M_DATA2

this paper generated a challenging two-dimensional imbalanced dataset M_DATA2 based on the normal distribution. Its scatter plot is shown in Figure 1. The dataset has a sample size of 1000, 2 feature variables, and the ratio between the number of samples in the two classes is 1:4. Among them, the samples of the minority class are divided into four parts by the majorityclasssamples, and there is a certain mixed area, which makes it difficult for the GNB classifier to classify effectivel本研究基于正态分布构建了一款极具挑战性的二维非平衡数据集M_DATA2，其散点图如图1所示。该数据集共包含1000个样本与2个特征变量，两类样本的数量之比为1:4。其中，少数类样本被多数类样本分割为四个部分，且存在一定的样本混合区域，这使得GNB（高斯朴素贝叶斯，Gaussian Naive Bayes）分类器难以实现有效的分类。