Faster subgraph isomorphism detection by well-founded total order indexing

  • Authors:
  • Markus Weber;Marcus Liwicki;Andreas Dengel

  • Affiliations:
  • German Research Center for Artificial Intelligence (DFKI) GmbH, Trippstadter Straíe 122, 67663 Kaiserslautern, Germany;German Research Center for Artificial Intelligence (DFKI) GmbH, Trippstadter Straíe 122, 67663 Kaiserslautern, Germany;German Research Center for Artificial Intelligence (DFKI) GmbH, Trippstadter Straíe 122, 67663 Kaiserslautern, Germany and Knowledge-Based Systems Group, Department of Computer Science, Unive ...

  • Venue:
  • Pattern Recognition Letters
  • Year:
  • 2012

Quantified Score

Hi-index 0.10

Visualization

Abstract

In this paper an extension to index-based subgraph matching is proposed. This extension significantly speeds up the indexing time for graphs where the nodes are labeled with a rather small amount of different classes. Furthermore, the needed storage amount is significantly reduced. In order to reduce the complexity, we introduce a weight function for labeled graphs. Using this weight function, a well-founded total order is defined on the weights of the labels. Inversions which violate the order are not allowed. A computational complexity analysis of the new preprocessing is given and its completeness is proven. Furthermore, in a number of practical experiments with randomly generated graphs the improvement of the new approach is shown. In experiments performed on random sample graphs, and on real-world datasets. The number of permutations for the real-world datasets have been decreased to a fraction of 10^-^5 and 10^-^8 in average compared to the original approach by Messmer. The novel indexing strategy makes indexing of larger graphs feasible, allowing for fast detection of subgraphs.