Proceedings of the 7th World Conference on Social Sciences and Humanities
Year: 2024
DOI:
[PDF]
A Contrastive Study Between the Terminology Extraction Method Based on Dependency Parsing and Two Traditional Methods
Yunhan Hao
ABSTRACT:
English maritime terminology represents structured knowledge units containing core information within English maritime texts. Effective terminology extraction is pivotal in the analysis of such texts. Previous research has demonstrated success in extracting terminology from Chinese non-literary texts and research topics from English research papers using dependency parsing. This paper introduces, for the first time, a terminology extraction method based on dependency parsing, and conducts an empirical study by using an English maritime text to compare this method with two traditional methods, namely the n-gram pattern method and the keyness analysis method. The results show that the three methods are complementary, and that the method based on dependency parsing yields the best performance in terms of efficiency, coverage, and accuracy of terminology extraction. For non-specialist terminology developers and users, combining dependency parsing with keyness analysis proves to be a reliable strategy. These findings offer valuable insights for optimizing terminology extraction software and advancing terminology extraction methodologies.
keywords: dependency parsing, terminology extraction, English maritime texts, Maritime English, terminology