Tu C.-YHSUAN-TIEN LIN2021-09-022021-09-022020https://www.scopus.com/inward/record.uri?eid=2-s2.0-85103812750&doi=10.1109%2fTAAI51410.2020.00017&partnerID=40&md5=5b74bda8ff534927cc9bf6c2ac6fb13bhttps://scholars.lib.ntu.edu.tw/handle/123456789/581365This paper proposes a method to improve the performance of imbalanced classification via reinforcement learning and cost-sensitive learning. Since the cost information is usually unavailable for cost-sensitive learning, we incorporate reinforcement learning to optimize the specified metric by adjusting the cost-matrix for the underlying cost-sensitive classifiers. Our experiment results show that, with the learned cost-matrix, the cost-sensitive classifiers can achieve better performance on several benchmark imbalanced data sets. ? 2020 IEEE.Benchmarking; Classification (of information); Learning systems; Cost information; Cost learning; Cost matrices; Cost-sensitive; Cost-sensitive learning; Imbalanced classification; Imbalanced Data-sets; Reinforcement learningconference paper10.1109/TAAI51410.2020.000172-s2.0-85103812750