Back to Search
Start Over
An Examination of the Statistical Laws of Semantic Change in Clinical Notes
- Source :
- AMIA Annu Symp Proc
- Publication Year :
- 2021
-
Abstract
- Natural language is continually changing. Given the prevalence of unstructured, free-text clinical notes in the healthcare domain, understanding the aspects of this change is of critical importance to clinical Natural Language Processing (NLP) systems. In this study, we examine two previously described semantic change laws based on word frequency and polysemy, and analyze how they apply to the clinical domain. We also explore a new facet of change: whether domain-specific clinical terms exhibit different change patterns compared to general-purpose English. Using a corpus spanning eighteen years of clinical notes, we find that the previously described laws of semantic change hold for our data set. We also find that domain-specific biomedical terms change faster compared to general English words.
Details
- ISSN :
- 21534063
- Volume :
- 2021
- Database :
- OpenAIRE
- Journal :
- AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science
- Accession number :
- edsair.pmid..........ab6a5bc783dee7fdc0833a6a05c81656