NUIG at EmoInt-2017: BiLSTM and SVR Ensemble to Detect Emotion Intensity

• Vladimir Andryushechkin ^[1] ; Ian David Wood ^[1] ; James O’ Neill ^[1]
  1. [1] National University of Ireland

     National University of Ireland

     Irlanda

     
• Localización: 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis WASSA 2017: Proceedings of the Workshop / Alexandra Balahur Dobrescu (ed. lit.), Saif M. Mohammad (ed. lit.), Erik van der Goot (ed. lit.), 2017, ISBN 978-1-945626-95-1, págs. 175-179
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • This paper describes the entry NUIG in the WASSA 20171 shared task on emo- tion recognition. The NUIG system used an SVR (SVM regression) and BiLSTM ensemble, utilizing primarily n-grams (for SVR features) and tweet word embeddings (for BiLSTM features). Experiments were carried out on several other candidate fea- tures, some of which were added to the SVR model. Parameter selection for the SVR model was run as a grid search whilst parameters for the BiLSTM model were selected through a non-exhaustive ad-hoc search.