Combining Recurrent and Convolutional Neural Networks for Relation Classification

TL;DR

This paper explores combining convolutional and recurrent neural networks for relation classification, introducing new architectures and a voting scheme that achieves state-of-the-art results on a benchmark dataset.

Contribution

It introduces a new context representation for CNNs, proposes connectionist bi-directional RNNs with ranking loss, and demonstrates improved accuracy through model combination.

Findings

Achieved state-of-the-art results on SemEval 2010 dataset

New context representation for CNNs improves classification

Combining CNN and RNN models enhances accuracy

Abstract

This paper investigates two different neural architectures for the task of relation classification: convolutional neural networks and recurrent neural networks. For both models, we demonstrate the effect of different architectural choices. We present a new context representation for convolutional neural networks for relation classification (extended middle context). Furthermore, we propose connectionist bi-directional recurrent neural networks and introduce ranking loss for their optimization. Finally, we show that combining convolutional and recurrent neural networks using a simple voting scheme is accurate enough to improve results. Our neural models achieve state-of-the-art results on the SemEval 2010 relation classification task.