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. Aimed at the problem of classification in data mining, based on the idea of ​​combinatorial classification, a multi-decision tree classification method based on genetic algorithm is proposed.In this method, multiple decision trees of probability metric level are combined in parallel and then Genetic algorithm was used to optimize the connection weight matrix in the combined algorithm, and two sets of simulation data were used to test and evaluate the method.The experimental results show that the combined classification method has higher classification accuracy than single decision tree, The results are well interpretable based on the optimization of the classification rules.