Uncovering the Temporal Dynamics of Diffusion Networks

By Manuel Gomez-Rodriguez, David Balduzzi, Bernhard Schölkopf. 2011

617 citations recorded by the source corpus.

Time plays an essential role in the diffusion of information, influence and disease over networks. In many cases we only observe when a node copies information, makes a decision or becomes infected – but the connectivity, transmission rates between nodes and transmission sources are unknown. Inferring the underlying dynamics is of outstanding interest since it enables forecasting, influencing and retarding infections, broadly construed. To this end, we model diffusion processes as discrete networks of continuous temporal processes occurring at different rates. Given cascade data – observed infection times of nodes – we infer the edges of the global diffusion network and estimate the transmission rates of each edge that best explain the observed data. The optimization problem is convex. The model naturally (without heuristics) imposes sparse solutions and requires no parameter tuning. The problem decouples into a collection of independent smaller problems, thus scaling easily to networks on the order of hundreds of thousands of nodes. Experiments on real and synthetic data show that our algorithm both recovers the edges of diffusion networks and accurately estimates their transmission rates from cascade data.

Subfields: Computational Theory and Mathematics, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Statistical and Nonlinear Physics

Citations in the corpus

Scalable Influence Estimation in Continuous-Time Diffusion Networks (2013, methodology): “On each training set, we learn the continuous-time model using NETRATE [10] with exponential transmission functions.”
Information diffusion in online social networks: a survey (2013, background): “[14] extend NETINF and propose to model the di↵usion process as a spatially discrete network of continuous, conditionally independent temporal processes occurring at di↵erent rates.”
Structure and dynamics of information pathways in online media (2012, methodology): “In contrast with the original model [9], we will later allow transmission rates α j , i to change over time.”
Learning content–social influential features for influence analysis (2016, neutral): “The second is based on network structure learning [9,12,28,32], which infer diffusion network or even estimate diffusion probabilities.”
Information diffusion and external influence in networks (2012, background): “Recently, [13] used hazard functions as a basis for disease propagation in continuous time across social networks.”
Learning Graphs From Data: A Signal Representation Perspective (2018, methodology): “A notable example is the work in [38] where the authors have proposed a generative probabilistic model of diffusion that aims to realistically describe how infections occur over time in a network.”
Learning Networks of Heterogeneous Influence (2012, methodology): “2 Real world dataset Finally, we use the MemeTracker dataset [3] to compare NETINF, NETRATE and KernelCascade.”
Mixture of Mutually Exciting Processes for Viral Diffusion (2013, methodology): “(Gomez-Rodriguez et al., 2011; Gomez-Rodriguez & Schölkopf, 2012) presented probabilistic models for diffusion network based on survival and event history analysis and showed that the problem can be solved via submodular optimization.”
Topological Recurrent Neural Network for Diffusion Prediction (2017, methodology): “There are many extensions of the IC and LT models, such as continuoustime IC [7].”
Modeling Information Propagation with Survival Theory (2013, methodology): “Network inference methods (Gomez-Rodriguez et al., 2011; 2013; Du et al., 2012; Wang et al., 2012) model information propagation using continuous time generative probabilistic models of diffusion.”
Influence Maximization in Continuous Time Diffusion Networks (2012, methodology): “We trace the information in two different ways and then infer two different diffusion networks using NET- RATE (Gomez-Rodriguez et al., 2011).”
Influence Maximization in Continuous Time Diffusion Networks (2012, methodology): “The exponential model is a well-known parametric model for modeling diffusion and influence in social and information networks (Gomez-Rodriguez et al., 2010).”
Quantifying Information Overload in Social Media and Its Impact on Social Contagions (2014, background): “…temporal dynamics of information propagation, and in particular, the speed at which information propagates depends crucially on the time users take to process the information they receive and determine if they would like to forward it to other users (Gomez-Rodriguez, Balduzzi, and Sch¨olkopf 2011).”
Uncover Topic-Sensitive Information Diffusion Networks (2013, methodology): “In both synthetic and real-world data, we show that our method significantly improves over the previous state-of-the-art models in terms of both recovering the hidden diffusion networks and predicting the transmission times of memes.”