基于模糊神经网的决策树生成

Induction of decision trees based on a fuzzy neural network

数据挖掘技术能从大量数据中挖掘和发现有价值和隐含的知识 ,因而得到广泛的研究和应用。提出了一种基于五层模糊神经网络的决策树生成方法 :首先运用五层模糊神经网络学习变量间的模糊映射关系 ,然后从中生成模糊决策树。这种方法利用了五层模糊神经网络学习后的模糊映射强度 ,并能实现模糊决策树的剪枝优化 ,提高了算法的正确率和效率。

Data mining technique can mine and discover valuable and hidden knowledge from databases, so it has been widely studied and applied. In this paper, an approach for induction of decision trees based on a fuzzy neural network is presented. First, fuzzy mapping relations among variables are learnt by a five-layer fuzzy neural network, then fuzzy decision trees are extracted from that network. The approach makes use of the weights of fuzzy mappings in the learnt fuzzy neural network, can realize the optimization of fuzzy decision trees by branches cutting, and improves the ratio of correctness and efficiency of induction of decision trees. \;