Lung-Hao Lee, Yuh-Shyang Wang, Po-Chen Lin, Chih-Te Hung, and Yuen-Hsien Tseng.
In Proceedings of the 28th International Conference on Computers in Education (ICCE’20), pages 558-560.
Abstract
In this paper, we proposed a Multi-Channel Convolutional Neural Network with Bidirectional Long Short-Term Memory (MC-CNN-BiLSTM) model for Chinese grammatical error detection. The TOCFL learner corpus is adopted to measure the system capability of indicating whether a sentence contains errors or not. Our model performs better than a previous CNN-LSTM model that reflects the effectiveness of multi-channel embedding representation.
