Ontology learning from relational databases

An ontology is a formal, explicit specification of a shared conceptualization. Ontologies are used in many fields, such as software engineering, information extraction, semantic search, knowledge management, recommender systems, etcetera. Since manual ontology building is a very costly, time-consuming, and error-prone task, automating the process of ontology building, or in other words, learning ontology from existing resources, is a good option. Nowadays, a large amount of data on the web is stored in relational databases, but databases cannot be used directly in the semantic web. Hence, in this paper, we have proposed a new approach to automatically creating an OWL ontology from a relational database. We have defined a set of rules to analyze all database components and convert them to corresponding ontology components. The core contribution of our work is the set of rules which can analyze and extract ontology elements from stored procedures, user-defined functions, views, multiple inheritance, the specific representation of single inheritance, common attributes, and the constraints on tables and their columns. The proposed approach has been compared with existing approaches using three frameworks.