Abstract
Online class imbalance learning deals with data streams having very skewed class distributions in a timely fashion. Although a few methods have been proposed to handle such problems, most of them focus on two-class cases. Multi-class imbalance imposes additional challenges in learning. This paper studies the combined challenges posed by multiclass imbalance and online learning, and aims at a more effective and adaptive solution. First, we introduce two resampling-based ensemble methods, called MOOB and MUOB, which can process multi-class data directly and strictly online with an adaptive sampling rate. Then, we look into the impact of multi-minority and multi-majority cases on MOOB and MUOB in comparison to other methods under stationary and dynamic scenarios. Both multi-minority and multi-majority make a negative impact. MOOB shows the best and most stable Gmean in most stationary and dynamic cases.
Original language | English |
---|---|
Title of host publication | IJCAI'16: Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence |
Publisher | AAAI press |
Pages | 2118-2124 |
Number of pages | 7 |
ISBN (Print) | 9781577357704 |
DOIs | |
Publication status | Published - Jul 2016 |
Externally published | Yes |
Event | 25th International Joint Conference on Artificial Intelligence, IJCAI'16 - , United States Duration: 9 Jul 2016 → 15 Jul 2016 |
Conference
Conference | 25th International Joint Conference on Artificial Intelligence, IJCAI'16 |
---|---|
Country/Territory | United States |
Period | 9/07/16 → 15/07/16 |
Funding
This work was supported by EPSRC (Grant No. EP/K001523/1). Xin Yao was also supported by a Royal Society Wolfson Research Merit Award