This study uses natural language processing (NLP) and machine learning to analyze sexual harassment in twelve Middle Eastern novels written in English. A machine learning framework classifies physical and non-physical harassment, while lexicon-based sentiment and emotion analysis labels sentences. An LSTM-GRU deep learning model classifies sentiment related to harassment. The harassment classification model achieved 75.8% accuracy, outperforming five other algorithms. The sentiment classification model reached 84.5% accuracy. Analysis showed that most statements, even those involving physical harassment, displayed negative sentiment.
Publisher
Humanities and Social Sciences Communications
Published On
Apr 10, 2024
Authors
Hui Qi Low, Pantea Keikhosrokiani, Moussa Pourya Asl
Tags
sexual harassment
natural language processing
machine learning
sentiment analysis
Middle Eastern novels
deep learning
classification
Related Publications
Explore these studies to deepen your understanding of the subject.
