A survey of predictive modeling on imbalanced domains

Abstract

Many real-world data-mining applications involve obtaining predictive models using datasets with stronglyimbalanced distributions of the target variable. Frequently, the least-common values of this target variableare associated with events that are highly relevant for end users (e.g., fraud detection, unusual returnson stock markets, anticipation of catastrophes, etc.). Moreover, the events may have different costs andbenefits, which, when associated with the rarity of some of them on the available training data, createsserious problems to predictive modeling techniques. This article presents a survey of existing techniques forhandling these important applications of predictive analytics. Although most of the existing work addressesclassification tasks (nominal target variables), we also describe methods designed to handle similar problemswithin regression tasks (numeric target variables). In this survey, we discuss the main challenges raisedby imbalanced domains, propose a definition of the problem, describe the main approaches to these tasks,propose a taxonomy of the methods, summarize the conclusions of existing comparative studies as well assome theoretical analyses of some methods, and refer to some related problems within predictive modeling.