This paper describes our participation in the ITAmoji task at EVALITA2018 (Ronzano et al., 2018). Our approach is based on three sets of features,i.e. micro-blog and keyword features, sentiment lexicon features and semantic features. We exploit these features to train and combine several classifiers using different libraries. The results show how the selected features are not appropriate for training a linear classifier to properly address the emoji prediction task.
The UNIBA System at the EVALITA 2018 Italian Emoji Prediction Task
Lucia Siciliani
;Daniela Girardi
2018-01-01
Abstract
This paper describes our participation in the ITAmoji task at EVALITA2018 (Ronzano et al., 2018). Our approach is based on three sets of features,i.e. micro-blog and keyword features, sentiment lexicon features and semantic features. We exploit these features to train and combine several classifiers using different libraries. The results show how the selected features are not appropriate for training a linear classifier to properly address the emoji prediction task.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
