Open Access

ARTICLE

Adversarial Learning for Distant Supervised Relation Extraction

Daojian Zeng^1,3, Yuan Dai^1,3, Feng Li^1,3, R. Simon Sherratt², Jin Wang^3,*

1 Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation, School of Computer & Communication Engineering, Changsha University of Science & Technology, Changsha, 410114, P. R. China. Email: zengdj@csust.edu.cn.
2 Department of Biomedical Engineering, University of Reading, UK. Email: r.s.sherratt@reading.ac.uk.
3 School of Computer & Communication Engineering, Changsha University of Science & Technology, Changsha, 410114, P. R. China. Email: jinwang@csust.edu.cn.

* Corresponding author: Jin Wang. Email: .

Computers, Materials & Continua 2018, 55(1), 121-136. https://doi.org/10.3970/cmc.2018.055.121

Abstract

Recently, many researchers have concentrated on using neural networks to learn features for Distant Supervised Relation Extraction (DSRE). These approaches generally use a softmax classifier with cross-entropy loss, which inevitably brings the noise of artificial class NA into classification process. To address the shortcoming, the classifier with ranking loss is employed to DSRE. Uniformly randomly selecting a relation or heuristically selecting the highest score among all incorrect relations are two common methods for generating a negative class in the ranking loss function. However, the majority of the generated negative class can be easily discriminated from positive class and will contribute little towards the training. Inspired by Generative Adversarial Networks (GANs), we use a neural network as the negative class generator to assist the training of our desired model, which acts as the discriminator in GANs. Through the alternating optimization of generator and discriminator, the generator is learning to produce more and more discriminable negative classes and the discriminator has to become better as well. This framework is independent of the concrete form of generator and discriminator. In this paper, we use a two layers fully-connected neural network as the generator and the Piecewise Convolutional Neural Networks (PCNNs) as the discriminator. Experiment results show that our proposed GAN-based method is effective and performs better than state-of-the-art methods.

---

Keywords

---