Adaptive feature-space conformal transformation for imbalanced-data learning
Gang Wu, Edward Yi Chang · 2003
When the training instances of the target class are heavily outnumbered by non-target training instances, SVMs can be ineffective in determin-ing the class boundary. To remedy this problem, we propose an adaptive conformal transformation (ACT) algorithm. ACT considers feature-space distance and the class-imbalance ratio when it per-forms conformal transformation on a kernel func-tion. Experimental results on UCI and real-world datasets show ACT to be effective in improving class prediction accuracy. 1.