Anytime Information Cascade Popularity Prediction via Self-Exciting Processes

Xi Zhang, Akshay Aravamudan, Georgios C Anagnostopoulos
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:26028-26047, 2022.

Abstract

One important aspect of understanding behaviors of information cascades is to be able to accurately predict their popularity, that is, their message counts at any future time. Self-exciting Hawkes processes have been widely adopted for such tasks due to their success in describing cascading behaviors. In this paper, for general, marked Hawkes point processes, we present closed-form expressions for the mean and variance of future event counts, conditioned on observed events. Furthermore, these expressions allow us to develop a predictive approach, namely, Cascade Anytime Size Prediction via self-Exciting Regression model (CASPER), which is specifically tailored to popularity prediction, unlike existing generative approaches {–} based on point processes {–} for the same task. We showcase CASPER’s merits via experiments entailing both synthetic and real-world data, and demonstrate that it considerably improves upon prior works in terms of accuracy, especially for early-stage prediction.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-zhang22a, title = {Anytime Information Cascade Popularity Prediction via Self-Exciting Processes}, author = {Zhang, Xi and Aravamudan, Akshay and Anagnostopoulos, Georgios C}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {26028--26047}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/zhang22a/zhang22a.pdf}, url = {https://proceedings.mlr.press/v162/zhang22a.html}, abstract = {One important aspect of understanding behaviors of information cascades is to be able to accurately predict their popularity, that is, their message counts at any future time. Self-exciting Hawkes processes have been widely adopted for such tasks due to their success in describing cascading behaviors. In this paper, for general, marked Hawkes point processes, we present closed-form expressions for the mean and variance of future event counts, conditioned on observed events. Furthermore, these expressions allow us to develop a predictive approach, namely, Cascade Anytime Size Prediction via self-Exciting Regression model (CASPER), which is specifically tailored to popularity prediction, unlike existing generative approaches {–} based on point processes {–} for the same task. We showcase CASPER’s merits via experiments entailing both synthetic and real-world data, and demonstrate that it considerably improves upon prior works in terms of accuracy, especially for early-stage prediction.} }
Endnote
%0 Conference Paper %T Anytime Information Cascade Popularity Prediction via Self-Exciting Processes %A Xi Zhang %A Akshay Aravamudan %A Georgios C Anagnostopoulos %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-zhang22a %I PMLR %P 26028--26047 %U https://proceedings.mlr.press/v162/zhang22a.html %V 162 %X One important aspect of understanding behaviors of information cascades is to be able to accurately predict their popularity, that is, their message counts at any future time. Self-exciting Hawkes processes have been widely adopted for such tasks due to their success in describing cascading behaviors. In this paper, for general, marked Hawkes point processes, we present closed-form expressions for the mean and variance of future event counts, conditioned on observed events. Furthermore, these expressions allow us to develop a predictive approach, namely, Cascade Anytime Size Prediction via self-Exciting Regression model (CASPER), which is specifically tailored to popularity prediction, unlike existing generative approaches {–} based on point processes {–} for the same task. We showcase CASPER’s merits via experiments entailing both synthetic and real-world data, and demonstrate that it considerably improves upon prior works in terms of accuracy, especially for early-stage prediction.
APA
Zhang, X., Aravamudan, A. & Anagnostopoulos, G.C.. (2022). Anytime Information Cascade Popularity Prediction via Self-Exciting Processes. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:26028-26047 Available from https://proceedings.mlr.press/v162/zhang22a.html.

Related Material