Recall is the Proper Evaluation Metric for Word Segmentation

Yan Shao, Christian Hardmeier, Joakim Nivre


Abstract
We extensively analyse the correlations and drawbacks of conventionally employed evaluation metrics for word segmentation. Unlike in standard information retrieval, precision favours under-splitting systems and therefore can be misleading in word segmentation. Overall, based on both theoretical and experimental analysis, we propose that precision should be excluded from the standard evaluation metrics and that the evaluation score obtained by using only recall is sufficient and better correlated with the performance of word segmentation systems.
Anthology ID:
I17-2015
Volume:
Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 2: Short Papers)
Month:
November
Year:
2017
Address:
Taipei, Taiwan
Editors:
Greg Kondrak, Taro Watanabe
Venue:
IJCNLP
SIG:
Publisher:
Asian Federation of Natural Language Processing
Note:
Pages:
86–90
Language:
URL:
https://aclanthology.org/I17-2015
DOI:
Bibkey:
Cite (ACL):
Yan Shao, Christian Hardmeier, and Joakim Nivre. 2017. Recall is the Proper Evaluation Metric for Word Segmentation. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 2: Short Papers), pages 86–90, Taipei, Taiwan. Asian Federation of Natural Language Processing.
Cite (Informal):
Recall is the Proper Evaluation Metric for Word Segmentation (Shao et al., IJCNLP 2017)
Copy Citation:
PDF:
https://aclanthology.org/I17-2015.pdf