A new approach based on a rough set and a decision tree to bank customer credit evaluation

This paper proposes a new approach to customer credit evaluation by synthesizing the rough set theory and decision tree theory. It adopts and improves a algorithm by Yuan Zhen, et al, (2005) while applying the rough set theory in attribute reduction. It also applies the C4.5 Algorithm proposed by Quinlan to build a decision tree model and adjusts relevant parameters during tree pruning period. Experimental results show that the approach has a better performance in terms of efficiency as well as prediction accuracy.