-
When Classical Chinese Meets Machine Learning: Explaining the Relative Performances of Word and Sentence Segmentation Tasks
- Author(s):
- Wei-Ting Chang, Chang-Ting Chu, CHAO-LIN LIU (see profile) , Ti-Yong Zheng
- Date:
- 2020
- Group(s):
- DH2020, Digital Humanists
- Subject(s):
- Artificial intelligence, China, Educaton, Digital humanities, Research, Methodology, Machine learning, Natural language processing (Computer science)
- Item Type:
- Video
- Tag(s):
- deep learning, explainable artificial intelligence, information extraction, text mining, Chinese studies, Digital humanities research and methodology, Natural language processing
- Permanent URL:
- http://dx.doi.org/10.17613/r3kx-ng78
- Abstract:
- We consider three major text sources about the Tang Dynasty of China in our experiments that aim to segment text written in classical Chinese. These corpora include a collection of Tang Tomb Biographies, the New Tang Book, and the Old Tang Book. We show that it is possible to achieve satisfactory segmentation results with the deep learning approach. More interestingly, we found that some of the relative superiority that we observed among different designs of experiments may be explainable. The relative relevance among the training corpora provide hints/explanation for the observed differences in segmentation results that were achieved when we employed different combinations of corpora to train the classifiers.
- Notes:
- A written summary for this talk is available at an ADHO site (http://dh2020.adho.org/abstracts/) and the arXiv (https://arxiv.org/).
- Metadata:
- xml
- Status:
- Published
- Last Updated:
- 3 years ago
- License:
- All Rights Reserved
dh2020.clliu_.mp4