Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Representation Learning for Classification in Heterogeneous Graphs with Application to Social Networks

Representation Learning for Classification in Heterogeneous Graphs with Application to Social... We address the task of node classification in heterogeneous networks, where the nodes are of different types, each type having its own set of labels, and the relations between nodes may also be of different types. A typical example is provided by social networks where node types may for example be users, content, or films, and relations friendship, like, authorship. Learning and performing inference on such heterogeneous networks is a recent task requiring new models and algorithms. We propose a model, Labeling Heterogeneous Network (LaHNet), a transductive approach to classification that learns to project the different types of nodes into a common latent space. This embedding is learned so as to reflect different characteristics of the problem such as the correlation between node labels, as well as the graph topology. The application focus is on social graphs, but the algorithm is general and can be used for other domains. The model is evaluated on five datasets representative of different instances of social data. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Knowledge Discovery from Data (TKDD) Association for Computing Machinery

Representation Learning for Classification in Heterogeneous Graphs with Application to Social Networks

Download PDF
33 pages

Loading next page...
 
/lp/association-for-computing-machinery/representation-learning-for-classification-in-heterogeneous-graphs-0sLuHm8hkv

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Association for Computing Machinery
Copyright
Copyright © 2018 ACM
ISSN
1556-4681
eISSN
1556-472X
DOI
10.1145/3201603
Publisher site
See Article on Publisher Site

Abstract

We address the task of node classification in heterogeneous networks, where the nodes are of different types, each type having its own set of labels, and the relations between nodes may also be of different types. A typical example is provided by social networks where node types may for example be users, content, or films, and relations friendship, like, authorship. Learning and performing inference on such heterogeneous networks is a recent task requiring new models and algorithms. We propose a model, Labeling Heterogeneous Network (LaHNet), a transductive approach to classification that learns to project the different types of nodes into a common latent space. This embedding is learned so as to reflect different characteristics of the problem such as the correlation between node labels, as well as the graph topology. The application focus is on social graphs, but the algorithm is general and can be used for other domains. The model is evaluated on five datasets representative of different instances of social data.

Journal

ACM Transactions on Knowledge Discovery from Data (TKDD) – Association for Computing Machinery

Published: Jul 20, 2018

Keywords: Representation learning

References

  • WITCH: A new approach to web spam detection
    Abernethy, Jacob; Chapelle, Olivier; Castillo, Carlos
  • Graffiti: Graph-based classification in heterogeneous networks
    Angelova, Ralitsa; Kasneci, Gjergji; Weikum, Gerhard
  • Graph-based text classification: Learn from your neighbors
    Angelova, Ralitsa; Weikum, Gerhard
  • Continuous optimization of hyper-parameters
    Bengio, Yoshua
  • Representation learning: A review and new perspectives
    Bengio, Yoshua; Courville, Aaron; Vincent, Pierre
  • Node classification in social networks
    Bhagat, Smriti; Cormode, Graham; Muthukrishnan, S.
  • Sparse local embeddings for extreme multi-label classification
    Bhatia, Kush; Jain, Himanshu; Kar, Purushottam; Varma, Manik; Jain, Prateek
  • Learning social network embeddings for predicting information diffusion
    Bourigault, Simon; Lagnier, Cedric; Lamprier, Sylvain; Denoyer, Ludovic; Gallinari, Patrick
  • GraRep: Learning graph representations with global structural information
    Cao, Shaosheng; Lu, Wei; Xu, Qiongkai
  • Deep computational phenotyping
    Che, Zhengping; Kale, David; Li, Wenzhe; Bahadori, Mohammad Taha; Liu, Yan
  • Multi-relational link prediction in heterogeneous information networks
    Davis, Darcy; Lichtenwalter, Ryan; Chawla, N. V.
  • Random walks based modularity: Application to semi-supervised learning
    Devooght, Robin; Mantrach, Amin; Kivimäki, Ilkka; Bersini, Hugues; Jaimes, Alejandro; Saerens, Marco
  • Recurrent collective classification
    Fan, Shuangfei; Huang, Bert
  • Using ghost edges for classification in sparsely labeled networks
    Gallagher, Brian; Tong, Hanghang; Eliassi-Rad, Tina; Faloutsos, Christos
  • Introduction to Statistical Relational Learning
    Getoor, Lise
  • Node2Vec: Scalable feature learning for networks
    Grover, Aditya; Leskovec, Jure
  • A heterogeneous label propagation algorithm for disease gene discovery
    Hwang, Taehyun; Kuang, Rui
  • Learning latent representations of nodes for classifying in heterogeneous social networks
    Jacob, Yann; Denoyer, Ludovic; Gallinari, Patrick
  • Graph regularized transductive classification on heterogeneous information networks
    Ji, Ming; Sun, Yizhou; Danilevsky, Marina; Han, Jiawei; Gao, Jing
  • Influence learning for cascade diffusion models: Focus on partial orders of infections
    Lamprier, Sylvain; Bourigault, Simon; Gallinari, Patrick
  • Mining topic-level influence in heterogeneous networks
    Liu, Lu; Tang, Jie; Han, Jiawei; Jiang, Meng
  • Link-based classification
    Lu, Qing; Getoor, Lise
  • Scalable gradient-based tuning of continuous regularization hyperparameters
    Luketina, Jelena; Raiko, Tapani; Berglund, Mathias; Greff, Klaus
  • Generalized expectation criteria for semi-supervised learning with weakly labeled data
    Mann, Gideon S.; McCallum, Andrew
  • Semi-supervised collective classification via hybrid label regularization
    McDowell, Luke; Aha, David
  • Distributed representations of words and phrases and their compositionality
    Mikolov, Tomas; Sutskever, Ilya; Chen, Kai; Corrado, Greg S.; Dean, Jeff
  • Deep collective inference
    Moore, John; Neville, Jennifer
  • A practical hypertext categorization method using links and incrementally available class information
    Myaeng, Sung Hyon; Lee, Mann-ho
  • Structural neighborhood based classification of nodes in a network
    Nandanwar, Sharad; Murty, M. N.
  • Iterative classification in relational data
    Neville, Jennifer; Jensen, David
  • Topical link analysis for web search
    Nie, Lan; Davison, Brian D.; Qi, Xiaoguang
  • Learning convolutional neural networks for graphs
    Niepert, Mathias; Ahmed, Mohamed; Kutzkov, Konstantin
  • Deepwalk: Online learning of social representations
    Perozzi, Bryan; Al-Rfou, Rami; Skiena, Steven
  • Iterative annotation of multi-relational social networks
    Peters, Stephane; Denoyer, Ludovic; Gallinari, Patrick
  • Overcoming relational learning biases to accurately predict preferences in large scale networks
    III, Joseph J. Pfeiffer; Neville, Jennifer; Bennett, Paul N.
  • Column networks for collective classification
    Pham, Trang; Tran, Truyen; Phung, Dinh; Venkatesh, Svetha
  • Learning to propagate rare labels
    Pimplikar, Rakesh; Garg, Dinesh; Bharani, Deepesh; Parija, Gyana
  • Collective classification in network data
    Sen, Prithviraj; Namata, Galileo; Bilgic, Mustafa; Getoor, Lise; Galligher, Brian; Eliassi-Rad, Tina
  • Ranking-based clustering of heterogeneous information networks with star network schema
    Sun, Yizhou; Yu, Yintao; Han, Jiawei
  • PTE: Predictive text embedding through large-scale heterogeneous text networks
    Tang, Jian; Qu, Meng; Mei, Qiaozhu
  • LINE: Large-scale information network embedding
    Tang, Jian; Qu, Meng; Wang, Mingzhe; Zhang, Ming; Yan, Jun; Mei, Qiaozhu
  • Max-margin DeepWalk: Discriminative learning of network representation
    Tu, Cunchao; Zhang, Weicheng; Liu, Zhiyuan; Sun, Maosong
  • Structural deep network embedding
    Wang, Daixin; Cui, Peng; Zhu, Wenwu
  • Graph transduction via alternating minimization
    Wang, Jun; Jebara, Tony; Chang, Shih-Fu
  • Multi-label relational neighbor classification using social context features
    Wang, Xi; Sukthankar, Gita
  • Modeling spatial-temporal clues in a hybrid deep learning framework for video classification
    Wu, Zuxuan; Wang, Xi; Jiang, Yu-Gang; Ye, Hao; Xue, Xiangyang
  • Revisiting semi-supervised learning with graph embeddings
    Yang, Zhilin; Cohen, William; Salakhutdinov, Ruslan
  • Robust semi-supervised learning on multiple networks with noise
    Ye, Junting; Akoglu, Leman
  • Learning with local and global consistency
    Zhou, Dengyong; Bousquet, Olivier; Lal, Thomas Navin; Weston, Jason; Schölkopf, Bernhard
  • Learning from labeled and unlabeled data on a directed graph
    Zhou, Dengyong; Huang, Jiayuan; Schölkopf, Bernhard
  • Deep sparse coding network for image classification
    Zhou, Sijun; Zhang, Shizhou; Wang, Jinjun
  • Activity-edge centric multi-label classification for mining heterogeneous information networks
    Zhou, Yang; Liu, Ling
  • Learning from Labeled and Unlabeled Data with Label Propagation
    Zhu, Xiaojin; Ghahramani, Zoubin