【摘要】 针对数据挖掘中的分类问题,依据组合分类方法的思想,提出一种基于遗传算法的多重决策树组合分类方法.在这种组合分类方法中,先将概率度量水平的多重决策树并行组合,然后在组合算法中采用遗传算法优化连接权值矩阵.并且采用两组仿真数据对该方法进行测试和评估.实验结果表明,该组合分类方法比单个决策树具有更高的分类精度,并在保持分类结果良好可解释性的基础上优化了分类规则.更多还原

【Abstract】 For classification problems in data mining, based on thought of combination classification method, this paper proposes a combination classification method of multiple decision trees based on genetic algorithm. In the proposed combination classification method, multiple decision trees that adopt the method of probability measurement level output are parallel combined. Then genetic algorithm is used for the optimization of connection weight matrix in combination algorithm. Further more, two sets of simulation experiment data are used to test and evaluate the proposed combination classification method. Results of the experiments indicate that the proposed combination classification method has higher classification accuracy level than single decision tree. Moreover, it optimizes classification rules and sustains good interpretability for classification results.更多还原