HawkesTopic: A Joint Model for Network Inference and Topic Modeling from Text-Based Cascades

Xinran He, Theodoros Rekatsinas, James Foulds, Lise Getoor, Yan Liu
Proceedings of the 32nd International Conference on Machine Learning, PMLR 37:871-880, 2015.

Abstract

Understanding the diffusion of information in social network and social media requires modeling the text diffusion process. In this work, we develop the HawkesTopic model (HTM) for analyzing text-based cascades, such as "retweeting a post" or "publishing a follow-up blog post". HTM combines Hawkes processes and topic modeling to simultaneously reason about the information diffusion pathways and the topics characterizing the observed textual information. We show how to jointly infer them with a mean-field variational inference algorithm and validate our approach on both synthetic and real-world data sets, including a news media dataset for modeling information diffusion, and an ArXiv publication dataset for modeling scientific influence. The results show that HTM is significantly more accurate than several baselines for both tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v37-he15, title = {HawkesTopic: A Joint Model for Network Inference and Topic Modeling from Text-Based Cascades}, author = {He, Xinran and Rekatsinas, Theodoros and Foulds, James and Getoor, Lise and Liu, Yan}, booktitle = {Proceedings of the 32nd International Conference on Machine Learning}, pages = {871--880}, year = {2015}, editor = {Bach, Francis and Blei, David}, volume = {37}, series = {Proceedings of Machine Learning Research}, address = {Lille, France}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v37/he15.pdf}, url = {https://proceedings.mlr.press/v37/he15.html}, abstract = {Understanding the diffusion of information in social network and social media requires modeling the text diffusion process. In this work, we develop the HawkesTopic model (HTM) for analyzing text-based cascades, such as "retweeting a post" or "publishing a follow-up blog post". HTM combines Hawkes processes and topic modeling to simultaneously reason about the information diffusion pathways and the topics characterizing the observed textual information. We show how to jointly infer them with a mean-field variational inference algorithm and validate our approach on both synthetic and real-world data sets, including a news media dataset for modeling information diffusion, and an ArXiv publication dataset for modeling scientific influence. The results show that HTM is significantly more accurate than several baselines for both tasks.} }
Endnote
%0 Conference Paper %T HawkesTopic: A Joint Model for Network Inference and Topic Modeling from Text-Based Cascades %A Xinran He %A Theodoros Rekatsinas %A James Foulds %A Lise Getoor %A Yan Liu %B Proceedings of the 32nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Francis Bach %E David Blei %F pmlr-v37-he15 %I PMLR %P 871--880 %U https://proceedings.mlr.press/v37/he15.html %V 37 %X Understanding the diffusion of information in social network and social media requires modeling the text diffusion process. In this work, we develop the HawkesTopic model (HTM) for analyzing text-based cascades, such as "retweeting a post" or "publishing a follow-up blog post". HTM combines Hawkes processes and topic modeling to simultaneously reason about the information diffusion pathways and the topics characterizing the observed textual information. We show how to jointly infer them with a mean-field variational inference algorithm and validate our approach on both synthetic and real-world data sets, including a news media dataset for modeling information diffusion, and an ArXiv publication dataset for modeling scientific influence. The results show that HTM is significantly more accurate than several baselines for both tasks.
RIS
TY - CPAPER TI - HawkesTopic: A Joint Model for Network Inference and Topic Modeling from Text-Based Cascades AU - Xinran He AU - Theodoros Rekatsinas AU - James Foulds AU - Lise Getoor AU - Yan Liu BT - Proceedings of the 32nd International Conference on Machine Learning DA - 2015/06/01 ED - Francis Bach ED - David Blei ID - pmlr-v37-he15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 37 SP - 871 EP - 880 L1 - http://proceedings.mlr.press/v37/he15.pdf UR - https://proceedings.mlr.press/v37/he15.html AB - Understanding the diffusion of information in social network and social media requires modeling the text diffusion process. In this work, we develop the HawkesTopic model (HTM) for analyzing text-based cascades, such as "retweeting a post" or "publishing a follow-up blog post". HTM combines Hawkes processes and topic modeling to simultaneously reason about the information diffusion pathways and the topics characterizing the observed textual information. We show how to jointly infer them with a mean-field variational inference algorithm and validate our approach on both synthetic and real-world data sets, including a news media dataset for modeling information diffusion, and an ArXiv publication dataset for modeling scientific influence. The results show that HTM is significantly more accurate than several baselines for both tasks. ER -
APA
He, X., Rekatsinas, T., Foulds, J., Getoor, L. & Liu, Y.. (2015). HawkesTopic: A Joint Model for Network Inference and Topic Modeling from Text-Based Cascades. Proceedings of the 32nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 37:871-880 Available from https://proceedings.mlr.press/v37/he15.html.

Related Material