Chinese discourse parsing: Model and evaluation
Journal
LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings
Pages
1019-1024
Date Issued
2020
Author(s)
Abstract
Chinese discourse parsing, which aims to identify the hierarchical relationships of Chinese elementary discourse units, has not yet a consistent evaluation metric. Although Parseval is commonly used, variations of evaluation differ from three aspects: micro vs. macro F1 scores, binary vs. multiway ground truth, and left-heavy vs. right-heavy binarization. In this paper, we first propose a neural network model that unifies a pre-trained transformer and CKY-like algorithm, and then compare it with the previous models with different evaluation scenarios. The experimental results show that our model outperforms the previous systems. We conclude that (1) the pre-trained context embedding provides effective solutions to deal with implicit semantics in Chinese texts, and (2) using multiway ground truth is helpful since different binarization approaches lead to significant differences in performance. ? European Language Resources Association (ELRA), licensed under CC-BY-NC
Subjects
Binarizations; Chinese text; Discourse parsing; Effective solution; Evaluation metrics; Ground truth; Implicit semantics; Neural network model; Semantics
Type
conference paper