Vol. 28:2 (2022) ► pp.299–327
Automatic medical term extraction from Vietnamese clinical texts
In this paper, we propose the first method for automatic Vietnamese medical term discovery and extraction from clinical texts. The method combines linguistic filtering based on our defined open patterns with nested term extraction and statistical ranking using C-value. It does not require annotated corpora, external data resources, parameter settings, or term length restriction. Beside its specialty in handling Vietnamese medical terms, another novelty is that it uses Pointwise Mutual Information to split nested terms and the disjunctive acceptance condition to extract them. Evaluated on real Vietnamese electronic medical records, it achieves a precision of about 74% and recall of about 92% and is proved stably effective with small datasets. It outperforms the previous works in the same category of not using annotated corpora and external data resources. Our method and empirical evaluation analysis can lay a foundation for further research and development in Vietnamese medical term discovery and extraction.
Article outline
- 1.Introduction
- 2.Related works
- 2.1Linguistics-based
- 2.2Statistics-based
- 2.3Machine learning-based
- 2.4Hybrid
- 3.The proposed method
- 3.1Method overview
- 3.2Preprocessing
- 3.3Linguistics-based candidate term extraction
- Part-of-Speech tagging
- Open pattern-based term extraction
- PMI-based nested term extraction
- Stop word-based filtering
- 3.4Statistics-based term ranking
- 4.Empirical evaluation
- 4.1Data descriptions
- 4.2Experiment settings and results
- Self-Evaluation
- Comparative evaluation
- 5.Conclusions
-
References
https://doi.org/10.1075/term.20037.vo