Ontology Population for Open-Source Intelligence (Discussion Paper)

We present an approach based on GATE (General Architecture for Text Engineering) for the automatic population of ontologies from text documents. We describe some experimental results, which are encouraging in terms of extracted correct instances of the ontology. We then focus on a phase of our pipeline and discuss a variant thereof, which aims at reducing the manual effort needed to generate pre-defined dictionaries used in document annotation. Our additional experiments show promising results also in this case.