Distance Patterns in Structural Similarity
The Journal of Machine Learning Research
The history of histograms (abridged)
VLDB '03 Proceedings of the 29th international conference on Very large data bases - Volume 29
Graph Matching Based on Node Signatures
GbRPR '09 Proceedings of the 7th IAPR-TC-15 International Workshop on Graph-Based Representations in Pattern Recognition
A graph matching method and a graph matching distance based on subgraph assignments
Pattern Recognition Letters
Kernel functions for case-based planning
Artificial Intelligence
Comparing graph similarity measures for graphical recognition
GREC'09 Proceedings of the 8th international conference on Graphics recognition: achievements, challenges, and evolution
Towards performance evaluation of graph-based representation
GbRPR'11 Proceedings of the 8th international conference on Graph-based representations in pattern recognition
Skeleton comparisons: the junction neighbourhood histogram
Proceedings of the 11th ACM symposium on Document engineering
Simulation-Based graph similarity
TACAS'06 Proceedings of the 12th international conference on Tools and Algorithms for the Construction and Analysis of Systems
Querying business process model repositories
World Wide Web
| Hi-index | 0.00 |
Objects like road networks, CAD/CAM components, electrical or electronic circuits, molecules, can be represented as graphs, in many modern applications. In this paper, we propose an efficient and effective graph manipulation technique that can be used in graph-based similarity search. Given a query graph G(V,E), we would like to determine fast which are the graphs in the database that are similar to G(V,E), with respect to a similarity measure. First, we study the similarity measure between two graphs. Then, we discuss graph representation techniques by means of multidimensional vectors. It is shown that no false dismissals are introduced by using the vector representation. Finally we illustrate some representative queries that can be handled by our approach, and present experimental results, based on the proposed graph similarity algorithm. The results show that considerable savings are obtained with respect to computational effort and I/O operations, in comparison to conventional searching techniques.