A wireless attack is a malicious action against wireless systems and wireless networks. In the last decade of years, wireless attacks are increasing day by day and it is now a very big problem for modern wireless communication systems. In this paper, the author’s used the Aegean Wi-Fi Intrusion Dataset (AWID3). There are two versions of this dataset, one is over 200 million tuples of full data and one is 1.8 million tuples of reduced data. As our dataset has millions of tuples and over 100 columns, it is easy to become overwhelmed because of its size. The authors used the reduced version and predicted if an attack was one of four types using the k-nearest-neighbors classifier. All of the attack types we used had a distribution that highly favored the non-attack class. Our best results were for the attack “arp” type where we attained the best accuracy with recall. The author’s primary goal of the paper was to attain the highest accuracy possible when creating a model that is capable of classifying the 4 attack types and detecting and classifying wireless attacks using Machine Learning models on the AWID3 dataset. One of the goals that supported this main objective was determining a way to avoid the curse of dimensionality.