Neonatal Health Dataset

该数据集包含了母亲、胎儿、产科以及产程中的风险因素，用于预测不良分娩结果。该数据集共有34个特征维度，并融合了使用CTGAN和ADASYN生成的真实数据与合成数据。在平衡的测试集上，模型平均F1分数达到了0.784。数据规模较小，异常类别样本有112个，测试集中有38个样本。该数据集的任务是预测不良分娩结果并提供反事实解释。

This dataset contains risk factors related to mothers, fetuses, obstetrics, and labor stages, and is used to predict adverse birth outcomes. It has a total of 34 feature dimensions, and combines real and synthetic data generated via CTGAN and ADASYN. On the balanced test set, the model achieved an average F1-score of 0.784. The dataset is relatively small in scale, with 112 samples belonging to the anomalous class and 38 samples in the test set. The task of this dataset is to predict adverse birth outcomes and provide counterfactual explanations.