A Study of Word Embeddings for Biomedical Question Answering

Sanjay Kamath (1, 2) , Brigitte Grau (1) , Yue Ma (2)

1 LIMSI - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur
2 LaHDAK - LRI - Données et Connaissances Massives et Hétérogènes (LRI)

Different word embeddings for question answering task can give different impacts on the performance. This article discusses the issue, compares different word embeddings on a question answering task for biomedical domain and reports better performing word embeddings.

Domains: Computer Science [cs]; Computer Science [cs]; Computation and Language [cs.CL]; Computer Science [cs]; Machine Learning [cs.LG]; Computer Science [cs]; Artificial Intelligence [cs.AI]

Sanjay Kamath, Brigitte Grau, Yue Ma. A Study of Word Embeddings for Biomedical Question Answering. 4e édition du Symposium sur l'Ingénierie de l'Information Médicale, Nov 2017, Toulouse, France. ⟨hal-01704570⟩