Wb-MSF

Wb-MSF: Multi-Source Information Diffusion Dataset on Weibo

Dataset Description

(Re)Post Cascade

Each cascade is saved in a csv file with following properties:

Property	Description
origin_id	precursor post id (be reposted)
origin_uid	user id of precursor post
id	post id
uid	user id
created_at	(re)post time

• The properties of italics is the private user information, which are hashed;
• Property created_at is format as YYYY-mm-DD HH:MM:SS ;
• If origin_id is NULL and origin_uid == uid , means this post is original and not reposted.

Followership

Each dataset has a global followership network, format as edge list: each line of {dataname}_followership.txt is a pair of comma separated user ids $u,v$ , means $u$ is followed by $v$ and $v$ is one of $u$'s fans.

Hash

To protect user privacy, we anonymize the data: i.e., we hash the fields related to user-id and post-id.

Download

• The small-scale version (MSC and SSC both contain 1K cascades): WB-MSF-1k
• Full version: WB-MSF

Cite Us

@inproceedings{WBMSF_2022,
  author     = {Wu, Zhen and Zhou, Jingya and Wang, Jie and Sun, Xigang},
  title      = {Wb-MSF: A Large-scale Multi-source Information Diffusion Dataset for Social Information Diffusion Prediction},
  booktitle  = {2022 Tenth International Conference on Advanced Cloud and Big Data (CBD)},
  year       = {2022},
}

@inproceedings{HERIGCN_2022,
  author          = {Wu, Zhen and Zhou, Jingya and Liu, Ling and Li, Chaozhuo and Gu, Fei},
  title           = {Deep Popularity Prediction in Multi-Source Cascade with HERI-GCN},
  booktitle       = {2022 IEEE 38th International Conference on Data Engineering (ICDE)},
  year            = {2022}
}