Mining Semantic Web Based Ontological Data

Ontologies are the structures where linked data is generated and stored. The key concept of our work on this research is to utilize linked data through semantic web. Semantic web makes machines to understand human language with meaning. As semantic web contains metadata which is called annotation, data is labelled with URI/IRI. A triple in RDF (Resource Description Framework) can be utilized for opening fields for AI (Artificial Intelligence). By querying these data, we can short out meaning of their relation between one data to another data. A connection between data to data will bring another data automatically. The study shows a web-based interface where semantic data from DBpedia is queried in “ontobee.org”, multiple ontologies are merged thorough queries in “query.wikidata.org” to find out which medicines and symptoms are related to what kind of disease or vice versa and that resultant merged data are tested for accuracy and association rules. Accuracy and association rules are generated using Decision tree, Apriori, Fp growth algorithm. Queries are done by smart SPARQL query. Mining those resultant merged data are visualized and also generated association rules on WEKA and Rapid minor software. The Protégé tool is used for ontology development. Knowledge acquisition can be adapted for editing models in different Semantic Web languages.