Current Volume 9
A High-Performance Machine Learning Framework for Cyberbullying Detection Using Probabilistic Classification and Evaluation Metrics
Abstract
Cyberbullying is a burning issue on online communication sites, and this situation needs the expertise of experts to develop effective detection systems that can detect this issue. The study presents a machine learning architecture that is highly efficient in detecting cyberbullying by binary classification with supervision in two stages. The model is tested on a labeled dataset that demonstrates its capability to achieve high accuracy in predicting results when it goes through different evaluation tests. The experimental findings reveal that the system achieves a classification accuracy of 95.3 percent. The confusion matrix indicates that the system made 136 correct normal detections, 169 correct bullying detections, 11 false positive errors, and 4 false negative errors. The model is very robust, as shown by the results of the evaluations that apply both probabilistic measures and ranking-based measures. The Receiver Operating Characteristic (ROC) curve has an AUC of 0.993, indicating excellent separation of classes, and the Precision Recall (PR) curve has an AUC of 0.994, indicating that there is high precision at all recall levels. The analysis of the score distribution shows that there are normal and bullying classes as separate groups, which confirms that the model is well-calibrated and has the ability to make decisions. The findings reveal that the suggested approach allows recognizing the content of cyberbullying correctly, so it can be used in the real life, as the task of social media monitoring and online safety systems. The ongoing studies will perform in two primary directions: first, it will come up with improved ways of generalizing findings to other datasets and secondly, it will come up with real time detection systems.
Keywords
Cyberbullying Detection, Machine Learning, Binary Classification, Natural Language Processing (NLP), Supervised Learning, Text Classification, Precision Recall Curve, ROC Curve, AUC, Sentiment Analysis, Online Safety, Social Media Analysis, Artificial Intelligence, Data Mining.
Citations
IRE Journals:
Gasikanti Sathwik, Kanukanti Pandu Ranga Sai, Sandila Sai Vipul Varma, Varkala Satheesh, Dr. K. Shirisha "A High-Performance Machine Learning Framework for Cyberbullying Detection Using Probabilistic Classification and Evaluation Metrics" Iconic Research And Engineering Journals Volume 9 Issue 10 2026 Page 3397-3405 https://doi.org/10.64388/IREV9I10-1716912
IEEE:
Gasikanti Sathwik, Kanukanti Pandu Ranga Sai, Sandila Sai Vipul Varma, Varkala Satheesh, Dr. K. Shirisha
"A High-Performance Machine Learning Framework for Cyberbullying Detection Using Probabilistic Classification and Evaluation Metrics" Iconic Research And Engineering Journals, 9(10) https://doi.org/10.64388/IREV9I10-1716912
About IRE Journals
IRE Journals is an open access online journals established with an aim to publish high quality of research work in various diciplines. We have more than 6 years in publications. Over 2500+ papers already published.
Important Links
For Authors
Volume 8:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 7:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 6:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 5:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 4:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12Volume 3:
Issue 1, Issue 2, Issue 3, Issue 4, Issue 5, Issue 6, Issue 7, Issue 8, Issue 9, Issue 10, Issue 11, Issue 12