Given a corpus of 100 tweets by a given user, the goal of this algorithm is to classify it bewteen BOT and HUMAN, and for the latter between MALE and FEMALE.
This is the result of my participation at the shared task "Bots and Gender Profiling" for PAN @ CLEF 2019, in the context on my class of "Applications of Computational Linguistics" from the Master of Artificial intelligence, Pattern Recognition and Digital Imaging (MIARFID) at the UPV, Valencia, Spain.
The training dataset is available at https://datasetsearch.research.google.com/search?query=pan19-bots-and-gender-profiling&docid=3Oxre9WrBUkRcQRTAAAAAA%3D%3D
- Clean the code, which was not written for the whole world to see
- Add a simple interface to use the models with any twitter user