Resumen de LIPN-UAM at EmoInt-2017: Combination of Lexicon-based features and Sentence-level Vector Representations for Emotion Intensity Determination

Davide Buscaldi, Ángeles Belém Priego

• This paper presents the combined LIPN- UAM participation in the WASSA 2017 Shared Task on Emotion Intensity. In particular, the paper provides some high- lights on the system that was presented to the shared task, partly based on the Tweetaneuse system used to participate in a French Sentiment Analysis task (DEFT2017). We combined lexicon-based features with sentence-level vector rep- resentations to obtain a random forest model.