Anomaly Detection in High-dimensional Big Data

With the world increasingly moving toward a data-driven setting, and with no generic approach for big data anomaly detection, the problem of high dimensionality is inevitable in many application areas. Hence, this thesis presents the state of anomaly detection in high-dimensional big data and proposes novel techniques for estimating locally relevant subspaces, searching candidate subspaces, and identifying outliers. However, high-dimensional big data brings a huge computational burden. Therefore, this thesis's main research objective is developing a scalable framework for anomaly detection that solves the problem of the curse of dimensionality in large data sets.