Anytime Information Cascade Popularity Prediction via Self-Exciting Processes

被引:0
作者
Zhang, Xi [1 ]
Aravamudan, Akshay [1 ]
Anagnostopoulos, Georgios C. [1 ]
机构
[1] Florida Inst Technol, Dept Comp Engn & Sci, Melbourne, FL 32901 USA
来源
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 162 | 2022年
关键词
POINT; SPECTRA;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
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.
引用
收藏
页数:20
相关论文
共 35 条
[1]  
ATHAR MS, 2020, SOCIAL NETWORKING CO, P775
[2]   First- and Second-Order Statistics Characterization of Hawkes Processes and Non-Parametric Estimation [J].
Bacry, Emmanuel ;
Muzy, Jean-Francois .
IEEE TRANSACTIONS ON INFORMATION THEORY, 2016, 62 (04) :2184-2202
[3]  
Bao P, 2013, PROCEEDINGS OF THE 22ND INTERNATIONAL CONFERENCE ON WORLD WIDE WEB (WWW'13 COMPANION), P177
[4]   DeepHawkes: Bridging the Gap between Prediction and Understanding of Information Cascades [J].
Cao, Qi ;
Shen, Huawei ;
Cen, Keting ;
Ouyang, Wentao ;
Cheng, Xueqi .
CIKM'17: PROCEEDINGS OF THE 2017 ACM CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, 2017, :1149-1158
[5]   MARKED SELF-EXCITING POINT PROCESS MODELLING OF INFORMATION DIFFUSION ON TWITTER [J].
Chen, Feng ;
Tan, Wai Hong .
ANNALS OF APPLIED STATISTICS, 2018, 12 (04) :2175-2196
[6]   Information Diffusion Prediction via Recurrent Cascades Convolution [J].
Chen, Xueqin ;
Zhou, Fan ;
Zhang, Kunpeng ;
Trajcevski, Goce ;
Zhong, Ting ;
Zhang, Fengli .
2019 IEEE 35TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE 2019), 2019, :770-781
[7]   AN ELEMENTARY DERIVATION OF MOMENTS OF HAWKES PROCESSES [J].
Cui, Lirong ;
Hawkes, Alan ;
Yi, He .
ADVANCES IN APPLIED PROBABILITY, 2020, 52 (01) :102-137
[8]  
Dassios A, 2011, ADV APPL PROBAB, V43, P814
[9]   Affine Point Processes and Portfolio Credit Risk [J].
Errais, Eymen ;
Giesecke, Kay ;
Goldberg, Lisa R. .
SIAM JOURNAL ON FINANCIAL MATHEMATICS, 2010, 1 (01) :642-665
[10]   Taxonomy and Evaluation for Microblog Popularity Prediction [J].
Gao, Xiaofeng ;
Cao, Zhenhao ;
Li, Sha ;
Yao, Bin ;
Chen, Guihai ;
Tang, Shaojie .
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2019, 13 (02)