TiDeH: Time-Dependent Hawkes Process for Predicting Retweet Dynamics

Abstract

Online social networking services allow their users to post content in the form of text, images or videos. The main mechanism driving content diffusion is the possibility for users to re-share the content posted by their social connections, which may then cascade across the system. A fundamental problem when studying information cascades is the possibility to develop sound mathematical models, whose parameters can be calibrated on empirical data, in order to predict the future course of a cascade after a window of observation. In this paper, we focus on Twitter and, in particular, on the temporal patterns of retweet activity for an original tweet. We model the system by Time-Dependent Hawkes process (TiDeH), which properly takes into account the circadian nature of the users and the aging of information. The input of the prediction model are observed retweet times and structural information about the underlying social network. We develop a procedure for parameter optimization and for predicting the future profiles of retweet activity at different time resolutions. We validate our methodology on a large corpus of Twitter data and demonstrate its systematic improvement over existing approaches in all the time regimes.

arXiv commentThe manuscript has been accepted in the 10th International AAAI Conference on Web and Social Media (ICWSM 2016)

Cited by in corpus (29)

• A Survey of Information Cascade Analysis: Models, Predictions, and Recent Advances
• Expecting to be HIP: Hawkes Intensity Processes for Social Media Popularity
• Predictive Analysis on Twitter: Techniques and Applications
• Modelling structure and predicting dynamics of discussion threads in online boards
• Explicit Time Embedding Based Cascade Attention Network for Information Popularity Prediction
• Ratioing the President: An exploration of public engagement with Obama and Trump on Twitter
• Universal Components of Real-world Diffusion Dynamics based on Point Processes
• Capturing Dynamics of Information Diffusion in SNS: A Survey of Methodology and Techniques
• Detecting False Rumors from Retweet Dynamics on Social Media
• Will You Dance To The Challenge? Predicting User Participation of TikTok Challenges
• Approximation of bayesian Hawkes process models with Inlabru
• Stationarity of the inter-event power-law distributions
• Modeling Popularity in Asynchronous Social Media Streams with Recurrent Neural Networks
• Quantifying Uncertainty in a Predictive Model for Popularity Dynamics
• Dynamic Hawkes Processes for Discovering Time-evolving Communities' States behind Diffusion Processes
• Interval-censored Hawkes processes
• Impact of individual actions on the collective response of social systems
• Impact of environmental changes on the dynamics of temporal networks
• Damping effect in innovation processes: case studies from Twitter
• SMP Challenge: An Overview of Social Media Prediction Challenge 2019
• Predicting Relationship Labels and Individual Personality Traits from Telecommunication History in Social Networks using Hawkes Processes
• Marked Self-Exciting Point Process Modelling of Information Diffusion on Twitter
• Hawkes Process Kernel Structure Parametric Search with Renormalization Factors
• Popularity Prediction for Social Media over Arbitrary Time Horizons
• Make Hawkes Processes Explainable by Decomposing Self-Triggering Kernels
• Cannot Predict Comment Volume of a News Article before (a few) Users Read It
• Transform Analysis for Hawkes Processes with Applications in Dark Pool Trading
• Tensor Kernel Recovery for Spatio-Temporal Hawkes Processes
• A Graph Approach to Simulate Twitter Activities with Hawkes Processes