Regularized Graph Convolutional Networks for Short Text Classification [conference paper]

Conference

Proceedings of the 28th International Conference on Computational Linguistics - December 12, 2020

Authors

Kshitij Tayal (Ph.D. student), Nikhil Rao, Saurabh Agarwal, Xiaowei Jia, Karthik Subbian, Vipin Kumar (professor)

Abstract

Short text classification is a fundamental problem in natural language processing, social network analysis, and e-commerce. The lack of structure in short text sequences limits the success of popular NLP methods based on deep learning. Simpler methods that rely on bag-of-words representations tend to perform on par with complex deep learning methods. To tackle the limitations of textual features in short text, we propose a Graph-regularized Graph Convolution Network (GR-GCN), which augments graph convolution networks by incorporating label dependencies in the output space. Our model achieves state-of-the-art results on both proprietary and external datasets, outperforming several baseline methods by up to 6%. Furthermore, we show that compared to baseline methods, GR-GCN is more robust to noise in textual features.

Link to full paper

Regularized Graph Convolutional Networks for Short Text Classification

Keywords

data mining, machine learning