LEVERAGING EFFICIENT ALGORITHMIC TECHNIQUES IN EFFICIENT FEATURE SELECTION IN IOT BASED IDS

Prachi Juneja

Vol. 6, Jan-Dec 2020

Page Number: 073-082

Abstract:

IoT is an arising innovation that includes checking the climate, and the IoT networks are generally powerless against assaults because of the number of devices incorporated with the organization. The Intrusion detection method has been applied to investigate the abnormality in the organization. The Existing models have the limit of failure in the interruption discovery because of the models' overfitting. In this analysis, the Flower Pollination Algorithm (FPA) has been applied in the interruption identification technique to expand the IoT organization's productivity. The FPA technique has the benefit of significant distance fertilization and blossom consistency to investigate the highlights viably. The FPA chooses the IoT network includes and applies the highlights for the classifier to identify the charges. The classifiers like Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF) and Artificial Neural Network (ANN) are utilized to distinguish the interruptions in the organization. This trial result shows that the proposed FPA strategy with ANN has an accuracy of 99.5 % in the location, and the current ANN has 99.4 % precision in recognition. The FPA technique has the upsides of significant distance fertilization and blossom consistency, which adequately investigations the company.

Back Download