Paper Info
Title
Unlocking medical ontologies for non-ontology experts
Abstract
Ontology authoring is a specialised task requiring amongst other things a deep knowledge of the ontology language being used. Understanding and reusing ontologies can thus be difficult for domain experts, who tend not to be ontology experts. To address this problem, we have developed a Natural Language Generation system for transforming the axioms that form the definitions of ontology classes into Natural Language paragraphs. Our method relies on deploying ontology axioms into a top-level Rhetorical Structure Theory schema. Axioms are ordered and structured with specific rhetorical relations under rhetorical structure trees. We describe here an implementation that focuses on a sub-module of SNOMED CT. With some refinements on articles and layout, the resulting paragraphs are fluent and coherent, offering a way for subject specialists to understand an ontology's content without need to understand its logical representation.
Year
Venue
Keywords
2011
BioNLP@ACL
ontology class,ontology axiom,natural language generation system,ontology expert,rhetorical structure tree,ontology authoring,non-ontology expert,ontology language,specific rhetorical relation,reusing ontology,medical ontology,natural language paragraph
Field
DocType
Citations 
Ontology (information science),Ontology-based data integration,Process ontology,Ontology chart,Computer science,Ontology Inference Layer,Natural language processing,Artificial intelligence,Suggested Upper Merged Ontology,Upper ontology,Ontology components
Conference
7
PageRank 
References 
Authors
0.77
10
4
Name
Order
Citations
PageRank
Shao Fen Liang1334.53
Donia Scott263471.58
Robert Stevens35538499.01
Alan Rector41489161.78