Yaanii: Effective Keyword Search over Semantic Dataset.
Roberto De Virgilio, Paolo Cappellari, Michele Miscione · Iris (Roma Tre University) · 2010
Nowadays data is disseminated in a number of different sources, from databases systems to the Web, from a traditional structured organization (relational) to a semi-structured (XML), up to the unstructured ones (text in Web documents). Although availability of data is constantly increasing, one principal difficulty users have to face is to find and retrieve the information they are looking for. To this aim keywords search based systems are increasingly capturing the attention of researchers. In this paper, we present Yaanii1, a tool for the effective Keyword Search over semantic datasets. It is based on a novel keyword search paradigm for graph-structured data, focusing in particular on the RDF data model. While many techniques search the best answer trees, we propose an effective algorithm for the exploration and computation of all matching subgraphs. We provide a clustering technique that identifies and groups graph substructures based on template match. A scoring function, IR inspired, evaluates the relevance of the substructures and the clusters. A strong point of our approach is that the ranking supports the generation of Top-k solutions during its execution.