Dataset and code for the paper: Recurrent Interaction Network for Jointly Extracting Entities and Classifying Relations. Kai Sun, Richong Zhang, Samuel Mensah, Yongyi Mao, Xudong Liu. EMNLP 2020. [pdf]
In this paper, we focus on leveraging the multitask learning method to solve the joint entities and relations extraction task by taking entity recognition (ER) as one task, and relation classification (RC) as another task. The ER aims to extract all entities in a sentence, and the RC aims to classify the relations between all word pairs. Existing methods using multi-task learning techniques to address the problem learn implicit interactions among the two tasks through a shared network, where the shared information is passed into the task-specific networks for prediction. However, such an approach hinders the model from learning explicit interactions between the two tasks to improve the performance on the individual tasks. As a solution, we design a multitask learning model which we refer to as recurrent interaction network (RIN) which allows the dynamically learning of explicit interactions between the output of two tasks, to effectively model task-specific features for classification.
- Python 3.6.7
- PyTorch 1.2.0
- NumPy 1.17.2
- 100-dimensional word vectors for nyt and webnlg here.
- GloVe pre-trained word vectors for nyt10 and nyt11:
- Download pre-trained word vectors here.
- Extract the glove.840B.300d.zip to the
dataset/glove/folder.
Training the model:
python train.py --dataset [dataset]Prepare vocabulary files for nyt10 and nyt11 datasets (nyt10 and nyt11 will be uploaded soon):
python prepare_vocab.py --dataset [dataset]If this work is helpful, please cite as:
@inproceedings{Sun2020RIN,
author = {Kai Sun and
Richong Zhang and
Samuel Mensah and
Yongyi Mao and
Xudong Liu},
title = {Recurrent Interaction Network for Jointly Extracting Entities and Classifying Relations},
year = {2020}
}MIT