weg2vec: Event embedding for temporal networks

Network embedding techniques are powerful to capture structural regularities in networks and to identify similarities between their local fabrics. However, conventional network embedding models are developed for static structures, commonly consider nodes only and they are seriously challenged when the network is varying in time. Temporal networks may provide an advantage in the description of real systems, but they code more complex information, which could be effectively represented only by a handful of methods so far. Here, we propose a new method of event embedding of temporal networks, called weg2vec , which builds on temporal and structural similarities of events to learn a low dimensional representation of a temporal network. This projection successfully captures latent structures and similarities between events involving different nodes at different times and provides ways to predict the final outcome of spreading processes unfolding on the temporal structure.

[1]  Jari Saramäki,et al.  Temporal Networks , 2011, Encyclopedia of Social Network Analysis and Mining.

[2]  Frank Diederich,et al.  Mathematical Epidemiology Of Infectious Diseases Model Building Analysis And Interpretation , 2016 .

[3]  R. Bro,et al.  A new efficient method for determining the number of components in PARAFAC models , 2003 .

[4]  Andrew Mellor,et al.  The Temporal Event Graph , 2017, J. Complex Networks.

[5]  Ido Guy,et al.  Node Embedding over Temporal Graphs , 2019, IJCAI.

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

[7]  Jari Saramäki,et al.  Mapping temporal-network percolation to weighted, static event graphs , 2017, Scientific Reports.

[8]  Vineeth N. Balasubramanian,et al.  STwalk: learning trajectory representations in temporal graphs , 2017, COMAD/CODS.

[9]  Jari Saramäki,et al.  Small But Slow World: How Network Topology and Burstiness Slow Down Spreading , 2010, Physical review. E, Statistical, nonlinear, and soft matter physics.

[10]  Jure Leskovec,et al.  Learning Dynamic Embeddings from Temporal Interactions , 2018, ArXiv.

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

[12]  Sune Lehmann,et al.  Link communities reveal multiscale complexity in networks , 2009, Nature.

[13]  Renaud Lambiotte,et al.  Diffusion on networked systems is a question of time or structure , 2013, Nature Communications.

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

[15]  Ciro Cattuto,et al.  DyANE: Dynamics-aware node embedding for temporal networks , 2019, ArXiv.

[16]  Yaneer Bar-Yam,et al.  Dynamics Of Complex Systems , 2019 .

[17]  Jari Saramäki,et al.  Temporal motifs in time-dependent networks , 2011, ArXiv.

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

[19]  Jari Saramäki,et al.  From calls to communities: a model for time-varying social networks , 2015, The European Physical Journal B.

[20]  Ciro Cattuto,et al.  Dynamics of Person-to-Person Interactions from Distributed RFID Sensor Networks , 2010, PloS one.

[21]  Taro Takaguchi,et al.  Randomized reference models for temporal networks , 2018, SIAM Rev..

[22]  Naoki Masuda,et al.  Optimal Containment of Epidemics in Temporal and Adaptive Networks , 2016, arXiv.org.

[23]  Naoki Masuda,et al.  A Guide to Temporal Networks , 2016, Series on Complexity Science.

[24]  András A. Benczúr,et al.  Node embeddings in dynamic graphs , 2019, Appl. Netw. Sci..

[25]  Mark E. J. Newman,et al.  The Structure and Function of Complex Networks , 2003, SIAM Rev..

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

[27]  Alexander Grey,et al.  The Mathematical Theory of Infectious Diseases and Its Applications , 1977 .

[28]  Dániel Fogaras,et al.  Scaling link-based similarity search , 2005, WWW '05.

[29]  Jure Leskovec,et al.  Motifs in Temporal Networks , 2016, WSDM.

[30]  R. Lambiotte,et al.  Line graphs, link partitions, and overlapping communities. , 2009, Physical review. E, Statistical, nonlinear, and soft matter physics.

[31]  Martin Rosvall,et al.  Modelling sequences and temporal networks with dynamic community structures , 2015, Nature Communications.

[32]  Albert-László Barabási,et al.  The origin of bursts and heavy tails in human dynamics , 2005, Nature.

[33]  Alessandro Vespignani,et al.  Time varying networks and the weakness of strong ties , 2013, Scientific Reports.

[34]  S. Yakowitz,et al.  An epidemic with individual infectivities and susceptibilities , 2000 .

[35]  András A. Benczúr,et al.  Temporal walk based centrality metric for graph streams , 2018, Applied Network Science.

[36]  N. Ling.  The Mathematical Theory of Infectious Diseases and its applications , 1978 .

[37]  Albert-László Barabási,et al.  Statistical mechanics of complex networks , 2001, ArXiv.

[38]  R. Pastor-Satorras,et al.  Activity driven modeling of time varying networks , 2012, Scientific Reports.

[39]  Ciro Cattuto,et al.  Detecting the Community Structure and Activity Patterns of Temporal Networks: A Non-Negative Tensor Factorization Approach , 2013, PloS one.

[40]  Jure Leskovec,et al.  Higher-order organization of complex networks , 2016, Science.

[41]  Francesco Vaccarino,et al.  Topological Strata of Weighted Complex Networks , 2013, PloS one.

[42]  P. Kaye.  Infectious diseases of humans: Dynamics and control , 1993 .

[43]  Ingo Scholtes,et al.  Causality-driven slow-down and speed-up of diffusion in non-Markovian temporal networks , 2013, Nature Communications.

[44]  Ciro Cattuto,et al.  Activity clocks: spreading dynamics on temporal networks of human contact , 2013, Scientific Reports.