Cyber attacks have been the major concern with the growing advancement in technology. Complex security models have been developed to combat these attacks, yet none exhibit a full-proof performance. Recently, several machine learning (ML) methods have gained significant popularity in offering effective and efficient intrusion detection schemes which assist in proactive detection of multiple network intrusions, such as Denial of Service (DoS), Probe, Remote to User (R2L), User to Root attack (U2R). Multiple research works have been surveyed based on adopted ML methods (either signature-based or anomaly detection) and some of the useful observations, performance analysis and comparative study are highlighted in this paper. Among the different ML algorithms in survey, PSO-SVM algorithm has shown maximum accuracy. Using RBF-based classifier and C-means clustering algorithm, a new model i.e., combination of serial and parallel IDS is proposed in this paper. The detection rate to detect known and unknown intrusion is 99.5% and false positive rate is 1.3%. In PIDS (known intrusion classifier), the detection rate for DOS, probe, U2R and R2L is 99.7%, 98.8%, 99.4% and 98.5% and the False positive rate is 0.6%, 0.2%, 3% and 2.8% respectively. In SIDS (unknown intrusion classifier), the rate of intrusion detection is 99.1% and false positive rate is 1.62%. This proposed model has known intrusion detection accuracy similar to PSO – SVM and is better than all other models. Finally the future research directions relevant to this domain and contributions have been discussed.

Serial and Parallel based Intrusion Detection System using Machine Learning

One of the most common way of verification is ‘Face Recognition’ which can also be implemented for communication between machines and humans. In industries, corporate sectors, financial exchanges etc., face detection and recognition are used for the purpose of authentication. Another area of interest is the communication of information securely. In this paper, for face detection process Haar-classifiers are used which produces an accuracy of around 99.4% for still images and 98.6% for video recordings. Using two algorithms i.e. LBPH algorithm and Eigenfaces face recognition is done. The former produces a recognition accuracy of 99.39% for still images and 99.32% for video recordings and latter produces accuracies of 99.28% and 98.87% for the same. The paper also proposes a more secure modified AES, whose modification done by 43 variable S - Box generation out of which 9 S - Box return correct inverse S – Box with the help of variable irreducible polynomial in GF (28 ) finite field. Thus every time using the variable irreducible polynomial in GF (28 ) a new S- Box and inverse S-Box is generated which makes it difficult for the intruders to crack modified AES.

Authentication and secure communication by Haar Cascade Classifier, Eigen Face, LBP Histogram and variable irreducible polynomial in (2 8 ) finite field

Ayantika Sarkar

Papers

Serial and Parallel based Intrusion Detection System using Machine Learning

Authentication and secure communication by Haar Cascade Classifier, Eigen Face, LBP Histogram and variable irreducible polynomial in (2 8 ) finite field