Indian Institute of Technology Guwahati

About: Indian Institute of Technology Guwahati is a education organization based out in Guwahati, Assam, India. It is known for research contribution in the topics: Catalysis & Computer science. The organization has 6933 authors who have published 17102 publications receiving 257351 citations.

A Neural Network based system for Intrusion Detection and attack classification

Abstract: Anomaly based Intrusion Detection Systems (IDSs) are known to achieve high accuracy and detection rate. However, a significant computational overhead is incurred in training and deploying them. In this paper, we aim to address this issue by proposing a simple Artificial Neural Network (ANN) based IDS model. The proposed IDS model uses the feed forward and the back propagation algorithms along with various other optimization techniques to minimize the overall computational overhead, while at the same time maintain a high performance level. Experimental results on the benchmark NSL-KDD dataset shows that the performance (accuracy and detection rate) of the proposed ANN based IDS model is at par and in some cases even better than other IDS models. Owing to its high performance and low computational overhead, the proposed ANN based IDS model is a suitable candidate for real time deployment and intrusion detection analysis.

Extended Born-Oppenheimer equation for a three-state system.

Abstract: We present explicit forms of nonadiabatic coupling (NAC) elements of nuclear Schrodinger equation (SE) for a coupled three-state electronic manifold in terms of mixing angles of real electronic basis functions. If the adiabatic-diabatic transformation (ADT) angles are the mixing angles of electronic bases, ADT matrix transforms away the NAC terms and brings diabatic form of SE. ADT and NAC matrices are shown to satisfy a curl condition with nonzero divergence. We have demonstrated that the formulation of extended Born-Oppenheimer (EBO) equation from any three-state BO system is possible only when there exists a coordinate-independent ratio of the gradients for each pair of mixing angles. On the contrary, since such relations among the mixing angles lead to zero curl, we explore its validity analytically around conical intersection(s) and support numerically considering two nuclear-coordinate-dependent three surface BO models. Numerical calculations are performed by using newly derived diabatic and EBO equations and expected transition probabilities are obtained.