With the increasing prevalence of network attacks specifically low-rate and high-rate Distributed Denial of Service (DDoS) attacks, there coins an emerging and immediate need to arrive with computationally efficient but simpler methods and those that can yield better classifier evaluation parameters. Although existing deep learning techniques have shown promising results in identifying such attacks, but the lack of interpretability and explainability in such models is the major limitation and this consequently hinders their widespread adoption in security critical domains by industry. Also, the current need for industry is to have explainable model at hand to make better decision making. This leads to the design of explainable AI systems as addressed by MIT, USA in recent times. Researchers across the globe and the IT industry are now looking to come out with explainable AI models which form the basic motivation for the current work. The objective of this research is to propose an ensemble machine learning system to detect both low rate and high-rate network attacks with better detection rates when compared to state-of-the-art learning algorithms and studies that are employed for network intrusion detection.