A novel GPU based intrusion detection system using deep autoencoder with Fruitfly optimization

Abstract Intrusion Detection Systems (IDSs) have received more attention to safeguarding the vital information in a network system of an organization. Generally, hackers are easily entering into secured through loopholes and smart attacks. In such situation, predicting attacks from normal packets is tedious, much challenging, time consuming highly technical. As result, different algorithms with varying learning training capacity been explored literature. However, existing methods could not meet desired performance requirements. Hence, this work proposes new technique using Deep Autoencoder Fruitfly Optimization. Initially, missing values dataset imputed Fuzzy C-Means Rough Parameter (FCMRP) algorithm which handles imprecision datasets exploit fuzzy rough sets while preserving crucial information. Then, robust features extracted multiple hidden layers. Finally, obtained fed Back Propagation Neural Network (BPN) classify Furthermore, neurons layers optimized population based Optimization algorithm. Experiments conducted on NSL_KDD UNSW-NB15 dataset. The computational results proposed intrusion detection deep autoencoder BPN compared Naive Bayes, Support Vector Machine (SVM), Radial Basis Function (RBFN), BPN, Softmax. Article Highlights A hybridized model introduced Missing method. discriminate