Ontology model of dependency graphs

Abstract

On the field of  natural language processing, one of the hardest task is to implement an efficient text to semantic engine. The semantic content is generated  from the free text document using a chain of processing phases.  The first processing steps relate to text preprocessing, morphology and grammar  analysis, while the result is a semantic graph usually stored in ontology format. The paper presents  an analysis about the role of dependency graph in this conversion process.  The  work contains proposals for a semantic model suitable for straightforward  conversion  from  the dependency graph. The implemented prototype system uses RDF-based ontology model to store the final semantic model.