Analysis of Word Embeddings and Sequence Features for Clinical Information Extraction
Lance De Vine, Mahnoosh Kholghi, Guido Zuccon, Laurianne Sitbon, Anthony N Nguyen · QUT ePrints (Queensland University of Technology) · 2015
This study investigates the use of unsuper-vised features derived from word embed-ding approaches and novel sequence rep-resentation approaches for improving clin-ical information extraction systems. Our results corroborate previous findings that indicate that the use of word embeddings significantly improve the effectiveness of concept extraction models; however, we further determine the influence that the corpora used to generate such features have. We also demonstrate the promise of sequence-based unsupervised features for further improving concept extraction. 1