A Multi-task Ensemble Framework for Emotion, Sentiment and Intensity Prediction

Abstract

In this paper, through multi-task ensemble framework we address three problems of emotion and sentiment analysis i.e. "emotion classification & intensity", "valence, arousal & dominance for emotion" and "valence & arousal} for sentiment". The underlying problems cover two granularities (i.e. coarse-grained and fine-grained) and a diverse range of domains (i.e. tweets, Facebook posts, news headlines, blogs, letters etc.). The ensemble model aims to leverage the learned representations of three deep learning models (i.e. CNN, LSTM and GRU) and a hand-crafted feature representation for the predictions. Experimental results on the benchmark datasets show the efficacy of our proposed multi-task ensemble frameworks. We obtain the performance improvement of 2-3 points on an average over single-task systems for most of the problems and domains.

References in corpus (6)

• Adam: A Method for Stochastic Optimization
• Improving neural networks by preventing co-adaptation of feature detectors
• Twitter mood predicts the stock market
• Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales
• Multitask Learning for Fine-Grained Twitter Sentiment Analysis
• Crowdsourcing a Word-Emotion Association Lexicon

Cited by in corpus (3)

• A Lexical, Syntactic, and Semantic Perspective for Understanding Style in Text
• A Multi-Task Incremental Learning Framework with Category Name Embedding for Aspect-Category Sentiment Analysis
• Visual Interest Prediction with Attentive Multi-Task Transfer Learning