UBC Entity Linking at TAC-KBP 2013: random forests for high accuracy
Ander Barrena, Eneko Agirre, Aitor Soroa · Theory and applications of categories · 2013
This paper describe our different runs submitted for the Entity Linking task at TAC-KBP 2013. We developed two systems, one is a generative entity linking model and the other is a supervised system reusing the scores of the previous model using random forests. Our main research interest is Named Entity Disambiguation and we thus performed a very naive clustering of NIL instances. In fact, our best run scores at a par to the best system on accuracy (ignoring NIL clustering), and close to the top performance on KB mentions, both in accuracy and B-cubed F1.