SSSNET: Semi-Supervised Signed Network Clustering

Node embeddings are a powerful tool in the analysis of networks; yet, their full potential for the important task of node clustering has not been fully exploited. In particular, most state-of-the-art methods generating node embeddings of signed networks focus on link sign prediction, and those that pertain to node clustering are usually not graph neural network (GNN) methods. Here, we introduce a novel probabilistic balanced normalized cut loss for training nodes in a GNN framework for semi-supervised signed network clustering, called SSSNET. The method is end-to-end in combining embedding generation and clustering without an intermediate step; it has node clustering as main focus, with an emphasis on polarization effects arising in networks. The main novelty of our approach is a new take on the role of social balance theory for signed network embeddings. The standard heuristic for justifying the criteria for the embeddings hinges on the assumption that an “enemy’s enemy is a friend”. Here, instead, a neutral stance is assumed on whether or not the enemy of an enemy is a friend. Experimental results on various data sets, including a synthetic signed stochastic block model, a polarized version of it, and real-world data at different scales, demonstrate that SSSNET can achieve comparable or better results than state-of-the-art spectral clustering methods, for a wide range of noise and sparsity levels. SSSNET complements existing methods through the possibility of including exogenous information, in the form of node-level features or labels.

[1]  Huan Liu,et al.  "Bridge": Enhanced Signed Directed Network Embedding , 2018, CIKM.

[2]  Sami Abu-El-Haija,et al.  Graph Traversal with Tensor Functionals: A Meta-Algorithm for Scalable Learning , 2021, ICLR.

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

[4]  Huan Liu,et al.  Is distrust the negation of trust?: the value of distrust in social media , 2014, HT.

[5]  Yanhui Hu,et al.  Integrating protein-protein interaction networks with phenotypes reveals signs of interactions , 2013, Nature Methods.

[6]  Yiqi Chen,et al.  BASSI: Balance and Status Combined Signed Network Embedding , 2018, DASFAA.

[7]  Inderjit S. Dhillon,et al.  Clustering to forecast sparse time-series data , 2015, 2015 IEEE 31st International Conference on Data Engineering.

[8]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[9]  Aristides Gionis,et al.  Discovering conflicting groups in signed networks , 2020, NeurIPS.

[10]  Xueqi Cheng,et al.  SDGNN: Learning Node Representation for Signed Directed Networks , 2021, AAAI.

[11]  Xueqi Cheng,et al.  Signed Graph Attention Networks , 2019, ICANN.

[12]  Katarzyna Musial,et al.  Semi-supervised stochastic blockmodel for structure analysis of signed networks , 2020, Knowl. Based Syst..

[13]  Aristides Gionis,et al.  Searching for polarization in signed graphs: a local spectral approach , 2020, WWW.

[14]  Mihai Cucuringu,et al.  An MBO scheme for clustering and semi-supervised clustering of signed networks , 2019, Communications in Mathematical Sciences.

[15]  Yanfeng Wang,et al.  Decoupled Variational Embedding for Signed Directed Networks , 2020, ACM Trans. Web.

[16]  Yu Li,et al.  Learning Signed Network Embedding via Graph Attention , 2020, AAAI.

[17]  Adam P. Harrison,et al.  High Performance Rearrangement and Multiplication Routines for Sparse Tensor Arithmetic , 2018, SIAM J. Sci. Comput..

[18]  Jure Leskovec,et al.  Exploiting Social Network Structure for Person-to-Person Sentiment Analysis , 2014, TACL.

[19]  Gesine Reinert,et al.  Efficient method for estimating the number of communities in a network , 2017, Physical review. E.

[20]  Jure Leskovec,et al.  GNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings , 2021, ICML.

[21]  Aymen Ben Rejeb,et al.  Oil, Gold, US dollar and Stock market interdependencies: A global analytical insight , 2017 .

[22]  Dimitris N. Metaxas,et al.  Rethinking Kernel Methods for Node Representation Learning on Graphs , 2019, NeurIPS.

[23]  Gary L. Miller,et al.  Scalable Constrained Clustering: A Generalized Spectral Method , 2016, ArXiv.

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

[25]  Bivas Mitra,et al.  On the Network Embedding in Sparse Signed Networks , 2019, PAKDD.

[26]  F. Harary On the notion of balance of a signed graph. , 1953 .

[27]  Peter Davies,et al.  SPONGE: A generalized eigenproblem for clustering signed networks , 2019, AISTATS.

[28]  Yaoping Hou,et al.  On the Laplacian Eigenvalues of Signed Graphs , 2003 .

[29]  Aristides Gionis,et al.  Discovering Polarized Communities in Signed Networks , 2019, CIKM.

[30]  Amitabha Bagchi,et al.  Balance Maximization in Signed Networks via Edge Deletions , 2020, ArXiv.

[31]  Jiliang Tang,et al.  Signed Graph Convolutional Networks , 2018, 2018 IEEE International Conference on Data Mining (ICDM).

[32]  M. Cugmas,et al.  On comparing partitions , 2015 .

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

[34]  Peter J. Bickel,et al.  Pseudo-likelihood methods for community detection in large sparse networks , 2012, 1207.2340.

[35]  Clara Pizzuti,et al.  Is normalized mutual information a fair measure for comparing community detection methods? , 2015, 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM).

[36]  Jure Leskovec,et al.  Signed networks in social media , 2010, CHI.

[37]  Matthias Hein,et al.  Clustering Signed Networks with the Geometric Mean of Laplacians , 2016, NIPS.

[38]  Riko Jacob,et al.  The I/O Complexity of Sparse Matrix Dense Matrix Multiplication , 2010, LATIN.

[39]  James Bailey,et al.  Standardized Mutual Information for Clustering Comparisons: One Step Further in Adjustment for Chance , 2014, ICML.

[40]  Quan Z. Sheng,et al.  A Comprehensive Survey on Community Detection With Deep Learning , 2021, IEEE Transactions on Neural Networks and Learning Systems.

[41]  Yiwen Zhang,et al.  MUSE: Multi-faceted Attention for Signed Network Embedding , 2021, ArXiv.

[42]  Inderjit S. Dhillon,et al.  Scalable clustering of signed networks using balance normalized cut , 2012, CIKM.

[43]  Charu C. Aggarwal,et al.  Attributed Signed Network Embedding , 2017, CIKM.

[44]  John W. Lau,et al.  Climate inference on daily rainfall across the Australian continent, 1876–2015 , 2019, The Annals of Applied Statistics.

[45]  Kamal Kant Bharadwaj,et al.  Predicting Friends and Foes in Signed Networks Using Inductive Inference and Social Balance Theory , 2012, 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining.

[46]  Kevin Chen-Chuan Chang,et al.  ROSE: Role-based Signed Network Embedding , 2020, WWW.

[47]  Fan Chung Graham,et al.  Spectral Clustering of Graphs with General Degrees in the Extended Planted Partition Model , 2012, COLT.

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

[49]  Hemant Tyagi,et al.  Regularized spectral methods for clustering signed networks , 2020, J. Mach. Learn. Res..

[50]  Kyungsik Han,et al.  ASiNE: Adversarial Signed Network Embedding , 2020, SIGIR.

[51]  Ying Wah Teh,et al.  Time-series clustering - A decade review , 2015, Inf. Syst..

[52]  Ramanathan V. Guha,et al.  Propagation of trust and distrust , 2004, WWW '04.

[53]  Jure Leskovec,et al.  Inductive Representation Learning on Large Graphs , 2017, NIPS.

[54]  Charu C. Aggarwal,et al.  Recommendations in Signed Social Networks , 2016, WWW.

[55]  Bo Du,et al.  Social Trust Network Embedding , 2019, 2019 IEEE International Conference on Data Mining (ICDM).