Alberto Fernández - Learning from Imbalanced Data Sets

This Springer imprint is published by the registered company Springer Nature Switzerland AG

The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Learning with imbalanced data refers to the scenario in which the amounts of instances that represent the concepts in a given problem follow a different distribution. The main issue when addressing such a learning problem is when the accuracy achieved for each class is also different. This situation occurs since the learning process of most classification algorithm is often biased toward the majority class examples, so that minority ones are not well modeled into the final system. Being a very common scenario in real-life applications, the interest of researchers and practitioners on the topic has grown significantly during these years.

Based on the experience of the authors after several years focused on imbalanced classification, this book aims at offering a general and comprehensible overview for anyone interested in this area of study. It contains a formal description of the problem and focuses on its main features and the most relevant proposed solutions. Additionally, it considers the different scenarios in Data Science for which the imbalanced classification can suppose a real challenge.

After a gentle introduction to the KDD process and current state of Data Science in the first chapter, the book then stresses the gap with standard classification tasks by establishing the foundations and reviewing the case studies with a direct application in this area in Chap. on the extension of the problem for multi-class problems, where the former classical methods are no longer to be applied in a straightforward way.

The book includes in Chap..

This thorough review on the current and future state of imbalanced classification aims giving this topic the significance it deserves. In particular, the interest of research and academia is clearly shown by the rising number of publications and citations year by year. In the foreseeable future, it predictably will continue expanding with novel significant developments, as many contemporary real-world applications must be addressed from the viewpoint of imbalanced classification.

The intended audience of this book are developers and engineers aiming to apply imbalance-learning techniques to solve different kinds of real-world problems, as well as researchers and students needing a comprehensive review on techniques, methodologies, and tools for learning from imbalanced data.

We wish to thank all our collaborators of the research group Soft Computing and Intelligent Information Systems. We are also thankful to our families for their helpful support.