Organizational Social Network Datasets
Many organizations possess social media accounts on different social networks,
but these profiles are not always linked. Most existing works on social network
entity linking focus on linking individuals, and do not model features specific
for organizational linking. We address this gap to not only link official
social media accounts, but also discover and solve the identification
and linking of associated affiliate accounts -- such as geographical
divisions and brands -- which are important to distinguish. We have released
experimental datasets used in our papers. If you are interested in linking
organizational profiles on social networks, please try our dataset for your experiments!
- Twitter profiles (228 organizations [Official: 232, Affiliate: 675, Unrelated: 2,474])
- Facebook profiles (216 organizations [Official: 145, Affiliate: 491, Unrelated: 2,767])
These are the MongoDB label collections for Facebook and Twitter profiles, where each entry consists of:
- Profile ID
- "Query" (company name)
- The actual label: "unrelated", "official", or "affiliate"
- Stop words list ("company", "corporation", and so on)
- According to "Terms of Services" for developers in both Twitter and Facebook, we cannot release the actual profiles.
But those who hope to use our dataset can retrieve them from the profile IDs.
Jerome Cheng (National University of Singapore)