Cluster-Based Exploration for Effective Keyword Search over Semantic Datasets

  • Authors:

  • Roberto Virgilio;Paolo Cappellari;Michele Miscione

  • Affiliations:

  • Dipartimento di Informatica e Automazione, Universitá Roma Tre, Rome, Italy;Department of Computing Science, University of Alberta, Canada;Dipartimento di Informatica e Automazione, Universitá Roma Tre, Rome, Italy

  • Venue:
  • ER '09 Proceedings of the 28th International Conference on Conceptual Modeling
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The amount of data available in the Web, in databases as well as other systems, is constantly increasing as increasing is the number of users that wish to access such data. Data is available in forms that may not be of easy access for not expert users. Keyword Search approaches are an effort to abstract from specific data representations, allowing users to retrieve information by providing a few terms of interest. Many solutions build on dedicated indexing techniques as well as search algorithms aiming at finding substructures that connect the data elements matching the keywords. In this paper, we present the development of Yaanii, a tool for effective Keyword Search over semantic datasets. Yaaniiis based on a novel keyword search paradigm for graph-structured data, focusing in particular on the RDF data model. 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 of the clusters, and supports the generation of Top-k solutions during its execution in the first k steps. Experiments demonstrate the effectiveness of our approach.