EmoNet: Fine-Grained Emotion Detection with Gated Recurrent Neural Networks

By Muhammad Abdul-Mageed, Lyle Ungar. Annual Meeting of the Association for Computational Linguistics, 2017

320 citations recorded by the source corpus.

Accurate detection of emotion from natural language has applications ranging from building emotional chatbots to better understanding individuals and their lives. However, progress on emotion detection has been hampered by the absence of large labeled datasets. In this work, we build a very large dataset for fine-grained emotions and develop deep learning models on it. We achieve a new state-of-the-art on 24 fine-grained types of emotions (with an average accuracy of 87.58%). We also extend the task beyond emotion types to model Robert Plutick’s 8 primary emotion dimensions, acquiring a superior accuracy of 95.68%.

Subfields: Artificial Intelligence, Experimental and Cognitive Psychology

View Paper: https://www.semanticscholar.org/paper/6a6f1b8ef3eefaf8027a76134859b44886a5f795

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Sven Buechel
2. Udo Hahn
3. Susanna Rücker
4. Saif M. Mohammad
5. Shuai Wang
6. B. Liu
7. Lei Zhang
8. Bing Liu

In Experimental and Cognitive Psychology

1. Sven Buechel
2. Udo Hahn
3. Susanna Rücker
4. Saif M. Mohammad
5. B. Liu
6. Meghana Joshi
7. Chunyan Miao
8. Shuai Wang

Citations in the corpus

• Word Emotion Induction for Multiple Languages as a Deep Multi-Task Learning Problem (2018, methodology): “For this problem, DL approaches are meanwhile fully established as the method of choice (Wang et al., 2016b; Abdul-Mageed and Ungar, 2017; Felbo et al., 2017; Mohammad and Bravo-Marquez, 2017).”
• Deep learning for sentiment analysis: A survey (2018, methodology): “, 2015a; Zhang, Zhang, & Fu, 2016; Zhang, Zhang, & Vo, 2016) Abdul-Mageed and Ungar (2017) first built a large dataset for emotion detection automatically by using distant supervision and then used a GRU network for fine-grained emotion detection. Felbo, Mislove, Søgaard, Rahwan, and Lehmann (2017) ”
• Knowledge-Enriched Transformer for Emotion Detection in Textual Conversations (2019, background): “, 2018) to deep learning methods (Abdul-Mageed and Ungar, 2017; Zhang et al., 2018b) for emotion detection in text.”
• GoEmotions: A Dataset of Fine-Grained Emotions (2020, background): “Other datasets are automatically weakly-labeled, based on emotion-related hashtags on Twitter (Wang et al., 2012; Abdul-Mageed and Ungar, 2017).”
• Learning and Evaluating Emotion Lexicons for 91 Languages (2020, background): “…an increasing number of researchers focus on more expressive encodings for affective states inspired by distinct lines of work in psychology (Yu et al., 2016; Buechel and Hahn, 2017; Sedoc et al., 2017; Abdul-Mageed and Ungar, 2017; Bostan and Klinger, 2018; Mohammad, 2018; Troiano et al., 2019).”
• SemEval-2019 Task 3: EmoContext Contextual Emotion Detection in Text (2019, background): “Abdul-Mageed and Ungar (2017) and Köper et al.”
• CARER: Contextualized Affect Representations for Emotion Recognition (2018, methodology): “Moreover, we re-implemented the GRNN model proposed in (Abdul-Mageed and Ungar, 2017).”
• Sentiment Analysis: Detecting Valence, Emotions, and Other Affectual States from Text (2016, background): “Abdul-Mageed and Ungar (2017) collected about 250 million tweets pertaining to 24 emotions and 665 hashtags and developed a Gated Recurrent Neural Network (GRNN) model for emotion classification.”
• Suicide Ideation Detection via Social and Temporal User Representations using Hyperbolic Learning (2021, methodology): “Thus, building on this correlation between emotions and suicidal ideation, we finetune BERT on EmoNet (Abdul-Mageed and Ungar, 2017) for capturing fine-grained (Plutchik-based) emotions (Plutchik, 1980; Sawhney et al., 2020).”