Deep learning for anomaly detection

Abstract

Anomaly detection methods are devoted to target detection schemes in which no priori information about the spectra of the targets of interest is available. This paper research on the 4 various types of anomaly detection machine learning anomaly models, namely Local Outlier Factor (LOF), Isolation Forest, One-Class Support Vector Machine (SVM), and Robust Covariance. Additionally, this paper shows the various steps in the implementation anomaly models and studies the effectiveness of each model in analysing an industrialized Multivariate Time-Series dataset.