Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Validation indices for graph clustering
Pattern Recognition Letters - Special issue: Graph-based representations in pattern recognition
Pattern Vectors from Algebraic Graph Theory
IEEE Transactions on Pattern Analysis and Machine Intelligence
Clustering Organisms Using Metabolic Networks
ICCS '08 Proceedings of the 8th international conference on Computational Science, Part II
A generative model for graph matching and embedding
Computer Vision and Image Understanding
Graph characteristics from the heat kernel trace
Pattern Recognition
Graph-Based Representations in Pattern Recognition and Computational Intelligence
IWANN '09 Proceedings of the 10th International Work-Conference on Artificial Neural Networks: Part I: Bio-Inspired Systems: Computational and Ambient Intelligence
Network Science: Complexity in Nature and Technology
Network Science: Complexity in Nature and Technology
High-dimensional spectral feature selection for 3D object recognition based on reeb graphs
SSPR&SPR'10 Proceedings of the 2010 joint IAPR international conference on Structural, syntactic, and statistical pattern recognition
Efficient and robust feature extraction by maximum margin criterion
IEEE Transactions on Neural Networks
Locality-Preserved Maximum Information Projection
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
In this paper we propose graph descriptors derived from B-matrices, which are built on the basis of distances between graph vertices. The B-matrices, being invariant under graph isomorphism, are a rich source of information about graph structure. We explore this representation and propose several new graph characteristics that can be used for efficient graph comparison. Experiments on clusterization and classification with synthetic and real-world data revealed, that new descriptors allow for distinguishing graphs with non trivial structural differences. Moreover, they appear to outperform descriptors based on heat kernel matrix, being at the same time more effective computationally.