基于模糊决策的随机森林算法  

Algorithm of random forest based on fuzzy decision

作　　者：史金余[1] 杨泽宇 谢兄[1] SHI Jin-yu;YANG Ze-yu;XIE Xiong(School of Information Science and Technology,Dalian Maritime University,Dalian 116026,China)

机构地区：[1]大连海事大学信息科学技术学院,辽宁大连116026

出　　处：《计算机工程与设计》2020年第8期2207-2212,共6页Computer Engineering and Design

基　　金：国家自然科学基金青年科学基金项目(61702074);辽宁省自然科学基金项目(20170520196);中央高校基本科研基金项目(3132016308、3132018197)。

摘　　要：为减轻现有随机森林算法中重要度低的属性对分类结果的影响,提出一种基于模糊决策的随机森林算法(FRF)。根据随机森林中对决策树设置的属性重要度的值,舍去重要度较低的属性,把完全生长的决策树转化为模糊决策树。由叶子结点中各个类别被赋予的相应权重和类别分布情况,计算得到能够判定样本类别的概率;进行随机森林算法优化,提高分类正确率。实验结果表明了优化后的随机森林算法方法在UCI标准数据集上的有效性,分类正确率得到了提高。A random forest algorithm based on fuzzy decision was proposed to solve the influence of low-importance attributes on classification results in existing random forest algorithms.According to the setting value of attribute’s importance in the decision tree of the random forest,the fully grown decision tree was transformed into a fuzzy decision tree by discarding the attri-butes with lower importance.The probability of determining the samples class was calculated according to granted weight and distribution of each class in the leaf nodes.The random forest algorithm was optimized to increase the classification accuracy.Experimental results show that the optimized random forest algorithm is effective on the UCI standard data sets and can improve the classification accuracy.

关 键 词：决策树 随机森林 模糊决策 分类器 属性重要度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]