An efficient IDS in cloud environment using feature selection based on DM algorithm

Cloud Computing provides the use of a wide array of applications to a designated server outside one’s personal computer. In the current technological era with the evolution of the Internet, it is being used on a wider range. With such popularity and wide use comes a threat to its security. Intrusion Detection System (IDS) helps to secure the cloud environment from intruders by classifying the packets as an attack or normal. The datasets used for such purpose are very large which contains many features hence takes a huge time in computation. It is important to choose pertinent features to feed into the model which can give better results than using all the features and take less computational time. The authors proposed a nature-inspired Dolphin Mating (DM) algorithm to determine pertinent features from the dataset. For this purpose, the authors have used the NSL-KDD dataset and Kyoto dataset. The selected features are trained and tested using several machine learning algorithms. The result obtained is compared with several existing algorithms and it was found that the proposed DM algorithm selects the most relevant feature subset which made the IDS efficient in the Cloud environment.