TweetGenie attracts Media Attention

Theo Meder

One of the private-public projects funded by the Royal Netherlands Academy for Arts and Sciences (KNAW) is called TINPOT:  ‘Taal, Indentiteit, Netwerken en Produktgeruchten Op Twitter” (Language, Identity, Networks and Product Rumors on Twitter). Three parties are participating in the project: the Meertens Instituut, the University of Twente and Teezir BV in Utrecht. Project leader is Theo Meder, and the researchers involved are Rilana Gravel, Dong Nguyen, and Dolf Trieschnigg, while the annotators were Daphne van Kessel and Charlotte van Tongeren. The project started in September 2012. One of the objectives of the project was to find out how much language on Twitter could tell us about the age and gender of the people involved. Sociolinguist Rilana Gravel did theoretical research and Dong Nguyen built a self-learning app called TweetGenie, that is able to guess the gender and age of people writing on Twitter, based on their use of language (so not based on names, photos or profile information).

If possible, TINPOT would like to do more research into the conduits that rumors and stories follow on Twitter, and what kind of micro-variation can occur.

Nguyen, Gravel, Trieschnigg and Meder wrote a paper called ‘”How Old Do You Think I Am?”: A Study of Language and Age in Twitter’ that can be downloaded here:

Soon after this article came online, the media started to show interest:

New Scientist (April 10):

United Press International (April 11):

UT Nieuws (April 12):

New York Times (April 15):

On May 13th TweetGenie was officially lanched: TweetGenie only works with Dutch accounts.

After the launch a lot of media attention followed once again. An anthology:

The Raw Story (May 13):

NRC (May 13):

Algemeen Dagblad (May 13):

Telegraaf (May 13):

NOS (May 13):

RTL Nieuws (May 13):

ScienceGuide (May 14):

Nu tech (May 15):