Transdisciplinary Approach Using Ensemble Learning Model approach in Machine Learning Technology to Predict and Abort Cyber-Abuse Against Women

Abstract

Cyber-abuse against women has become a pervasive issue in today's digital age, posing serious threats to their safety and well-being. To combat this problem, we propose an ensemble learning model approach in machine learning technology that aims to detect and abort instances of cyber-abuse targeting women on online platforms. The proposed model combines the machine learning techniques of both Support Vector Machine (SVM) and Random Forest (RF) algorithms to enhance performance and generalization. An ensemble model is constructed by combining the predictions of individual base models using aggregation techniques such as LightGBM (Light Gradient Machines). The trained ensemble learning model is integrated into a real-time monitoring system that continuously analyzes social media content. This system identifies and flags potentially abusive or harassing content directed towards women. By combining technological advancements, human expertise, and community engagement, our ensemble learning model approach offers a comprehensive solution to predict and prevent cyber-abuse against women, fostering safer and more inclusive online spaces. Additionally, our approach emphasizes the importance of jurisdictional considerations, and punitive measures implemented in different jurisdictions.