Resumen de IITP at EmoInt-2017: Measuring Intensity of Emotions using Sentence Embeddings and Optimized

Md Shad Akhtar, Palaash Sawant, Asif Ekbal, Jyoti Pawar, Pushpak Bhattacharyya

• This paper describes the system that we submitted as part of our participation in the shared task on Emotion Inten- sity (EmoInt-2017). We propose a Long short term memory (LSTM) based archi- tecture cascaded with Support Vector Re- gressor (SVR) for intensity prediction. We also employ Particle Swarm Optimization (PSO) based feature selection algorithm for obtaining an optimized feature set for training and evaluation. System evaluation shows interesting results on the four emo- tion datasets i.e. anger, fear, joy and sad- ness. In comparison to the other partici- pating teams our system was ranked 5th in the competition.