Paper Details

ABSTRACT

Characterization of “Non-ISA” factual sentences can be used in Ontology learning. Characterization identifies subjects, objects and relations between them. It also identifies subject modifiers and object modifiers. Ontology Learning (OL) as a research field has been motivated by the possibility of automated generation of formal knowledge based on top of Natural Language (NL) document content so as to support reasoning based knowledge discovery. Most of the work done in this field has been made in Light-Weight OL, not much attempt has been made in Heavy-Weight OL. Ontology Learning is automated generation of ontologies from documents that contain natural language text. Characterize of “Non-ISA” factual sentences in English involve different stages like Triple Extraction, Normalize, Singularize and Characterization. In this paper we are going to focus on Triple-Extraction part which helps in characterization of sentences. In this module it converts Complex and compound “Non-Isa” into simple sentences. Characterization of Simple sentences are far easier than compound and complex sentences. This characterization can be useful to further convert a “Non-ISA” factual sentence in English into its equivalent Description Logic (DL), which is a part of Heavy weight OL, which makes the information retrieval very effective and reliable.