Descriptive Analytics Solution for Attack Detection by Utilizing DL Strategies

Abstract

An intrusion detection system that employs a variety of system tasks and log files that are being generated on the host machine to detect HIDS refers to high-intensity distributed denial-of-service attacks. To enhance the capacity of intrusion detection systems, Big Data with Deep Learning Methods are combined. Deep Neural Network (DNN) and highly proficient approaches, Random Forest as well as Gradient Boosting Tree, are utilized to categories internet traffic datasets. Deep learning algorithms are widely used to develop an intrusion detection system (IDS) task of automatically recognizing and characterizing attacks at the host addressing performance in real time. Researchers utilize a homogeneity measure to analyze characteristics to identify its most productivity and organizational from dataset. As according to extensive experimental research, DNNs outperform classical machine learning classifiers in terms of performance. The findings shows that DNN has a good precision for different classifiers detection on datasets with accuracy rate for multi-class categorization. Employing Apache Flink to simplify the process and handling the streaming capabilities.