A Measure Optimized Cost-Sensitive Learning Framework for Imbalanced Data Classification

View Sample PDF

Author(s): Peng Cao (Northeastern University, China & University of Alberta, Canada), Osmar Zaiane (University of Alberta, Canada)and Dazhe Zhao (Northeastern University, China)
Copyright: 2014
Pages: 28
Source title: Biologically-Inspired Techniques for Knowledge Discovery and Data Mining
Source Author(s)/Editor(s): Shafiq Alam (University of Auckland, New Zealand), Gillian Dobbie (University of Auckland, New Zealand), Yun Sing Koh (University of Auckland, New Zealand)and Saeed ur Rehman (Unitec Institute of Technology, New Zealand)
DOI: 10.4018/978-1-4666-6078-6.ch003

Keywords: Data Mining / Data Mining and Databases / Information Science Reference / Library & Information Science

Purchase

View A Measure Optimized Cost-Sensitive Learning Framework for Imbalanced Data Classification on the publisher's website for pricing and purchasing information.

Abstract

Class imbalance is one of the challenging problems for machine-learning in many real-world applications. Many methods have been proposed to address and attempt to solve the problem, including sampling and cost-sensitive learning. The latter has attracted significant attention in recent years to solve the problem, but it is difficult to determine the precise misclassification costs in practice. There are also other factors that influence the performance of the classification including the input feature subset and the intrinsic parameters of the classifier. This chapter presents an effective wrapper framework incorporating the evaluation measure (AUC and G-mean) into the objective function of cost sensitive learning directly to improve the performance of classification by simultaneously optimizing the best pair of feature subset, intrinsic parameters, and misclassification cost parameter. The optimization is based on Particle Swarm Optimization (PSO). The authors use two different common methods, support vector machine and feed forward neural networks, to evaluate the proposed framework. Experimental results on various standard benchmark datasets with different ratios of imbalance and a real-world problem show that the proposed method is effective in comparison with commonly used sampling techniques.

The IRMA Community

Research IRM

A Measure Optimized Cost-Sensitive Learning Framework for Imbalanced Data Classification

Purchase

Abstract

Related Content

IRMA Sponsors