TEmoX: Classification of Textual Emotion Using Ensemble of Transformers
Textual emotion classification (TxtEC) refers to the classification of emotion expressed by individuals in textual form. The widespread use of the Internet and numerous Web 2.0 applications has emerged in an expeditious growth of textual interactions. However, determining emotion from texts is chall...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10264097/ |
| _version_ | 1797660878345076736 |
|---|---|
| author | Avishek Das Mohammed Moshiul Hoque Omar Sharif M. Ali Akber Dewan Nazmul Siddique |
| author_facet | Avishek Das Mohammed Moshiul Hoque Omar Sharif M. Ali Akber Dewan Nazmul Siddique |
| author_sort | Avishek Das |
| collection | DOAJ |
| description | Textual emotion classification (TxtEC) refers to the classification of emotion expressed by individuals in textual form. The widespread use of the Internet and numerous Web 2.0 applications has emerged in an expeditious growth of textual interactions. However, determining emotion from texts is challenging due to their unorganized, unstructured, and disordered forms. While research in textual emotion classification has made considerable breakthroughs for high-resource languages, it is yet challenging for low-resource languages like Bengali. This work presents a transformer-based ensemble approach (called TEmoX) to categorize Bengali textual data into six integral emotions: joy, anger, disgust, fear, sadness, and surprise. This research investigates 38 classifier models developed using four machine learning LR, RF, MNB, SVM, three deep-learning CNN, BiLSTM, CNN+BiLSTM, five transformer-based m-BERT, XLM-R, Bangla-BERT-1, Bangla-BERT-2, and Indic-DistilBERT techniques with two ensemble strategies and three embedding techniques. The developed models are trained, tuned, and tested on the three versions of the Bengali emotion text corpus BEmoC-v1, BEmoC-v2, BEmoC-v3. The experimental outcomes reveal that the weighted ensemble of four transformer models En-22: Bangla-BERT-2, XLM-R, Indic-DistilBERT, Bangla-BERT-1 outperforms the baseline models and existing methods by providing the maximum weighted <inline-formula> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula>-score (80.24%) on BEmoC-v3. The dataset, models, and fractions of codes are available at <uri>https://github.com/avishek-018/TEmoX</uri>. |
| first_indexed | 2024-03-11T18:36:14Z |
| format | Article |
| id | doaj.art-cd447683baf24827bca6d4cd7c5a3e0b |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T18:36:14Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-cd447683baf24827bca6d4cd7c5a3e0b2023-10-12T23:00:41ZengIEEEIEEE Access2169-35362023-01-011110980310981810.1109/ACCESS.2023.331945510264097TEmoX: Classification of Textual Emotion Using Ensemble of TransformersAvishek Das0https://orcid.org/0000-0002-1589-8322Mohammed Moshiul Hoque1https://orcid.org/0000-0001-8806-708XOmar Sharif2M. Ali Akber Dewan3https://orcid.org/0000-0001-6347-7509Nazmul Siddique4https://orcid.org/0000-0002-0642-2357Department of Computer Science and Engineering, Chittagong University of Engineering and Technology, Chittagong, BangladeshDepartment of Computer Science and Engineering, Chittagong University of Engineering and Technology, Chittagong, BangladeshDepartment of Computer Science and Engineering, Chittagong University of Engineering and Technology, Chittagong, BangladeshSchool of Computing and Information Systems, Faculty of Science and Technology, Athabasca University, Athabasca, CanadaSchool of Computing, Engineering and Intelligent Systems, Ulster University, Londonderry, U.K.Textual emotion classification (TxtEC) refers to the classification of emotion expressed by individuals in textual form. The widespread use of the Internet and numerous Web 2.0 applications has emerged in an expeditious growth of textual interactions. However, determining emotion from texts is challenging due to their unorganized, unstructured, and disordered forms. While research in textual emotion classification has made considerable breakthroughs for high-resource languages, it is yet challenging for low-resource languages like Bengali. This work presents a transformer-based ensemble approach (called TEmoX) to categorize Bengali textual data into six integral emotions: joy, anger, disgust, fear, sadness, and surprise. This research investigates 38 classifier models developed using four machine learning LR, RF, MNB, SVM, three deep-learning CNN, BiLSTM, CNN+BiLSTM, five transformer-based m-BERT, XLM-R, Bangla-BERT-1, Bangla-BERT-2, and Indic-DistilBERT techniques with two ensemble strategies and three embedding techniques. The developed models are trained, tuned, and tested on the three versions of the Bengali emotion text corpus BEmoC-v1, BEmoC-v2, BEmoC-v3. The experimental outcomes reveal that the weighted ensemble of four transformer models En-22: Bangla-BERT-2, XLM-R, Indic-DistilBERT, Bangla-BERT-1 outperforms the baseline models and existing methods by providing the maximum weighted <inline-formula> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula>-score (80.24%) on BEmoC-v3. The dataset, models, and fractions of codes are available at <uri>https://github.com/avishek-018/TEmoX</uri>.https://ieeexplore.ieee.org/document/10264097/Natural language processingtext classificationtextual emotion classificationBengali emotion text corpusensemble of transformers |
| spellingShingle | Avishek Das Mohammed Moshiul Hoque Omar Sharif M. Ali Akber Dewan Nazmul Siddique TEmoX: Classification of Textual Emotion Using Ensemble of Transformers IEEE Access Natural language processing text classification textual emotion classification Bengali emotion text corpus ensemble of transformers |
| title | TEmoX: Classification of Textual Emotion Using Ensemble of Transformers |
| title_full | TEmoX: Classification of Textual Emotion Using Ensemble of Transformers |
| title_fullStr | TEmoX: Classification of Textual Emotion Using Ensemble of Transformers |
| title_full_unstemmed | TEmoX: Classification of Textual Emotion Using Ensemble of Transformers |
| title_short | TEmoX: Classification of Textual Emotion Using Ensemble of Transformers |
| title_sort | temox classification of textual emotion using ensemble of transformers |
| topic | Natural language processing text classification textual emotion classification Bengali emotion text corpus ensemble of transformers |
| url | https://ieeexplore.ieee.org/document/10264097/ |
| work_keys_str_mv | AT avishekdas temoxclassificationoftextualemotionusingensembleoftransformers AT mohammedmoshiulhoque temoxclassificationoftextualemotionusingensembleoftransformers AT omarsharif temoxclassificationoftextualemotionusingensembleoftransformers AT maliakberdewan temoxclassificationoftextualemotionusingensembleoftransformers AT nazmulsiddique temoxclassificationoftextualemotionusingensembleoftransformers |