Predictive security model using data mining

摘要: We propose a practical and predictive security model for intrusion detection in computer networking environment using data mining. This uses classification regression technique The goal of the proposed is to identify significant variables that measure network from wealth raw perform an efficient vulnerability evaluation based on those variables. Analysis experimental results conducted DARPA benchmark dataset shows CART (classification trees) approach performs better compared other models, like random projection principal component analysis. also indicate performance not significantly affected, even as dimension input decreases, without compromising prediction success rate.