Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

HDEL: a hierarchical deep ensemble approach for text-based emotion detection

HDEL: a hierarchical deep ensemble approach for text-based emotion detection Emotion detection from social media data plays a crucial role in studying societal emotions concerning different events, aiding in predicting the reactions of specific social groups. However, it is complex to automatically extract implicit emotional information from noisy social media text data. This study introduces the Hierarchical Deep Ensemble Learning (HDEL) system to identify emotions in text data. The proposed HDEL model utilizes BiLSTM (Bidirectional Long Short-Term Memory), CNN (Convolutional Neural Network), BiGRU (Bidirectional Gated Recurrent Unit), and RCNN (Recurrent Convolutional Neural Network) in the first level of its hierarchy. The predicted probabilities of the four models are embedded with input data to prepare the intermediate hybrid data. This hybrid data is fed to the next layer of the proposed system, which utilizes a Random Forest (RF) algorithm to predict the emotion. The proposed approach is tested using three emotion datasets: the WASSA-2017 Emotion Intensity (EmoInt) dataset, the International Survey on Emotion Antecedents and Reactions (ISEAR) dataset, and the CrowdFlower (CF) dataset. EmoInt and ISEAR are clean and balanced, while CF is noisy and imbalanced. The results are compared with various state-of- the-art Machine Learning models. The outperforming results depict the superiority of the proposed approach. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Multimedia Tools and Applications Springer Journals

HDEL: a hierarchical deep ensemble approach for text-based emotion detection

Download PDF
22 pages

Loading next page...
 
/lp/springer-journals/hdel-a-hierarchical-deep-ensemble-approach-for-text-based-emotion-p77gznJPo6

References (43)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024
ISSN
1380-7501
eISSN
1573-7721
DOI
10.1007/s11042-024-19032-y
Publisher site
See Article on Publisher Site

Abstract

Emotion detection from social media data plays a crucial role in studying societal emotions concerning different events, aiding in predicting the reactions of specific social groups. However, it is complex to automatically extract implicit emotional information from noisy social media text data. This study introduces the Hierarchical Deep Ensemble Learning (HDEL) system to identify emotions in text data. The proposed HDEL model utilizes BiLSTM (Bidirectional Long Short-Term Memory), CNN (Convolutional Neural Network), BiGRU (Bidirectional Gated Recurrent Unit), and RCNN (Recurrent Convolutional Neural Network) in the first level of its hierarchy. The predicted probabilities of the four models are embedded with input data to prepare the intermediate hybrid data. This hybrid data is fed to the next layer of the proposed system, which utilizes a Random Forest (RF) algorithm to predict the emotion. The proposed approach is tested using three emotion datasets: the WASSA-2017 Emotion Intensity (EmoInt) dataset, the International Survey on Emotion Antecedents and Reactions (ISEAR) dataset, and the CrowdFlower (CF) dataset. EmoInt and ISEAR are clean and balanced, while CF is noisy and imbalanced. The results are compared with various state-of- the-art Machine Learning models. The outperforming results depict the superiority of the proposed approach.

Journal

Multimedia Tools and Applications – Springer Journals

Published: Jan 1, 2025

Keywords: Emotion Classification; Deep Learning; Ensemble Learning; Random Forest

There are no references for this article.