TY - GEN
T1 - FINAL
T2 - 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2016
AU - Zhang, Si
AU - Tong, Hanghang
N1 - Funding Information:
This work is partially supported by the National Science Foundation under Grant No. IIS1017415, by DTRA under the grant number HDTRA1-16-0017, by Army Research Office under the contract number W911NF-16-1-0168, by National Institutes of Health under the grant number R01LM 011986, Region II University Transportation Center under the project number 49997-33 25 and a Baidu gift. We would like to sincerely thank Dr. Jie Tang and Dr. Yutao Zhang for their generosity to share the datasets, and anonymous reviewers for their insightful and constructive comments
Publisher Copyright:
© 2016 ACM.
PY - 2016/8/13
Y1 - 2016/8/13
N2 - Multiple networks naturally appear in numerous high-impact applications. Network alignment (i.e., finding the node correspondence across different networks) is often the very first step for many data mining tasks. Most, if not all, of the existing alignment methods are solely based on the topology of the underlying networks. Nonetheless, many real networks often have rich attribute information on nodes and/or edges. In this paper, we propose a family of algorithms (FINAL) to align attributed networks. The key idea is to leverage the node/edge attribute information to guide (topology-based) alignment process. We formulate this problem from an optimization perspective based on the alignment consistency principle, and develop effective and scalable algorithms to solve it. Our experiments on real networks show that (1) by leveraging the attribute information, our algo- rithms can significantly improve the alignment accuracy (i.e., up to a 30% improvement over the existing methods); (2) compared with the exact solution, our proposed fast alignment algorithm leads to a more than 10 speed-up, while preserving a 95% ac- curacy; and (3) our on-query alignment method scales linearly, with an around 90% ranking accuracy compared with our exact full alignment method and a near real-time response time.
AB - Multiple networks naturally appear in numerous high-impact applications. Network alignment (i.e., finding the node correspondence across different networks) is often the very first step for many data mining tasks. Most, if not all, of the existing alignment methods are solely based on the topology of the underlying networks. Nonetheless, many real networks often have rich attribute information on nodes and/or edges. In this paper, we propose a family of algorithms (FINAL) to align attributed networks. The key idea is to leverage the node/edge attribute information to guide (topology-based) alignment process. We formulate this problem from an optimization perspective based on the alignment consistency principle, and develop effective and scalable algorithms to solve it. Our experiments on real networks show that (1) by leveraging the attribute information, our algo- rithms can significantly improve the alignment accuracy (i.e., up to a 30% improvement over the existing methods); (2) compared with the exact solution, our proposed fast alignment algorithm leads to a more than 10 speed-up, while preserving a 95% ac- curacy; and (3) our on-query alignment method scales linearly, with an around 90% ranking accuracy compared with our exact full alignment method and a near real-time response time.
KW - Alignment consistency
KW - Attributed network alignment
KW - Onquery alignment
UR - http://www.scopus.com/inward/record.url?scp=84984996767&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84984996767&partnerID=8YFLogxK
U2 - 10.1145/2939672.2939766
DO - 10.1145/2939672.2939766
M3 - Conference contribution
AN - SCOPUS:84984996767
T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
SP - 1345
EP - 1354
BT - KDD 2016 - Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PB - Association for Computing Machinery
Y2 - 13 August 2016 through 17 August 2016
ER -