This paper describes the UWaterloo affect prediction system developed for EmoInt- 2017. We delve into our feature se- lection approach for affect intensity, af- fect presence, sentiment intensity and sentiment presence lexica alongside pre- trained word embeddings, which are uti- lized to extract emotion intensity signals from tweets in an ensemble learning ap- proach. The system employs emotion spe- cific model training, and utilizes distinct models for each of the emotion corpora in isolation. Our system utilizes gradient boosted regression as the primary learning technique to predict the final emotion in- tensities.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados