Chadza, Timothy Kyriakopoulos, Kostas Lambotharan, Sangarapillai Predicting Sequential Network Attacks Using Hidden Markov Model - MATLAB Code <p>This code reproduces the results for the research paper titled <strong>Contemporary Sequential Network Attacks Prediction Using Hidden Markov Model</strong> by Timothy Chadza, Konstantinos Kyriakopoulos & Sangarapillai Lambotharan presented at the 17th International Conference on Privacy, Security and Trust (PST), IEEE, Fredericton, NB, Canada, 2019. This is published under GNU GENERAL PUBLIC LICENSE Version 3. If you use this code, please, cite the above paper.<br></p><p><br></p><p>This work is on the evaluation of detection accuracy for determining all states, the current state, and the prediction of next state of an observation sequence, using the two conventional hidden Markov model training algorithms, namely, Baum Welch and Viterbi training. The training algorithms are initialised using uniform, random and count-based parameters. The experimental evaluation is conducted on the CSE-CIC-IDS2018, a modern dataset comprising seven different attack scenarios over a large network environment. The different attacks are sequentially aggregated to constitute an attack sequence. Viterbi decoding has been used to estimate the next state upon computation of the next attack manifestation.</p><p><br></p><p>The code is run by executing the main.m file in MATLAB.</p> Intrusion prediction;Hidden Markov model, HMM;Baum Welch;Viterbi training;Intrusion detection system (IDS);CSE-CIC-IDS2018 dataset;Coding and Information Theory;Computer Software;Networking and Communications;Computer System Security 2020-07-31
    https://repository.lboro.ac.uk/articles/software/Predicting_Sequential_Network_Attacks_Using_Hidden_Markov_Model_-_MATLAB_Code/12719681
10.17028/rd.lboro.12719681.v1