expioring features for arabic spam detection

Abstract

In recent years, significant research has been dedicated to the development of methodologies for identifying and managing spam product reviews, particularly utilizing traditional machine learning techniques. However, more recently, researchers have shifted their focus towards deep learning approaches for spam-related applications in product review datasets. Deep learning has exhibited superior performance across various domains, including spam review detection, making it a compelling area of inquiry. As deep learning has emerged as a new frontier in machine learning, it has unlocked new possibilities for enhancing spam detection systems in product review platforms. This thesis presents the design and implementation of a spam review detection system using innovative deep learning architectures. The objective of the research is to develop robust and adaptable solutions capable of accurately identifying and mitigating spam product reviews across online platforms. The problematic area of spam review detection is explored, and related work in the field is reviewed to provide a comprehensive understanding of the current state of the art. The deep learning-based spam review detection system achieves remarkable accuracy, demonstrating the effectiveness of the chosen methodologies and their suitability for real-world spam detection tasks in product review datasets. The results of this research contribute to advancing the field of spam review detection and hold great potential for practical applications in enhancing user satisfaction and maintaining the credibility of online platforms hosting product reviews.