Integrating Hybrid Systems for Robust Anomaly Detection in Big Data

Abstract

Robust systems capable of effectively identifying abnormalities in extensive data environments are necessary due to the rapid and significant growth of data across several sectors. Identifying data points that exhibit substantial deviations from the usual pattern is essential for anomaly detection, as it uncovers abnormal patterns that might potentially signify fraudulent activities, system failures, or other critical problems. However, traditional anomaly detection approaches fail to keep up with the volume, diversity, and velocity of big data growth. Recently, researchers have started investigating in hybrid systems that incorporate the greatest elements of several methodologies. In this paper, we present a new hybrid anomaly detection approach known as DT -SVMNB. To discriminate between normal and anomalous data in large data environments, our proposed system includes a range of machine learning approaches, such as decision trees, Support Vector Machines (SVM), and Naïve Bayesian classifiers (NBC). Our results show that the proposed model outperforms existing approaches in terms of A degree of accuracy (99.94 %) proving the usefulness and efficacy of our suggested solution.