Predicting Twitter User Demographics from Names Alone
By Zach Wood-Doughty, Nicholas Andrews, Rebecca Marvin, Mark Dredze. 2018
44 citations recorded by the source corpus.
Social media analysis frequently requires tools that can automatically infer demographics to contextualize trends. These tools often require hundreds of user-authored messages for each user, which may be prohibitive to obtain when analyzing millions of users. We explore character-level neural models that learn a representation of a user’s name and screen name to predict gender and ethnicity, allowing for demographic inference with minimal data. We release trained models1 which may enable new demographic analyses that would otherwise require enormous amounts of data collection
Subfields: Artificial Intelligence, Sociology and Political Science, Natural Language Processing, Social Sciences & Humanities
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Yulia Tsvetkov
- Anjalie Field
- Zeerak Talat
- Su Lin Blodgett
- Oskar van der Wal
- Willem H. Zuidema
- Katrin Schulz
- Alina Leidinger
In Sociology and Political Science
In Natural Language Processing
- Yulia Tsvetkov
- Anjalie Field
- Su Lin Blodgett
- Zeerak Talat
- Willem H. Zuidema
- Oskar van der Wal
- Katrin Schulz
- Noah A. Smith
In Social Sciences & Humanities
Citations in the corpus
- A Survey of Race, Racism, and Anti-Racism in NLP (2021, background): “For example, names drawn from the U.S. census generalize poorly to Twitter (Wood-Doughty et al., 2018), and names common among Black and white children were not distinctly different prior to the 1970s (Fryer Jr and Levitt, 2004; Sweeney, 2013).”
- Undesirable biases in NLP: Averting a crisis of measurement (2022, background): “, names) is another challenge at the operationalization level (Sweeney, 2013; Fryer Jr & Levitt, 2004; Wood-Doughty et al., 2018).”