EUVIMEDEUVIMED entdecken
Crossref

Detection and Classification of Cyberbullying in Social Media Using Transformer-Based NLP Models

Prof. Sagar Dhanake, Aniket Sawane, Romaan Shaikh, Shradha Suse

International Journal of Advanced Research in Science Communication and Technology · 26.06.2026 · 505

DOI: 10.48175/ijarsct-37251 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 26.06.2026. Verfasst von Prof. Sagar Dhanake und 3 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid growth of social media platforms has significantly increased the prevalence of cyberbullying, posing serious psychological and social risks to individuals. Detecting such harmful content automatically remains a challenging task due to the contextual, nuanced, and often ambiguous nature of online language. This paper presents a transformer-based Natural Language Processing (NLP) framework for the detection and classification of cyberbullying in social media text. The proposed approach leverages pre-trained transformer models, including Bidirectional Encoder Representations from Transformers (BERT), to capture contextual semantics and linguistic patterns more effectively than traditional machine learning methods. A labeled dataset comprising social media posts is utilized to train and evaluate the model across multiple categories of cyberbullying, such as harassment, hate speech, and offensive language. Text preprocessing techniques, including tokenization, normalization, and handling of imbalanced data, are applied to enhance model performance. The transformer model is fine-tuned on the dataset and compared with baseline classifiers such as Support Vector Machines (SVM) and Long Short-Term Memory (LSTM) networks. Experimental results demonstrate that the proposed transformer-based model achieves superior performance in terms of accuracy, precision, recall, and F1-score, highlighting its effectiveness in capturing contextual dependencies and subtle linguistic cues. Furthermore, the model shows robustness in handling noisy and informal social media text. The findings suggest that transformer-based NLP models provide a promising solution for real-time cyberbullying detection systems, contributing to safer online environments. Future work will focus on multilingual detection and explainable AI techniques to improve transparency and adaptability across diverse social media platforms.

Zitieren

APA

Dhanake, P. S., Sawane, A., Shaikh, R., & Suse, S. (2026). Detection and Classification of Cyberbullying in Social Media Using Transformer-Based NLP Models. International Journal of Advanced Research in Science Communication and Technology, 505. https://doi.org/10.48175/ijarsct-37251

Vancouver

Dhanake PS, Sawane A, Shaikh R, Suse S. Detection and Classification of Cyberbullying in Social Media Using Transformer-Based NLP Models. International Journal of Advanced Research in Science Communication and Technology. 2026:505. doi:10.48175/ijarsct-37251

BibTeX
@article{Dhanake2026,
  author = {Dhanake, Prof. Sagar and Sawane, Aniket and Shaikh, Romaan and Suse, Shradha},
  title = {Detection and Classification of Cyberbullying in Social Media Using Transformer-Based NLP Models},
  journal = {International Journal of Advanced Research in Science Communication and Technology},
  year = {2026},
  pages = {505},
  doi = {10.48175/ijarsct-37251}
}
Kostenlos starten →