Extracting and summarizing the frequent emerging graph patterns from a dataset of graphs

  • Authors:
  • Guillaume Poezevara;Bertrand Cuissart;Bruno Crémilleux

  • Affiliations:
  • Laboratoire GREYC-CNRS UMR 6072, Université de Caen Basse-Normandie, Caen, France;Laboratoire GREYC-CNRS UMR 6072, Université de Caen Basse-Normandie, Caen, France;Laboratoire GREYC-CNRS UMR 6072, Université de Caen Basse-Normandie, Caen, France

  • Venue:
  • Journal of Intelligent Information Systems
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Emerging patterns are patterns of great interest for discovering information from data and characterizing classes. Mining emerging patterns remains a challenge, especially with graph data. In this paper, we propose a method to mine the whole set of frequent emerging graph patterns, given a frequency threshold and an emergence threshold. Our results are achieved thanks to a change of the description of the initial problem so that we are able to design a process combining efficient algorithmic and data mining methods. Moreover, we show that the closed graph patterns are a condensed representation of the frequent emerging graph patterns and we propose a new condensed representation based on the representative pruned graph patterns: by providing shorter patterns, it is especially dedicated to represent a set of graph patterns. Experiments on a real-world database composed of chemicals show the feasibility and the efficiency of our approach.