Extracting relationship associations from semantic graphs in life sciences

Abstract

The rate of literature publication in life sciences is growing fast, and researchers in the bioinformatics and knowledge discovery fields have been studying how to use the existing literature to discover novel knowledge or generate novel hypothesis. Existing literature-based discovery methods and tools use text-mining techniques to extract non-specified relationships between two concepts. This paper presents a new approach to literature-based discovery, which adopts semantic web techniques to measure the relevance between two relationships with specified types that involve a particular entity. We extract pairs of highly relevant relationships, which we call relationship associations, from semantic graphs representing scientific papers. These relationship associations can be used to help researchers generate scientific hypotheses or create their own semantic graphs for their papers. We present the results of experiments for extracting relationship associations from 392 semantic graphs representing MEDLINE papers. © 2011 Springer-Verlag.