DAD: A Distributed Anomaly Detection system using ensemble one-class statistical learning in edge networks

Intrusion Detection in SCADA Based Power Grids: Recursive Feature Elimination Model With Majority Vote Ensemble Algorithm

“…Over a past few years, many Intrusion Detection Systems for various communication technologies have been proposed to detect the threats more accurately based on ensemble learning [34], [35], [36], [37], [38], [39], [40], [41].…”

“…Cloudbased solutions for distributed anomaly detection systems can be found in [40]. In [41], the authors propose a Gaussian mixture based anomaly detection technique that relies on ensemble one-class statistical learning model that is designed to effectively recognize zero day attacks in real-time using the concept of edge networks.…”

IoT anomaly detection methods and applications: A survey

“…The edge network, including routers and switches in locations such as the airport and gateway, is safeguarded with a proposed ensemble one-class statistical learning model by Moustafa et al [93] that implements Gaussian Mixture-based Correntropy. The proposed system aims to identify zero-day attacks and also attempts to develop a legitimate profile for new data flows as they occur.…”

Identifying performance anomalies in fluctuating cloud environments: A robust correlative-GNN-based explainable approach

“…With monitoring tools, performance data such as resource usage of cloud computing systems can be collected [2]. At the same time, anomaly detection to build a profile of performance data and detect deviations from the profile for cloud computing systems can be developed [3]. Considering it is tedious and time-consuming to label data manually because various anomalies exist, unsupervised learning involves picking up interesting structures in the data, and learning features without labels is popular [4].…”