A Comparative Study of Computational Linguistics Terminology in English Papers by Chinese and American Scholars

This study aims to explore the research hotspots and development trends in computational linguistics through a terminology-based comparison. The specific method involves constructing a corpus of English papers in computational linguistics by Chinese and American scholars from the past decade, then counting the coverage frequency of basic and high-frequency terms in the corpus, and finally conducting a diachronic and synchronic comparative analysis. Our findings are as follows. Over the past decade, there was a decreasing trend in the usage of basic terms among both Chinese and American scholars in computational linguistics. This may be attributed to the continuous production of new terms and the reduction in comprehension difficulty of papers. Nonetheless, certain periods have shown an increase in the usage of specific terms, such as the prevalence of algorithms and models in Chinese scholars’ papers from 2012 to 2013, and American scholars’ focus on text processing, application systems, and linguistic resources from 2014 to 2015. On the other hand, the high-frequency terms used in the past decade collectively reflect the prevailing research trends in computational linguistics: continuous ascent and evolution of deep learning, consistent research interest and ongoing innovation in machine translation, and significant growth in the study of speech and discourse.