Heterogeneous graph neural networks for noisy few-shot relation classification

Relation classification is an essential and fundamental task in natural language processing. Distant supervised methods have achieved great success on relation classification, which improve the performance of the task through automatically extending the dataset. However, the distant supervised methods also bring the problem of wrong labeling. Inspired by people learning new knowledge from only a few samples, we focus on predicting formerly unseen classes with a few labeled data. In this paper, we propose a heterogeneous graph neural network for few-shot relation classification, which contains sentence nodes and entity nodes. We build the heterogeneous graph based on the message passing between entity nodes and sentence nodes in the graph, which can capture rich neighborhood information of the graph. Besides, we introduce adversarial learning for training a robust model and evaluate our heterogeneous graph neural networks under the scene of introducing different rates of noise data. Experimental results have demonstrated that our model outperforms the state-of-the-art baseline models on the FewRel dataset. (C) 2020 Elsevier B.V. All rights reserved.