This article presents a unique ensemble SVM-based CGO algorithm to improve the classification accuracy of intrusion detection in a large data setting. Preprocessing is done on the dataset to remove redundant, noisy, and undesired data. The preprocessed data is then classified using ensemble SVM and the CGO method is used to increase the accuracy of the classification. The proposed solution uses a Hadoop infrastructure to quickly process the massive data instances. Several procedures are carried out before experiments, including data preparation, data exploration, and model construction. A dataset including a wide range of simulated incursions into a military network environment was made available for assessment.
