Knowledge graphs (KGs) are fast becoming the cornerstone of research for storing and retrieving information effectively due to their ability to link heterogeneous data, make inferences, and discover new knowledge without additional human input. Researchers use a plethora of KGs like NELL, DBPedia, YAGO to augment their information extraction activities. However, in such diverse and expressive graphs, the access to knowledge that matches user’s needs is not always obvious— i.e., user intent does not necessarily get translated into the query interpretation which can frustrate the user. In this work, we present a discourse enabled framework on a large scale KGs as means to start an investigation into modeling user intent for query processing. In this regarding we present a data aware query reformulation strategy with a faceted interface to enable discourse that helps users to specify their needs in an intuitive manner.