Natural language to ontology chart

MEASUR Ontology Charts are used over the past 30+ years to elicit and specify user's requirements in an accurate and precise form. However, these models are hard to develop manually as they require the analyst to go through documentation, usually written in a natural language and identify all the actors, entities and associations of the system. This paper proposes a semi-automatic approach for generating Ontology Charts from requirements documents, present in natural language. The proposed method is utilizing a dictionary which is used to identify words that can potentially participate in the model and a database with existing ontology charts that are used to determine how the different nodes will be connected. The remaining nodes are connected using a simple syntactic approach. Finally, these semi auto-generated units are merged together to form the ontology chart. The paper illustrates the proposed approach with a detailed example.