Title :
Cost-Sensitive Online Classification
Author :
Jialei Wang ; Peilin Zhao ; Hoi, Steven C. H.
Author_Institution :
Sch. of Comput. Eng., Nanyang Technol. Univ., Singapore, Singapore
Abstract :
Both cost-sensitive classification and online learning have been studied extensively in data mining and machine learning communities, respectively. It is a bit surprising that there was very limited comprehensive study for addressing an important intersecting problem, that is, cost-sensitive online classification. In this paper, we formally study this problem, and propose a new framework for cost-sensitive online classification by exploiting the idea of online gradient descent techniques. Based on the framework, we propose a family of cost-sensitive online classification algorithms, which are designed to directly optimize two well-known cost-sensitive measures: (i) maximization of weighted sum of sensitivity and specificity, and (ii) minimization of weighted misclassification cost. We analyze the theoretical bounds of the cost-sensitive measures made by the proposed algorithms, and extensively examine their empirical performance on a variety of cost-sensitive online classification tasks.
Keywords :
data mining; gradient methods; learning (artificial intelligence); minimisation; pattern classification; cost-sensitive measure; cost-sensitive online classification; data mining; machine learning; online gradient descent technique; online learning; sensitivity sum; specificity sum; weighted misclassification cost minimization; Accuracy; Data mining; Machine learning; Machine learning algorithms; Prediction algorithms; Sensitivity; classification; cost-sensitive learning; online learning;
Conference_Titel :
Data Mining (ICDM), 2012 IEEE 12th International Conference on
Conference_Location :
Brussels
Print_ISBN :
978-1-4673-4649-8
DOI :
10.1109/ICDM.2012.116
