Title: AI-Powered Sorting in E-commerce: Personalization and Performance Optimization
Authors: Nesreen S. Aljerjawi and Samy S. Abu-Naser
Volume: 9
Issue: 8
Pages: 110-113
Publication Date: 2025/08/28
Abstract:
Traditional e-commerce platforms primarily utilize static sorting algorithms based on attributes such as price, popularity, or recency. This one-size-fits-all approach fails to address the unique preferences and behavioral patterns of individual users, leading to suboptimal user experiences and missed revenue opportunities. This paper proposes a framework for an AI-powered sorting system that dynamically re-ranks products for each user in real-time, leveraging machine learning to personalize product discovery and optimize platform performance. By analyzing user data, including clickstream, search queries, and purchase history, the system learns to predict a user's likelihood of engaging with a product. The proposed methodology outlines the data collection, feature engineering, and model architecture required to build a robust and scalable solution. We discuss the potential for this system to significantly enhance key e-commerce metrics, such as click-through rate, conversion rate, and customer satisfaction, representing a major advancement over conventional, rule-based systems.