cite If you find this dataset and code useful in your research, please consider citing:
@article{hallac2021user,
title = {User Representation Learning for Social Networks: An Empirical Study},
author = {Hallac, Ibrahim Riza and Ay, Betul and Aydin, Galip},
journal = {Applied Sciences},
volume = {11},
number = {12},
pages = {5489},
year = {2021},
publisher = {Multidisciplinary Digital Publishing Institute}
}
https://www.mdpi.com/2076-3417/11/12/5489
Each user belongs to one of the 5 predefined categories: Economy, Crypto Economy, Technology, Fashion, and Politics.
Tweets are not publicly shared due to Twitter API’s Terms of Service. Because "Twitter does not allow for making large amounts of raw Twitter data available on the Web" [1]
In order to get the actual Text of the tweets you should rehydrate them with their IDs. You can use a tool like Twarch [2] for harvesting Tweets. Please contact to corresponding author for the list of ids used in the paper.
[1] Digital Collecting Toolkit, http://digitalcollecting.lib.virginia.edu/toolkit/
[2] "twarc is a command line tool and Python library for archiving Twitter", https://github.com/DocNow/twarc
- Mutlu Halil ERİŞEN
- Zeynep KOYUN
- Semiha MAKİNİST
- İbrahim Rıza HALLAÇ