A Novel Neural Network Model for Joint POS Tagging and Graph-based Dependency Parsing
This work addresses the feature-engineering problem in natural language processing for researchers and practitioners, though it is incremental as it builds on existing neural network methods.
The paper tackles the problem of joint POS tagging and graph-based dependency parsing by introducing a neural network model that uses bidirectional LSTMs to share feature representations, achieving a new state of the art by outperforming existing models on 19 languages from the Universal Dependencies project.
We present a novel neural network model that learns POS tagging and graph-based dependency parsing jointly. Our model uses bidirectional LSTMs to learn feature representations shared for both POS tagging and dependency parsing tasks, thus handling the feature-engineering problem. Our extensive experiments, on 19 languages from the Universal Dependencies project, show that our model outperforms the state-of-the-art neural network-based Stack-propagation model for joint POS tagging and transition-based dependency parsing, resulting in a new state of the art. Our code is open-source and available together with pre-trained models at: https://github.com/datquocnguyen/jPTDP