Real-valued negative selection algorithms: Ensuring data integrity through anomaly detection

Abstract

The Real-Valued Negative Selection algorithms which are the focal point of this work generate their detector set based on the points of self data. Self data is regarded as the normal behavioural pattern of the monitored system. An anomaly in data alters the confidentiality and integrity of its content thereby causing a defect for making useful and accurate decisions. Therefore, to correctly detect such an anomaly, this study applies the real-valued negative selection with; fixed-sized detectors (RNSA) and variable-sized detectors (V-Detector) for classification and detection of anomalies. Classifier algorithms of Support Vector Machine (SVM) and K-Nearest Neighbour (KNN) are used for benchmarking the performances of the real-valued negative selection algorithms. Experimental results illustrate that RNSA and V-Detector algorithms are suitable for the detection of anomalies, with the SVM and KNN producing significant efficiency rates. It was also gathered that V-Detector yielded superior performances with relation to the other algorithms.