Abstract
Graphs are popularly used to represent objects with dependency structures, yet all existing graph classification algorithms can only handle simple graphs where each node is a single attribute (or a set of independent attributes). In this paper, we formulate a new super-graph classification task where each node of the super-graph may contain a graph (a single-attribute graph), so a super-graph contains a set of inter-connected graphs. To support super-graph classification, we propose a Weighted Random Walk Kernel (WRWK) which generates a product graph between any two super-graphs, and uses the similarity (kernel value) of two single-attribute graph as the node weight. Then we calculate weighted random walks on the product graph to generate kernel value between two super-graphs as their similarity. Our method enjoys sound theoretical properties, including bounded similarity. Experiments confirm that our method significantly outperforms baseline approaches.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Angelova, R., Weikum, G.: Graph-based text classification: learn from your neighbors. In: ACM SIGIR, pp. 485–492 (2006)
Cai, Y., Cercone, N., Han, J.: An attribute-oriented approach for learning classification rules from relational databases. In: ICDE, pp. 281–288 (1990)
Cheng, H., Zhou, Y., Yu, J.X.: Clustering large attributed graphs: A balance between structural and attribute similarities. ACM TKDD 5(2) (2011)
Gärtner, T., Flach, P.A., Wrobel, S.: On graph kernels: Hardness results and efficient alternatives. In: COLT, pp. 129–143 (2003)
Luo, X., Xu, Z., Yu, J.: Building association link network for semantic link on web resources. IEEE Trans. Autom. Sci. Eng. 8(3), 482–494 (2011)
Riesen, K., Bunke, H.: Graph classification and clustering based on vector space embedding. World Scientific Publishing Co., Inc. (2010)
Schölkopf, B., Smola, A.J.: Learning with kernels. MIT Press (2002)
Xu, Z., Ke, Y., Wang, Y., Cheng, H., Cheng, J.: A model-based approach to attributed graph clustering. In: ACM SIGMOD, pp. 505–516 (2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Guo, T., Zhu, X. (2014). Super-Graph Classification. In: Tseng, V.S., Ho, T.B., Zhou, ZH., Chen, A.L.P., Kao, HY. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2014. Lecture Notes in Computer Science(), vol 8443. Springer, Cham. https://doi.org/10.1007/978-3-319-06608-0_27
Download citation
DOI: https://doi.org/10.1007/978-3-319-06608-0_27
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-06607-3
Online ISBN: 978-3-319-06608-0
eBook Packages: Computer ScienceComputer Science (R0)