View article

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 …