Augmentation of Chinese Character Representations with Compositional Graph Learning (Student Abstract)

Jason Wang, Kaiqun Fu, Zhiqian Chen, Chang-Tien Lu

AAAI 2022 pp. 13075-13076

doi:10.1609/AAAI.V36I11.21674 /aaai/2022/wang2022aaai-augmentation/

Abstract

Chinese characters have semantic-rich compositional information in radical form. While almost all previous research has applied CNNs to extract this compositional information, our work utilizes deep graph learning on a compact, graph-based representation of Chinese characters. This allows us to exploit temporal information within the strict stroke order used in writing characters. Our results show that our stroke-based model has potential for helping large-scale language models on some Chinese natural language understanding tasks. In particular, we demonstrate that our graph model produces more interpretable embeddings shown through word subtraction analogies and character embedding visualizations.

PDF AAAI Semantic Scholar

Cite

Text

Wang et al. "Augmentation of Chinese Character Representations with Compositional Graph Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2022. doi:10.1609/AAAI.V36I11.21674

Markdown

[Wang et al. "Augmentation of Chinese Character Representations with Compositional Graph Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2022.](https://mlanthology.org/aaai/2022/wang2022aaai-augmentation/) doi:10.1609/AAAI.V36I11.21674

BibTeX

@inproceedings{wang2022aaai-augmentation,
  title     = {{Augmentation of Chinese Character Representations with Compositional Graph Learning (Student Abstract)}},
  author    = {Wang, Jason and Fu, Kaiqun and Chen, Zhiqian and Lu, Chang-Tien},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2022},
  pages     = {13075-13076},
  doi       = {10.1609/AAAI.V36I11.21674},
  url       = {https://mlanthology.org/aaai/2022/wang2022aaai-augmentation/}
}