KGAT: An Enhanced Graph-Based Model for Text Classification

As a fundamental task in natural language processing, text classification, which is to predict the class label of a given text, has been intensively studied. Consequently, a host of techniques have been developed, among which techniques that are based on graph neural network and its variant e.g., graph attention network (GAT) achieved impressive performances, as they show superiority in dealing with complex graphstructured data. Despite effectiveness, most of these techniques suffer from several limitations, e.g., incapability in well-capturing correlation among words in a text. In light of these, we propose a comprehensive approach KGAT which incorporates multi-head GAT with enhanced attention and customized ReadOut operation for text classification. (1) Our approach constructs a text graph GT with edge weights from a text such that both semantic and structural information (with correlation degree) can be well captured. (2) On text graph GT, a novel attention mechanism is incorporated in a multi-head GAT for representation learning. (3) Our approach customizes ReadOut operation such that the representation of a text is refined by using a set of influential nodes of GT. Intensive experimental studies on both typical benchmark datasets and a newly created one (Sensitive) show that our approach substantially outperforms other baseline methods and yields a promising technique for text classification.