A Deep Learning-Based Approach for Detecting Cyberbullying in Low-Resource Environments

Kankanampati, Venkata Padmavathi (2025) A Deep Learning-Based Approach for Detecting Cyberbullying in Low-Resource Environments. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (563kB) | Preview

Abstract

Cyberbullying has become a serious issue in the digital era, with the rapid growth of online communication amplifying the spread of abusive and harmful content. Detecting such behavior is a challenging task due to the unstructured, context-dependent, and informal nature of social media text. Existing approaches based on traditional machine learning or deep learning frameworks often struggle to balance detection accuracy with computational efficiency, especially in low-resource environments. Many models fail to interpret subtle linguistic cues such as sarcasm, slang, or implicit aggression, limiting their reliability in real-world settings. This research presents a lightweight deep learning framework using a Gated Recurrent Unit (GRU) integrated with an Attention mechanism to enhance contextual understanding while maintaining low computational cost. The GRU component effectively captures sequential dependencies in text, while the Attention layer dynamically focuses on significant words that influence meaning, leading to improved interpretability and detection precision. A structured experimental design was implemented to compare this architecture with traditional models using standard evaluation metrics. Results indicate that the GRU–Attention framework achieved superior performance, with an accuracy, precision, recall, and F1-score of 0.94, surpassing all baseline models. The findings confirm that the GRU–Attention model provides an efficient, accurate, and scalable solution for cyberbullying detection in resource-constrained environments.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Hamill, David UNSPECIFIED
Subjects:	B Philosophy. Psychology. Religion > Psychology > Aggressiveness > Bullying Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet > World Wide Web > Websites > Online social networks T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet > World Wide Web > Websites > Online social networks
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	07 Sep 2026 13:20
Last Modified:	07 Sep 2026 13:20
URI:	https://norma.ncirl.ie/id/eprint/9867

Actions (login required)

View Item