An efficient cyber security system based on flow-based anomaly detection using Artificial neural network

Abstract

Cyber security is developing factor for protecting internet resources by handing various monitoring feature based support to improve the security. Increasing internet cries in the defined facts for need of advance met in cyber security. Most internet attacker’s theft the information through malicious activities, false data injection, hacking and make soon creating procedures. In most cases cyber sercuity failed to detect the malicious activities because the monitoring feature analyses improper to predict the result in previous machine learning algorithms. TO resolve this problem to propose an advance cyber security based on flow-based anomaly detection using Min max game theory optimized artificial neural network (MMGT-ANN). The reprocessing was carried out with KDD crime dataset. Then Data driven network model is applied to monitor the feature margins and defect scaling rate. Based on the feature scaling rate Transmission Flow defect rate is estimated and applied with Min max Game theory to select the feature limits. Then features are trained with optimized ANN to detect the crime rate. By the attention of the proposed system achieves higher performance in precision rate to attain higher detection accuracy with lower time complexity compared to the other system.

Ensemble-Learning Framework for Intrusion Detection to Enhance Internet of Things’ Devices Security

“…The IoT consists of a network of interconnected nodes that constantly communicate with one another and process information over a variety of network protocols. Because of the ease with which these protocols can be exploited, it has been discovered that they pose a considerable risk to the confidentiality of the transferred data [ 5 ].…”