Smart Intrusion Detection System Comprised of Machine Learning and Deep Learning

In the present world, digital intruders can exploit the vulnerabilities of a network and are capable to collapse even a country. Attack in Estonia by digital intruders, attack in Iran's nuclear plant and intrusion of spyware in smart phone depicts the efficiency of attackers. Furthermore, centralized firewall system is not enough for ensuring a secured network. Hence, in the age of big data, where availability of data is huge and computation capability of PC is also high, there machine learning and network security have become two inseparable issues. In this thesis, KDD Cup’99 intrusion detection dataset is used. Total 3, 11,030 numbers of records with 41 features are available in the dataset. For finding the anomalies of the network four machine learning methods are used like Classification and Regression Tree (CART), Random Forest, Naive Bayes and Multi-Layer Perception. Initially all 41 features are used to find out the accuracy. Among all the methods, Random Forest provides 98.547% accuracy in intrusion detection which is maximum, and CART shows maximum accuracy (99.086%) to find normal flow of data. Gradually selective 15 features were taken to test the accuracy and it was found that Random Forest is still efficient (accuracy 98.266%) in detecting the fault of the network. In both cases MLP found to be a stable method where accuracy regarding benign data and intrusion are always close to 95% (93.387%, 94.312% and 95.0075, 93.652% respectively). Finally, an IDS model is proposed where Random Forest of ML method and MLP of DL method is incorporated, to handle the intrusion in a most efficient manner.