The diverse nature and increasing reliance on virtual systems, cloud computing offers countless services to clients. However, this reliance also makes these services vulnerable to Distributed Denial of Service (DDoS) attacks. Detection of such attacks and mitigating their effects is crucial. Recent studies have found that ensemble techniques, like boosting and bagging, are used to detect DDoS traffic but lack generalization in predicting such attacks. Therefore, there is a need to generalize the prediction to detect DDoS attacks in cloud environments. Thus, this work proposes a Stacked-Ensemble (SE) model to identify DDoS attacks in cloud networks. The proposed SE model has a two-layered architecture with base models stacked at level-0 and a final model stacked at level-1. The predictions of the base models at level-0 are combined to train the final model at level-1, ensuring the generalization of predictions for the attacks. The proposed model is tested using the CICIDS-2017 dataset and outperforms existing techniques by achieving an accuracy of 99.9%. in detecting DDoS in cloud environments and outperforming state-of-the-art techniques.