Ayuki Katayama


2024

pdf bib
Evaluating Language Models in Location Referring Expression Extraction from Early Modern and Contemporary Japanese Texts
Ayuki Katayama | Yusuke Sakai | Shohei Higashiyama | Hiroki Ouchi | Ayano Takeuchi | Ryo Bando | Yuta Hashimoto | Toshinobu Ogiso | Taro Watanabe
Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities

Automatic extraction of geographic information, including Location Referring Expressions (LREs), can aid humanities research in analyzing large collections of historical texts. In this study, to investigate how accurate pretrained Transformer language models (LMs) can extract LREs from historical texts, we evaluate two representative types of LMs, namely, masked language model and causal language model, using early modern and contemporary Japanese datasets. Our experimental results demonstrated the potential of contemporary LMs for historical texts, but also suggest the need for further model enhancement, such as pretraining on historical texts.