Enhanced Negation Detection in Arabic Reviews Using Supervised Classification Approach

This paper introduces a novel approach for automated negation detection in Arabic reviews, leveraging advanced supervised classification techniques.We explore various methods, including na ve bayes (kernel), decision tree, and k-nearest neighbors, to analyze lexical and structural features from an Arabic text corpus.Our experimental results reveal that the decision tree model achieves the highest accuracy at 97.13%, significantly outperforming other classifiers.This advancement highlights the effectiveness of our approach in enhancing sentiment analysis for Arabic text, demonstrating a major improvement in negation detection capabilities.