Back to Patterns: Efficient Japanese Morphological Analysis with Feature-Sequence Trie

Naoki Yoshinaga


Abstract
Accurate neural models are much less efficient than non-neural models and are useless for processing billions of social media posts or handling user queries in real time with a limited budget. This study revisits the fastest pattern-based NLP methods to make them as accurate as possible, thus yielding a strikingly simple yet surprisingly accurate morphological analyzer for Japanese. The proposed method induces reliable patterns from a morphological dictionary and annotated data. Experimental results on two standard datasets confirm that the method exhibits comparable accuracy to learning-based baselines, while boasting a remarkable throughput of over 1,000,000 sentences per second on a single modern CPU. The source code is available at https://www.tkl.iis.u-tokyo.ac.jp/ynaga/jagger/
Anthology ID:
2023.acl-short.2
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
13–23
Language:
URL:
https://aclanthology.org/2023.acl-short.2
DOI:
10.18653/v1/2023.acl-short.2
Bibkey:
Cite (ACL):
Naoki Yoshinaga. 2023. Back to Patterns: Efficient Japanese Morphological Analysis with Feature-Sequence Trie. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 13–23, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Back to Patterns: Efficient Japanese Morphological Analysis with Feature-Sequence Trie (Yoshinaga, ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-short.2.pdf
Video:
 https://aclanthology.org/2023.acl-short.2.mp4