Proposed vision for Network Intrusion Detection System using Latent Semantic Analysis and data mining

In traditional and current Network Intrusion Detection Systems (NIDSs) the most important stage of them is; how to reduce the features space dimension to extract the only critical features to detect the intruders. Principle Component Analysis (PCA) is needed to detect intrusion by transform a set of features space to a lower dimension space retaining the variability of the original data from any change. But PCA doesn´t take the classes into account and is created for analyzing steady state processes, thus it is not able to handle any dynamic process, where wire, wireless and mobile traffic is dynamic (non-linear) therefore PCA is not feasible. In this research Latent Semantic Analysis (LSA) is proposed to reveal the variables in data. We are intending to introduce superior algorithm to frame Dynamic Principle Component Analysis (DPCA) in a heuristic fashion, this achievement will be explored in properties of emerging platforms such as smartness and mobility, and we need to merge DPCA and LSA to reveal semantics over variables; supported by ontology. Then using ID3 data mining and Artificial Intelligence showed how the intruding packets were detected and analyzed, this analysis has taken the stationary networks. The new stage of this research will take the mobility into account. So for, a group of algorithms have been created and correlated in parallel and serial configurations to present the proposed vision for NIDS. The results obtained from proposed system showing that accuracy and detection rate of ID3 classifiers is higher with (DPCA and LSA) than with traditional feature reduction methods.