This is a brief summary of paper for me to study and organize it, Compositional Sequence Labeling Models for Error Detection in Learner Writing (Rei and Yannakoudakis., ACL 2016) I read and studied.
This paper popose a method to detect error in learner writing by tackling it as sequence labeling task.
Empirically, they experimented the method with compoistion architectures which are convolution neural network, recurrent netword, and Long-short term memory as follows:
In paritcular, for LSTM, they used peephole LSTM.
Note(Abstract):
In this paper, they present the first experiments using neural network models for the task of error detection in learner writing. We perform a systematic comparison of alternative compositional architectures and propose a framework for error detection based on bidirectional LSTMs. Experiments on the CoNLL-14 shared task dataset show the model is able to outperform other participants on detecting errors in learner writing. Finally, the model is integrated with a publicly deployed self-assessment system, leading to performance comparable to human annotators.
Download URL:
The paper: Compositional Sequence Labeling Models for Error Detection in Learner Writing (Rei and Yannakoudakis., ACL 2016)
The paper: Compositional Sequence Labeling Models for Error Detection in Learner Writing (Rei and Yannakoudakis., ACL 2016)
Reference
- Paper
- How to use html for alert