國 立 陽 明 交 通 大 學 演 講

講    題：基於過採樣之不平衡資料分類(Imbalanced data classification based on oversampling)

主 講 人：林真如(Chen-ju Lin) 教授(元智大學工業工程與管理學系)

主 持 人：劉建良教授

主辦單位：國立陽明交通大學工業工程與管理學系

時間：2023年10月16日(星期一) 13:30~15:10

地    點：綜合一館AB103室

演講摘要：

Classification task is complicated by several facts including skewed class proportion and unclear decision regions due to noise, class overlap, small disjunct, caused by large within-class variation. These issues make data classification difficult, reducing overall performance, and challenging to draw meaningful insights. In this talk, I will introduce the evidence-based adaptive oversampling algorithm (EVA-oversampling) based on Dempster-Shafer theory of evidence for imbalance classification. This technique involves assigning probability regarding class belonging for each instance to represent uncertainty that each data point may hold. Synthetic data points are generated to make up for the under-representation of minority instances on the region with high confidence, thereby strengthening the minority class region. The experiments revealed that the proposed method worked effectively even in situations where imbalanced counts and data complexity would normally pose significant obstacles. This approach performs better than several existing algorithms in terms of F1-measure and G-mean for highly imbalanced data. The instances of the minority group can be properly identified while maintaining the overall classification performance.

演講性質：學術研究專題

歡迎聽講