SIDE: Representation Learning in Signed Directed Networks

Given a signed directed network, how can we learn node representations which fully encode structural information of the network including sign and direction of edges? Node representation learning or network embedding learns a mapping of each node to a vector. The mapping encodes structural information on network, providing low-dimensional dense node features for general machine learning and data mining frameworks. Since many social networks allow trust (friend) and distrust (enemy) relationships described by signed and directed edges, generalizing network embedding method to learn from sign and direction information in networks is crucial. In addition, social theories are critical tool in signed network analysis. However, none of the existing methods supports all of the desired properties: considering sign, direction, and social theoretical interpretation. In this paper, we propose SIDE, a general network embedding method that represents both sign and direction of edges in the embedding space. SIDE carefully formulates and optimizes likelihood over both direct and indirect signed connections. We provide socio-psychological interpretation for each component of likelihood function. We prove linear scalability of our algorithm and propose additional optimization techniques to reduce the training time and improve accuracy. Through extensive experiments on real-world signed directed networks, we show that SIDE effectively encodes structural information into the learned embedding.

[1]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[2]  Steven Skiena,et al.  DeepWalk: online learning of social representations , 2014, KDD.

[3]  Jure Leskovec,et al.  Predicting positive and negative links in online social networks , 2010, WWW '10.

[4]  Minji Yoon,et al.  Fast and Accurate Random Walk with Restart on Dynamic Graphs with Guarantees , 2017, WWW.

[5]  Lise Getoor,et al.  Collective Classification in Network Data , 2008, AI Mag..

[6]  Ruslan Salakhutdinov,et al.  Revisiting Semi-Supervised Learning with Graph Embeddings , 2016, ICML.

[7]  M. McPherson,et al.  Birds of a Feather: Homophily in Social Networks , 2001 .

[8]  Charu C. Aggarwal,et al.  Node Classification in Signed Social Networks , 2016, SDM.

[9]  Lee Sael,et al.  BEAR: Block Elimination Approach for Random Walk with Restart on Large Graphs , 2015, SIGMOD Conference.

[10]  Michael Mitzenmacher,et al.  A Brief History of Generative Models for Power Law and Lognormal Distributions , 2004, Internet Math..

[11]  Scott A. Golder,et al.  Structural Predictors of Tie Formation in Twitter: Transitivity and Mutuality , 2010, 2010 IEEE Second International Conference on Social Computing.

[12]  Lise Getoor,et al.  Collective Classi!cation in Network Data , 2008 .

[13]  Enhong Chen,et al.  Learning Deep Representations for Graph Clustering , 2014, AAAI.

[14]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[15]  Charu C. Aggarwal,et al.  Signed Network Embedding in Social Media , 2017, SDM.

[16]  K. E. Read,et al.  Cultures of the Central Highlands, New Guinea , 1954, Southwestern Journal of Anthropology.

[17]  Lee Sael,et al.  Personalized Ranking in Signed Networks Using Signed Random Walk with Restart , 2016, 2016 IEEE 16th International Conference on Data Mining (ICDM).

[18]  Minji Yoon,et al.  TPA: Fast, Scalable, and Accurate Method for Approximate Random Walk with Restart on Billion Scale Graphs , 2017, 2018 IEEE 34th International Conference on Data Engineering (ICDE).

[19]  David B. Skillicorn,et al.  Spectral Embedding of Signed Networks , 2015, SDM.

[20]  Jure Leskovec,et al.  node2vec: Scalable Feature Learning for Networks , 2016, KDD.

[21]  Qiongkai Xu,et al.  GraRep: Learning Graph Representations with Global Structural Information , 2015, CIKM.

[22]  Wenwu Zhu,et al.  Structural Deep Network Embedding , 2016, KDD.

[23]  Mingzhe Wang,et al.  LINE: Large-scale Information Network Embedding , 2015, WWW.

[24]  Xuanjing Huang,et al.  Incorporate Group Information to Enhance Network Embedding , 2016, CIKM.

[25]  Jinhong Jung,et al.  A comparative study of matrix factorization and random walk with restart in recommender systems , 2017, 2017 IEEE International Conference on Big Data (Big Data).

[26]  Inderjit S. Dhillon,et al.  Low rank modeling of signed networks , 2012, KDD.

[27]  Lee Sael,et al.  BePI: Fast and Memory-Efficient Method for Billion-Scale Random Walk with Restart , 2017, SIGMOD Conference.

[28]  U. Feige,et al.  Spectral Graph Theory , 2015 .

[29]  Jian Pei,et al.  Asymmetric Transitivity Preserving Graph Embedding , 2016, KDD.

[30]  Yang Xiang,et al.  SNE: Signed Network Embedding , 2017, PAKDD.

[31]  Padhraic Smyth,et al.  A Spectral Clustering Approach To Finding Communities in Graph , 2005, SDM.

[32]  Xiao Huang,et al.  Label Informed Attributed Network Embedding , 2017, WSDM.

[33]  Wei Lu,et al.  Deep Neural Networks for Learning Graph Representations , 2016, AAAI.

[34]  F. Harary,et al.  STRUCTURAL BALANCE: A GENERALIZATION OF HEIDER'S THEORY1 , 1977 .

[35]  Zhiyuan Liu,et al.  Comprehend DeepWalk as Matrix Factorization , 2015, ArXiv.

[36]  David Liben-Nowell,et al.  The link-prediction problem for social networks , 2007 .

[37]  Lee Sael,et al.  Random Walk with Restart on Large Graphs Using Block Elimination , 2016, ACM Trans. Database Syst..

[38]  Sahin Albayrak,et al.  Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization , 2010, SDM.

[39]  Jérôme Kunegis,et al.  What is the added value of negative links in online social networks? , 2013, WWW.