Home // DBKDA 2019, The Eleventh International Conference on Advances in Databases, Knowledge, and Data Applications // View article

Graph Learning for Prediction of Drug-Disease Interactions: Preliminary Results

Authors:
Andrej Kastrin
Dimitar Hristovski

Keywords: Complex networks; Network analysis; Network learn- ing; MEDLINE

Abstract:
One of the fundamental problems to complex network research is understanding of link formation. We study the problem of representation learning in a bipartite drug-disease network of semantic predications extracted from biomedical literature. We employ DeepWalk and node2vec node embedding methods with deep learning link predictor, as well as standard baseline predictors including common neighbors, Jaccard coefficient, and Adamic/Adar. Experimental results show that both network embedding algorithms outperform traditional link predictors.

Pages: 28 to 30

Copyright: Copyright (c) IARIA, 2019

Publication date: June 2, 2019

Published in: conference

ISSN: 2308-4332

ISBN: 978-1-61208-715-3

Location: Athens, Greece

Dates: from June 2, 2019 to June 6, 2019