Learning the Consistent Behavior of Common Users for Target Node Prediction across Social Networks

Shan-Hung Wu, Hao-Heng Chien, Kuan-Hua Lin, Philip Yu
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):298-306, 2014.

Abstract

We study the target node prediction problem: given two social networks, identify those nodes/users from one network (called the source network) who are likely to join another (called the target network, with nodes called target nodes). Although this problem can be solved using existing techniques in the field of cross domain classification, we observe that in many real-world situations the cross-domain classifiers perform sub-optimally due to the heterogeneity between source and target networks that prevents the knowledge from being transferred. In this paper, we propose learning the consistent behavior of common users to help the knowledge transfer. We first present the Consistent Incidence Co-Factorization (CICF) for identifying the consistent users, i.e., common users that behave consistently across networks. Then we introduce the Domain-UnBiased (DUB) classifiers that transfer knowledge only through those consistent users. Extensive experiments are conducted and the results show that our proposal copes with heterogeneity and improves prediction accuracy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-wu14, title = {Learning the Consistent Behavior of Common Users for Target Node Prediction across Social Networks}, author = {Wu, Shan-Hung and Chien, Hao-Heng and Lin, Kuan-Hua and Yu, Philip}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {298--306}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/wu14.pdf}, url = {https://proceedings.mlr.press/v32/wu14.html}, abstract = {We study the target node prediction problem: given two social networks, identify those nodes/users from one network (called the source network) who are likely to join another (called the target network, with nodes called target nodes). Although this problem can be solved using existing techniques in the field of cross domain classification, we observe that in many real-world situations the cross-domain classifiers perform sub-optimally due to the heterogeneity between source and target networks that prevents the knowledge from being transferred. In this paper, we propose learning the consistent behavior of common users to help the knowledge transfer. We first present the Consistent Incidence Co-Factorization (CICF) for identifying the consistent users, i.e., common users that behave consistently across networks. Then we introduce the Domain-UnBiased (DUB) classifiers that transfer knowledge only through those consistent users. Extensive experiments are conducted and the results show that our proposal copes with heterogeneity and improves prediction accuracy.} }
Endnote
%0 Conference Paper %T Learning the Consistent Behavior of Common Users for Target Node Prediction across Social Networks %A Shan-Hung Wu %A Hao-Heng Chien %A Kuan-Hua Lin %A Philip Yu %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-wu14 %I PMLR %P 298--306 %U https://proceedings.mlr.press/v32/wu14.html %V 32 %N 2 %X We study the target node prediction problem: given two social networks, identify those nodes/users from one network (called the source network) who are likely to join another (called the target network, with nodes called target nodes). Although this problem can be solved using existing techniques in the field of cross domain classification, we observe that in many real-world situations the cross-domain classifiers perform sub-optimally due to the heterogeneity between source and target networks that prevents the knowledge from being transferred. In this paper, we propose learning the consistent behavior of common users to help the knowledge transfer. We first present the Consistent Incidence Co-Factorization (CICF) for identifying the consistent users, i.e., common users that behave consistently across networks. Then we introduce the Domain-UnBiased (DUB) classifiers that transfer knowledge only through those consistent users. Extensive experiments are conducted and the results show that our proposal copes with heterogeneity and improves prediction accuracy.
RIS
TY - CPAPER TI - Learning the Consistent Behavior of Common Users for Target Node Prediction across Social Networks AU - Shan-Hung Wu AU - Hao-Heng Chien AU - Kuan-Hua Lin AU - Philip Yu BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-wu14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 298 EP - 306 L1 - http://proceedings.mlr.press/v32/wu14.pdf UR - https://proceedings.mlr.press/v32/wu14.html AB - We study the target node prediction problem: given two social networks, identify those nodes/users from one network (called the source network) who are likely to join another (called the target network, with nodes called target nodes). Although this problem can be solved using existing techniques in the field of cross domain classification, we observe that in many real-world situations the cross-domain classifiers perform sub-optimally due to the heterogeneity between source and target networks that prevents the knowledge from being transferred. In this paper, we propose learning the consistent behavior of common users to help the knowledge transfer. We first present the Consistent Incidence Co-Factorization (CICF) for identifying the consistent users, i.e., common users that behave consistently across networks. Then we introduce the Domain-UnBiased (DUB) classifiers that transfer knowledge only through those consistent users. Extensive experiments are conducted and the results show that our proposal copes with heterogeneity and improves prediction accuracy. ER -
APA
Wu, S., Chien, H., Lin, K. & Yu, P.. (2014). Learning the Consistent Behavior of Common Users for Target Node Prediction across Social Networks. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):298-306 Available from https://proceedings.mlr.press/v32/wu14.html.

Related Material