在许多优化问题中，目标值是连续的。对这类问题，首先对目标值进行离散化，再采用决策树方法提取规则。在一定程度上，相比直接对连续的目标值优化可提高正确率，并增加结果的可理解性。为了克服分段划分带来的突变性，可将目标值进行模糊划分，再采用决策树方法提取规则，这样进一步可提高正确率。 In many optimization problems, the target value is continuous. For such problems, we first discretize the target value and then use the decision tree method to extract the rules. To a certain extent, the accuracy can be improved and the comprehensibility of the result can be improved compared with directly optimizing the continuous target value. In order to overcome the mutation caused by segmentation, the target value can be divided into fuzzy points, and then the rules can be extracted by using the decision tree method, which further improves the correctness.