TGNet: Learning to Rank Nodes in Temporal Graphs

Node ranking in temporal networks are often impacted by heterogeneous context from node content, temporal, and structural dimensions. This paper introduces TGNet , a deep learning framework for node ranking in heterogeneous temporal graphs. TGNet utilizes a variant of Recurrent Neural Network to adapt context evolution and extract context features for nodes. It incorporates a novel influence network to dynamically estimate temporal and structural influence among nodes over time. To cope with label sparsity, it integrates graph smoothness constraints as a weak form of supervision. We show that the application of TGNet is feasible for large-scale networks by developing efficient learning and inference algorithms with optimization techniques. Using real-life data, we experimentally verify the effectiveness and efficiency of TGNet techniques. We also show that TGNet yields intuitive explanations for applications such as alert detection and academic impact ranking, as verified by our case study.

[1]  Ronald Rosenfeld,et al.  Semi-supervised learning with graphs , 2005 .

[2]  Yang Song,et al.  An Overview of Microsoft Academic Service (MAS) and Applications , 2015, WWW.

[3]  Michael R. Lyu,et al.  A generalized Co-HITS algorithm and its application to bipartite graphs , 2009, KDD.

[4]  Soumen Chakrabarti,et al.  Learning to rank networked entities , 2006, KDD '06.

[5]  Ryan A. Rossi,et al.  Dynamic PageRank Using Evolving Teleportation , 2012, WAW.

[6]  Kunle Olukotun,et al.  Map-Reduce for Machine Learning on Multicore , 2006, NIPS.

[7]  Ah Chung Tsoi,et al.  Adaptive ranking of web pages , 2003, WWW '03.

[8]  Tie-Yan Liu,et al.  Semi-supervised ranking on very large graphs with rich metadata , 2011, KDD.

[9]  James Caverlee,et al.  PageRank for ranking authors in co-citation networks , 2009, J. Assoc. Inf. Sci. Technol..

[10]  Ali A. Ghorbani,et al.  Toward developing a systematic approach to generate benchmark datasets for intrusion detection , 2012, Comput. Secur..

[11]  Richard S. Zemel,et al.  Gated Graph Sequence Neural Networks , 2015, ICLR.

[12]  Alvaro A. Cárdenas,et al.  Principled reasoning and practical applications of alert fusion in intrusion detection systems , 2008, ASIACCS '08.

[13]  Jure Leskovec,et al.  Supervised random walks: predicting and recommending links in social networks , 2010, WSDM '11.

[14]  Yi Chang,et al.  Positive-Unlabeled Learning in Streaming Networks , 2016, KDD.

[15]  Klaus-Robert Müller,et al.  Efficient BackProp , 2012, Neural Networks: Tricks of the Trade.

[16]  Shou-De Lin,et al.  Unsupervised Ranking using Graph Structures and Node Attributes , 2017, WSDM.

[17]  Leman Akoglu,et al.  Fast Memory-efficient Anomaly Detection in Streaming Heterogeneous Graphs , 2016, KDD.

[18]  Pietro Liò,et al.  Graph Attention Networks , 2017, ICLR.

[19]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[20]  Ramana Rao Kompella,et al.  Network Sampling: From Static to Streaming Graphs , 2012, TKDD.

[21]  Bo Zong,et al.  Towards scalable critical alert mining , 2014, KDD.

[22]  Gao Cong,et al.  Graph-based Point-of-interest Recommendation with Geographical and Temporal Influences , 2014, CIKM.

[23]  Padhraic Smyth,et al.  EventRank: a framework for ranking time-varying networks , 2005, LinkKDD '05.

[24]  Samuel S. Schoenholz,et al.  Neural Message Passing for Quantum Chemistry , 2017, ICML.

[25]  Thorsten Joachims,et al.  Training linear SVMs in linear time , 2006, KDD '06.

[26]  Tie-Yan Liu,et al.  A Ranking Approach on Large-Scale Graph With Multidimensional Heterogeneous Information , 2016, IEEE Transactions on Cybernetics.

[27]  Deepak S. Turaga,et al.  Consensus extraction from heterogeneous detectors to improve performance over network traffic anomaly detection , 2011, 2011 Proceedings IEEE INFOCOM.

[28]  David Yarowsky,et al.  Ranking and Semi-supervised Classification on Large Scale Graphs Using Map-Reduce , 2009, Graph-based Methods for Natural Language Processing.

[29]  Wei Cheng,et al.  Ranking Causal Anomalies via Temporal and Dynamical Analysis on Vanishing Correlations , 2016, KDD.

[30]  Hongyang Zhang,et al.  Approximate Personalized PageRank on Dynamic Graphs , 2016, KDD.

[31]  Hai Jin,et al.  Temporal-Based Ranking in Heterogeneous Networks , 2014, NPC.

[32]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[33]  Tie-Yan Liu,et al.  Learning to rank for information retrieval , 2009, SIGIR.

[34]  Johannes Gehrke,et al.  Edge-Weighted Personalized PageRank: Breaking A Decade-Old Performance Barrier , 2015, KDD.

[35]  Xavier Bresson,et al.  Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering , 2016, NIPS.

[36]  Shivani Agarwal,et al.  Ranking on graph data , 2006, ICML.

[37]  Kian-Lee Tan,et al.  Parallel Personalized Pagerank on Dynamic Graphs , 2017, Proc. VLDB Endow..

[38]  Jure Leskovec,et al.  The dynamics of viral marketing , 2005, EC '06.

[39]  Serge Abiteboul,et al.  Adaptive on-line page importance computation , 2003, WWW '03.

[40]  Soumen Chakrabarti,et al.  Learning random walks to rank nodes in graphs , 2007, ICML '07.