Comparative Analysis of RDF Query Languages

Abstract

The global rapid development and maturing of Internet prompts the exploding increase of information in each field and provides people with abundant sources for seeking and obtaining useful information. The concept of information retrieval using search methodologies thus becomes the most active part of network informatics. Semantic Search serves as a key to efficient and accurate information retrieval to the users as in today’s human-readable Web the information cannot be easily processed by machine and even highly sophisticated, efficient keyword based Syntactic Search techniques have not been able to provide accurate retrieval. The Semantic processing for data retrieval is accomplished by the use of a similarity measures known as Resource Description Framework (RDF), Ontology Web Language (OWL), etc. that have been layered on the top of Extensible Markup Language (XML). The annotations increased the efficiency of Semantic Search by enabling the metadata to be associated with the resources. The RDF data stored along with the annotations requires some methodologies to query the repository and thus the query languages like SPARQL, SeRQL, SquishQL etc. came into existence. In this thesis, the intent is to generate RDF data model for semantic retrieval of Homonyms by adding annotations and its validation using W3C Validation service. The six prominent query languages, SPARQL, SeRQL, SquishQL, LAQRS, XsRQL and RDQL, have been studied on the basis of some features which are necessary and required for being a robust query language and a comparative analysis has been done using the results obtained for the retrieval of Homonyms.