Towards a framework for graph-based keyword search over relational data

Keyword-based access to structured data has attracted research and industry as a means for facilitating access to information. In recent years, the research community and big data technology vendors put a lot of effort into developing new proof of concept systems for the task at hand. Two major limitations have been identified for such prototypes to transition into fully developed products: 1) systems are not designed to scale up; 2) the absence of a complete evaluation approach oriented towards effectiveness. This work presents a framework for supporting the development and the evaluation of graph-based keyword search systems. Furthermore, the implementation of a core module of this framework is detailed and shared open-source with the community.