EUVIMEDEUVIMED entdecken
Crossref

From Machine Learning to Transformers: A Comparative Study on Emotion Classification of Tweets

Beenish Zafar, Ali Saeed, Hafiza Maria Kiran, Mobashirah Nasir, Afifa Hameed

VAWKUM Transactions on Computer Sciences · 01.03.2026 · 14 (1) · 40-52

DOI: 10.21015/vtcs.v14i1.2236 ↗

Kurz gefasst

Fachartikel in der Zeitschrift VAWKUM Transactions on Computer Sciences, online erschienen am 01.03.2026. Verfasst von Beenish Zafar und 4 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this era of social media, it is essential that policymakers, businessmen, and researchers focus on understanding the sentiments of the people through their expressions on social media. An analytical comparison of several well-known machine learning (ML) algorithms along with Transformer-based models has been exploited in this study to determine the emotions and sentiments of people through their tweets. ML algorithms included in the study consider classical models—Random Forest, SVM, ID3, Naive Bayes, KNN, MLP—and transformer models, namely BERT-Base and BERT-Large. These algorithms have been applied to a carefully curated dataset of tweets that have been labeled for their emotion, such as joy, fear, anger, and sadness. Evaluation metrics, including precision, accuracy, kappa score, F1 score, and recall, have been used to compare and analyze the performance of each algorithm. The findings of this paper reveal that both Random Forest and SVM showed the best accuracy, at 79.4% and 79.1% while Random Forest slightly outperformed SVM in accuracy. In Transformers, BERT-Base outperforms not only BERT-Large but also the performance of all ML algorithms with an accuracy of 86.3%. The proposed study offers important insights for future applications and studies in the area of emotion classification. It also highlights the importance of choosing a suitable algorithm for tasks related to text categorization.

Zitieren

APA

Zafar, B., Saeed, A., Kiran, H. M., Nasir, M., & Hameed, A. (2026). From Machine Learning to Transformers: A Comparative Study on Emotion Classification of Tweets. VAWKUM Transactions on Computer Sciences, 14(1), 40-52. https://doi.org/10.21015/vtcs.v14i1.2236

Vancouver

Zafar B, Saeed A, Kiran HM, Nasir M, Hameed A. From Machine Learning to Transformers: A Comparative Study on Emotion Classification of Tweets. VAWKUM Transactions on Computer Sciences. 2026;14(1):40-52. doi:10.21015/vtcs.v14i1.2236

BibTeX
@article{Zafar2026,
  author = {Zafar, Beenish and Saeed, Ali and Kiran, Hafiza Maria and Nasir, Mobashirah and Hameed, Afifa},
  title = {From Machine Learning to Transformers: A Comparative Study on Emotion Classification of Tweets},
  journal = {VAWKUM Transactions on Computer Sciences},
  year = {2026},
  volume = {14},
  number = {1},
  pages = {40--52},
  doi = {10.21015/vtcs.v14i1.2236}
}
Kostenlos starten →