Recent evolution in computer science and web technology result in continually expanding resources on the web that pose unprecedented challenges in Big data to access it with the existing framework. In order to access the vastly increasing resource more effectively, clustering methods are required. Resource Description Framework (RDF) which is used to describe the resources on the web. Results suggest that linking resources in a semantic manner enhance searching capabilities over the web such that using keywords as linking information can be used to calculate the semantic similarity. Consequently, the traditional retrieval system often suffers from scalability and inefficiency problems when processing or analyzing such large-scale data. Hence, the basic idea is to propose a framework that retrieves the information effectively from the Resource Description Framework (RDF) using SPARQL ENGINE. Furthermore, we purpose pruning method that improves accuracy by quickly filtering out the false alarms (whenever there is a change in the source) and automatically update the RDF document that can be further used for semantically retrieving the root source content of the particular resource. To improve its scalability and efficiency in a big data environment, this framework is implemented on Hadoop, a widely-adopted distributed computing platform.