Experimental Study on Class Imbalance Problem Using an Oil Spill Training Data Set

Xi Qin Ouyang *

Department of Mathematics, Ganzhou Teachers College, Ganzhou 341000, Jiangxi, China.

Yuan Ping Chen

Department of Foundation, Jiangxi Environmental Engineering Vocational College, Ganzhou 341000, Jiangxi, China.

Bing Hui Wei

Department of Computer Science, Jiangxi University of Science and Technology, Ganzhou 341000, Jiangxi, China.

*Author to whom correspondence should be addressed.


Abstract

There is a paucity of research on one of the key issues in oil spill detection: the imbalanced training set learning problem. This paper performs experiments to show the influence of the imbalanced learning problem (ILP) on oil spill detection and devises a novel framework to tackle this problem. Experimental results show that an imbalanced training set degenerate the performance of oil spill detection, and our proposed framework achieves a better performance based on F-measure.

Keywords: Imbalanced problem, oil spill detection, data set.


How to Cite

Ouyang, Xi Qin, Yuan Ping Chen, and Bing Hui Wei. 2017. “Experimental Study on Class Imbalance Problem Using an Oil Spill Training Data Set”. Journal of Advances in Mathematics and Computer Science 21 (5):1-9. https://doi.org/10.9734/BJMCS/2017/32860.

Downloads

Download data is not yet available.