This document presents an overview of outlier detection, including definitions of outliers, common applications of outlier detection, and an algorithm called the depth-based approach. The depth-based approach models outliers as data objects located on the outer layers of convex hulls of the data space. It works by organizing data objects into convex hull layers and identifying outliers as those on outer layers, based on the assumption that outliers are located at the border of the data space while normal data is at the center.