TY - GEN
T1 - Learning misclassification costs for imbalanced datasets, application in gene expression data classification
AU - Lu, Huijuan
AU - Xu, Yige
AU - Ye, Minchao
AU - Yan, Ke
AU - Jin, Qun
AU - Gao, Zhigang
N1 - Funding Information:
Acknowledgments. This study is supported by National Natural Science Foundation of China (Nos. 61272315, 61402417, 61602431 and 61701468), Zhejiang Provincial Natural Science Foundation (Nos. Y1110342, LY15F020037) and International Cooperation Project of Zhejiang Provincial Science and Technology Department (No. 2017C34003).
Publisher Copyright:
© 2018, Springer International Publishing AG, part of Springer Nature.
PY - 2018
Y1 - 2018
N2 - Cost-sensitive algorithms have been widely used to solve imbalanced classification problem. However, the misclassification costs are usually determined empirically, leading to uncertain performance. Hence an effective method is desired to automatically calculate the optimal cost weights. Targeting at the highest weighted classification accuracy (WCA), we propose two approaches to search for the optimal cost weights, including grid searching and function fitting. In experiments, we classify imbalanced gene expression data using extreme learning machine to test the cost weights obtained by the two approaches. Comprehensive experimental results show that the function fitting is more efficient which can well find the optimal cost weights with acceptable WCA.
AB - Cost-sensitive algorithms have been widely used to solve imbalanced classification problem. However, the misclassification costs are usually determined empirically, leading to uncertain performance. Hence an effective method is desired to automatically calculate the optimal cost weights. Targeting at the highest weighted classification accuracy (WCA), we propose two approaches to search for the optimal cost weights, including grid searching and function fitting. In experiments, we classify imbalanced gene expression data using extreme learning machine to test the cost weights obtained by the two approaches. Comprehensive experimental results show that the function fitting is more efficient which can well find the optimal cost weights with acceptable WCA.
KW - Correct classification rate
KW - Cost-sensitive
KW - Misclassification cost
KW - Parameter fitting
UR - http://www.scopus.com/inward/record.url?scp=85051872559&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85051872559&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-95930-6_47
DO - 10.1007/978-3-319-95930-6_47
M3 - Conference contribution
AN - SCOPUS:85051872559
SN - 9783319959290
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 513
EP - 519
BT - Intelligent Computing Theories and Application - 14th International Conference, ICIC 2018, Proceedings
A2 - Premaratne, Prashan
A2 - Gupta, Phalguni
A2 - Huang, De-Shuang
A2 - Bevilacqua, Vitoantonio
PB - Springer Verlag
T2 - 14th International Conference on Intelligent Computing, ICIC 2018
Y2 - 15 August 2018 through 18 August 2018
ER -