Attacks on network infrastructure include attacks on the integrity and confidentiality of network packets, their destinations and origins, and attacks on network availability. A distributed denial of service (DDoS) attack poses a significant risk to service providers. A DDoS attack is designed to disrupt and limit legitimate users' access to services by flooding the target with enormous malicious requests. A cyber-attack of this magnitude would almost certainly result in massive economic losses for companies and service providers as operational and financial expenses rise. Machine learning (ML) approaches have been increasingly employed to counter DDoS attacks. Indeed, with ML approaches, many protection systems have been changed into smart and intelligent systems that can fight DDoS attacks. This paper presents a DDOS Attack Detection (DDAD) framework for detecting IoT DDoS attacks using ML techniques. The Label Encoding with standard scalar technique is used to preprocess the datasets. An improved firefly algorithm is used to extract the characteristics. A mixed ML Algorithm was used to determine the feature significance. Finally, classification was performed using ML algorithms such as RF, DT, SVM, and LR. The experimental findings are described in conjunction with several categorization metrics, including accuracy, precision, recall, and f-measure.