Dependency Parse Graph Neural Network for Text Classification

This paper proposed a text dependency parse graph convolutional network (text DP GCN) for text classification. First, the dependency syntactic information is integrated into the adjacency matrix of the text to obtain a text representation with stronger expression ability. Then, the text DP GCN is fine tuned using the pre training word vector and pooling operation. Experiments are carried out on several benchmark data sets, and the experimental results prove the effectiveness of the algorithm.