Session

Saturday, June 24, 2017 - 09:00 to 15:45
Extracting disease-phenotype relations from text with disease-phenotype concept recognisers and association rule mining
Abstract: 
Automatically extracting phenotypes (i.e., the composite of one’s observable characteristics/traits) from free text such as scientific literature or clinical notes and associating phenotypes with diseases is an important task. Such associations can be used in, for example, recommending candidate genes for diseases, investigating drug targets, or performing differential diagnosis. In this paper we focus on extracting disease-phenotype relations with association rule mining techniques and compare results with two other methods. We show that association rule mining offers promising alternative method for detecting disease-phenotype relations.
Simon Kocbek's picture
Simon Kocbek
Garvan Institute of Medical Research Sydney (AU)
Tudor Groza's picture
Tudor Groza