|
Activity Number:
|
9
- When Location Is Random: Advances in Statistical Modeling and Inference for Spatial Point Processes
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #322098
|
|
|
Title:
|
A New Clustered Temporal Point Process Model with Application to Social Media Data
|
|
Author(s):
|
Jingfei Zhang and Yongtao Guan* and Hansheng Wang and Xuening Zhu
|
|
Companies:
|
University of Miami and University of Miami and Peking University and Peking University
|
|
Keywords:
|
point process ;
social media ;
EM algorithm
|
|
Abstract:
|
In this paper, we propose a new class of clustered temporal point process model to model the posting patterns of users of Sina Weibo, the largest twitter-type online social media in China. The proposed model captures both inhomogeneity in the initial posting time as well as the clustering pattern in the subsequent posts following the initial post. We develop two EM-type algorithms for estimating the parameters in the proposed model, based on which we cluster the di?erent user patterns. In the application to real data, we discover interesting subgroups of users with distinct behaviors in terms of their initial posts and subsequent posts following the initial posts.
|
Authors who are presenting talks have a * after their name.
