一种新的模糊决策树模型及其应用

模糊决策树是决策树在模糊环境下的一种推广,虽然其表示形式更符合人类的思维，但在构造时会增加预处理的工作量和创建树时的开销。基于这种情况，提出了一种混合算法，算法保留了较少属性值的Shannon熵，计算多属性和连续属性值模糊化后的模糊熵。将该算法应用于滑坡数据的挖掘中，得到了更易于理解的决策树和有效的规则，与传统算法的性能比较也证明了该算法的有效性。

A new fuzzy decision tree model and its application

A fuzzy decision tree is the generalization of a decision tree in a fuzzy environment. The knowledge represented by a fuzzy decision tree is more natural to the way of human thinking, but there is the additional work of preprocessing and cost of constructing trees. A new hybrid fuzzy decision tree model was proposed. The new algorithm calculates the entropy of multi-valued and continuous-valued attributes after fuzzification and Shannon entropy of other attributes was calculated by this new algorithm. Simulation results confirm that the proposed model can lead to tmderstandable decision trees and extract effective rules. Experimental results show that the proposed model is more effective and efficient than a fuzzy decision tree and C4.5.