EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Comparative Analysis Of Convolutional Neural Networks, Long Short-Term Memory Networks, and Bert For Text-Based Emotion Classification

Porawat Visutsak, Tanajak Tongbai, Duongduen Ongrungruaeng, Nuttiruj Phongwuttisak, Surapong Wiriya, Keun Ho Ryu

ASEAN Journal of Scientific and Technological Reports · 10.05.2026 · 29 (6) · e262378

DOI: 10.55164/ajstr.v29i6.262378 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ASEAN Journal of Scientific and Technological Reports, online erschienen am 10.05.2026. Verfasst von Porawat Visutsak und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents a detailed experimental study and comparative analysis of three popular deep learning architectures (CNN, LSTM, and BERT) for emotion classification in written messages. Using a publicly available dataset of six unique emotional states (sadness, joy, love, anger, fear, and surprise), an effective ablation study was conducted to determine optimal architectural configurations, including a sequence length of 66 tokens and an embedding size of 200. To validate the results of a comparative analysis of model performance, a bootstrap technique (30 trials) and the Wilcoxon Signed-Rank test were used to eliminate potential bias. As shown by experiments, the tuned BERT architecture (with a learning rate of 2e-5) produced the most accurate and reliable result of 93.50% in classifying emotional states from texts. Moreover, with an appropriate sequence length configuration, the LSTM network (89.92%) significantly outperformed the CNN (89.65%), confirming the need to account for long-range dependencies in emotion classification. Overall, the research results show the key importance of hyperparameter tuning and the ability to handle complex information for emotion identification.

Zitieren

APA

Visutsak, P., Tongbai, T., Ongrungruaeng, D., Phongwuttisak, N., Wiriya, S., & Ryu, K. H. (2026). Comparative Analysis Of Convolutional Neural Networks, Long Short-Term Memory Networks, and Bert For Text-Based Emotion Classification. ASEAN Journal of Scientific and Technological Reports, 29(6), e262378. https://doi.org/10.55164/ajstr.v29i6.262378

Vancouver

Visutsak P, Tongbai T, Ongrungruaeng D, Phongwuttisak N, Wiriya S, Ryu KH. Comparative Analysis Of Convolutional Neural Networks, Long Short-Term Memory Networks, and Bert For Text-Based Emotion Classification. ASEAN Journal of Scientific and Technological Reports. 2026;29(6):e262378. doi:10.55164/ajstr.v29i6.262378

BibTeX
@article{Visutsak2026,
  author = {Visutsak, Porawat and Tongbai, Tanajak and Ongrungruaeng, Duongduen and Phongwuttisak, Nuttiruj and Wiriya, Surapong and Ryu, Keun Ho},
  title = {Comparative Analysis Of Convolutional Neural Networks, Long Short-Term Memory Networks, and Bert For Text-Based Emotion Classification},
  journal = {ASEAN Journal of Scientific and Technological Reports},
  year = {2026},
  volume = {29},
  number = {6},
  pages = {e262378},
  doi = {10.55164/ajstr.v29i6.262378}
}
Kostenlos starten →