key idea #
- Graph构建，包含病人、症状、疾病三种节点类型，症状和疾病通过TransE初始化，病人取症状的均值
- 过2层GAT，采用多个二分类的分类器，子类疾病的概率=父类概率×子类概率（涨了0.6个点）
architecture #
Graph Neural Network Based Multi-Label Hierarchical Classification for Disease Predictions in General Practice
MedInfo
2023