Exploring Diachronic Changes of Biomedical Knowledge using Distributed Concept Representations
Published in 18th ACL Workshop on Biomedical Natural Language Processing (BioNLP 2019), Florence, Italy, 2019
Vashisth, G., Mikhailov, M., Voigt-Antons, J.-N. & Roller, R.
We train time-sliced word embeddings on biomedical corpora to track how concepts evolve across years. By aligning vector spaces, we quantify semantic drift and highlight case studies where terminology and associations shift (e.g., emerging therapies, reclassified diseases). The approach supports literature monitoring and hypothesis generation in fast-moving domains.
Recommended citation: Vashisth, G., Mikhailov, M., Voigt-Antons, J.-N. & Roller, R. (2019, August). Exploring Diachronic Changes of Biomedical Knowledge using Distributed Concept Representations. Paper presented at the 18th ACL Workshop on Biomedical Natural Language Processing (BioNLP 2019), Florence, Italy. https://doi.org/10.18653/v1/W19-5037