Inexact Graph Matching Using Eigen-Subspace Projection Clustering

Caelli, Terry; Kosinov, Serhiy

doi:10.1142/s0218001404003186

Cited by 31 publications

(8 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarity graph search aims to find graphs that are similar to the query-graph. Existing techniques mainly fall into three categories, the propagationbased paradigm [2,15,17], whose intuition is that two nodes are similar if their neighbors are similar; the spectral-based paradigm [1,3,4,14,20,21], due to the fact that two graphs are isomorphic if their adjacency matrices have the same eigenvalues and eigenvectors; and optimization-based paradigm [26,27], which transfers graph matching to an optimization problem.…”

Section: Related Workmentioning

confidence: 99%

Scalable supergraph search in large graph databases

Lyu

Lin

et al. 2016

2016 IEEE 32nd International Conference on Data Engineering (ICDE)

View full text Add to dashboard Cite

Supergraph search is a fundamental problem in graph databases that is widely applied in many application scenarios. Given a graph database and a query-graph, supergraph search retrieves all data-graphs contained in the query-graph from the graph database. Most existing solutions for supergraph search follow the pruning-and-verification framework, which prunes false answers based on features in the pruning phase and performs subgraph isomorphism testings on the remaining graphs in the verification phase. However, they are not scalable to handle large-sized data-graphs and query-graphs due to three drawbacks. First, they rely on a frequent subgraph mining algorithm to select features which is expensive and cannot generate large features. Second, they require a costly verification phase. Third, they process features in a fixed order without considering their relationship to the query-graph. In this paper, we address the three drawbacks and propose new indexing and query processing algorithms. In indexing, we select features directly from the data-graphs without expensive frequent subgraph mining. The features form a feature-tree that contains all-sized features and both the cost sharing and pruning power of the features are considered. In query processing, we propose a verification-free algorithm, where the order to process features is query-dependent by considering both the cost sharing and the pruning power. We explore two optimization strategies to further improve the algorithm efficiency. The first strategy applies a lightweight graph compression technique and the second strategy optimizes the inclusion of answers. Finally, we conduct extensive performance studies on two real large datasets to demonstrate the high scalability of our algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Scalable supergraph search in large graph databases

Lyu

Lin

et al. 2016

2016 IEEE 32nd International Conference on Data Engineering (ICDE)

View full text Add to dashboard Cite

show abstract

“…For a more thorough introduction to kernel methods, and particularly to graph kernels, the reader is referred to Chapter 5. [156][157][158][159][160][161][162]. The general idea of this approach is based on the following observation.…”

Section: Graph Edit Distance and Related Approachesmentioning

confidence: 99%

Graph Classification and Clustering Based on Vector Space Embedding

Riesen

Bunke

2010

Series in Machine Perception and Artificial Intelligence

View full text Add to dashboard Cite

I am grateful to Prof. Dr. Horst Bunke for directing me through this work. On countless occasions, his excellent advise and open attitude towards new ideas have been a tremendous support. I would also like to thank Prof. Dr. Xiaoyi Jiang for acting as co-referee of this work and Prof. Dr. Matthias Zwicker for supervising the examination. Furthermore, I am very grateful to Susanne Thüler for her support in all administrative concerns. I would like to acknowledge the funding by the Swiss National Science Foundation (Project 200021-113198/1). Many colleagues have contributed invaluable parts to this work.

show abstract

“…To this end, different graph embedding procedures have been proposed in the literature so far. Some of them [4,25,33,39,42,46] are based on the spectral graph theory. Graphs are converted into a vector representation using some spectral features extracted from the adjacency or the Laplacian matrix of a graph.…”

Section: Introductionmentioning

confidence: 99%