Ontology provides a shared common vocabulary for a specific domain. It is widely used as a formal structured knowledge representation tool for domain knowledge, which can be shared and reused. Ontology development is equivalent to defining a set of data and their structure for creating a variety of semantic web related tasks. The creation and population of ontology is a tedious task, if the knowledge to be encoded is to be extracted manually. This paper proposes an automatic method to populate ontology from unstructured documents by Information Extraction. Various Natural Language Processing techniques, that make use of machine learning models, are used in the realization of the Information Extraction system. It has been observed that, customizing the IE system for the underlying domain ensures improved accuracy and efficiency in information extraction. The proposed work is done in the domain “New Appointments in companies”.

Keywords : Information Extraction; Ontology Population; Natural Language Processing