---

ATTACK DETECTION IN EDGE NETWORKS WITH CONVOLUTIONAL NEURAL NETWORK BASED INSTRUCTION DETECTION SYSTEM

Abstract This paper suggests a stronger intrusion detection system (IDS) using the Convolutional Neural Network (CNN). The suggested CNN-IDS defines ransomware attacks by carefully analysing malicious packets that are crawling to bolt the user's device into a network. The strategy suggested is evaluated on both sides by the Ransomware attacks in a network scenario. In such network situations, theoretical findings are checked where the proposed solution aims to see if malware attacks have been detected on the network. The suggested CNN-IDS for ransomware attacks is accurate at 92.4% and the incorrect average ratio at 10 times cross validation is less than 7.6%. The result shows that in terms of consistency, robustness and precision, the proposed approach is effective than current IDS Authors M Keerthana Paavai Engineering College, India Keywords Intrusion Detection System, CNN, Ransomware Attack

Published By : ICTACT Published In : ICTACT Journal on Data Science and Machine Learning ( Volume: 1 , Issue: 3 , Pages: 100-103 ) Date of Publication : June 2020