Multi-label text classification based on semantic-sensitive graph convolutional network

De-Long Zeng, Enze Zha, Jiayi Kuang, Ying Shen

Abstract: Highlights•Integrating relationships among texts, words, and labels to construct a global graph.•S-GCN Combines global information and internal contextual associations of documents.•Texts not in the graph can be classified by storing hidden representations of words.