Detection of Botnet Using Deep Learning Architecture Using Chrome 23 Pattern with IOT

Abstract

The Internet world has recently grown with a lot of unstructured data and a growing number of cyber-attacks are targeting those devices due to rapid growth and popularization of Internet of Things (IoT) devices. In this research, we proposed a botnet attack detection system based on deep learning (DL) with ensemble architecture. We use two separate training models here; one is used to train the image and the other is used to train the numerical data as we take twitter information as input. An efficient approach is adopted to implement an image enhancement technique using Gaussian filter with a high dimensionality reduction in the pre-processing in the proposed work than the conventional techniques. For botnet attack detection using DL classifier, including clustering using KNN, the overall detection efficiency reaches about 92 %.