Recognizing the crucial role of communication, data delivery, and access across various sectors, including government, business, and individual domains, it becomes imperative to pinpoint faults and vulnerabilities in cyber communication. To safeguard personal, governmental, and business data from the growing threats of advanced attacks on wireless communications, the implementation of Intrusion Detection System is essential. Information security serves as a robust defense for both host machines and networks. Analytical methods, particularly machine learning, are employed for the identification and prevention of diverse wireless network threats. In this paper, a novel approach is presented based on machine learning model, is integrated with different cross-validation techniques to find few performance matrix parameters in identifying and eliminating intrusions from a wireless network. This hybrid methodology employs various attributes, such as the Gini Index and Entropy of the decision tree (DT) model, different DT methodologies, including those using Gini Index, train-split method, and information entropy, along with their respective subdivisions like K-Fold validation and Stratified K-Fold validation, that are implemented and evaluated on a recent dataset containing DDoS network attack activities. The proposed algorithm mentioned in the paper establishes its performance by evaluating network performance parameters including packet delivery ratio and network throughput.