A critical overview of outlier detection methods

One of the opening steps towards obtaining a reasoned analysis is the detection of outlaying observations. Even if outliers are often considered as a miscalculation or noise, they may bring significant information. For that reason, it is important to spot them prior to modeling and analysis. In this paper, we will present a structured and comprehensive review of the research on outlier detection. We have clustered existing methods into different categories based on the underlying approach adopted by each technique. In addition, for each category, we provide a discussion on the advantages and disadvantages of each method. Our paper's purpose is to assist the novice researcher, to produce clear ideas and to facilitate a better understanding of the different directions in which research has been done on this topic. (C) 2020 Elsevier Inc. All rights reserved.