Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{192334,
        author = {Miss. Tanvi Diwakar Pal and Miss. Durga Dattatray Kondekar and Mr. Farhan Khan Tasleem Khan and Mr. Anshul Hanumant Kohchade and Mr. Dipak Pandurang Jadhav and Mr. Vishal Nandkishor Hage and Assistant Prof. Aakansha R Shukla},
        title = {Cyberbullying Detection on Social Media Using Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {1044-1052},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192334},
        abstract = {The exponential growth of social media platforms has revolutionized human communication, yet it has simultaneously facilitated the proliferation of cyberbullying, posing severe threats to mental health and well-being. Cyberbullying encompasses the use of digital technologies to harass, intimidate, threaten, or humiliate individuals through online messages, comments, and social media posts. Given the massive volume of user-generated content produced daily across platforms, manual content moderation is neither scalable nor practical, necessitating automated detection systems. This paper presents a comprehensive framework for cyberbullying detection utilizing Machine Learning (ML) and Natural Language Processing (NLP) techniques. The proposed system employs sophisticated text preprocessing, feature extraction using Term Frequency-Inverse Document Frequency (TF-IDF) and word embeddings, and classification through ensemble learning methods combining Naïve Bayes, Support Vector Machine (SVM), Random Forest, and deep learning architectures including Bidirectional Encoder Representations from Transformers (BERT). Experimental results demonstrate that the ensemble approach achieves an accuracy of 94.00% on benchmark datasets, with the BERT-based model achieving state-of-the-art performance with F1-scores exceeding 0.92. The system is designed for real-time integration into social media platforms, enabling proactive intervention through automated alerts, content filtering, and administrative reporting. By leveraging advanced AI techniques, this research contributes to creating safer digital environments and promoting responsible online behavior.},
        keywords = {Cyberbullying Detection, Machine Learning, Natural Language Processing, BERT, Ensemble Learning, Social Media, TF-IDF, Support Vector Machine, Deep Learning, Sentiment Analysis},
        month = {February},
        }