of DDoS attacks in IoT environments through the application of advanced machine learning techniques combined with real-time traffic monitoring and adaptive response strategies. Traditional detection mechanisms, including signature-based and anomaly-based systems, face challenges such as high false positive rates and inability to detect novel attack patterns effectively. To address these limitations, the proposed approach integrates supervised learning models for identifying known attack signatures alongside unsupervised anomaly detection algorithms that uncover previously unseen attack behaviors. Feature extraction and dimensionality reduction are utilized to handle the complexity and high dimensionality of IoT traffic data efficiently, enabling lightweight processing suitable for resource-constrained IoT devices and gateways. The mitigation framework incorporates dynamic traffic filtering, rate limiting, and automated blacklisting, supported by cloud and edge computing resourcees, to ensure scalable and responsive defense mechanisms. Experimental results on benchmark IoT datasets demonstrate that this hybrid detection and mitigation system achieves high accuracy in identifying diverse DDoS attack vectors, significantly reduces false alarms, and provides timely mitigation to protect IoT infrastructure from disruption. This comprehensive approach ensures the robustness, adaptability, and scalability of DDoS defenses, contributing to the enhanced security and reliability of IoT ecosystems in the face of evolving cyber threats.