Addressing Gaps in Fashion Recommendation Systems: A Review

Personalized fashion recommendation systems have become an important component of e-commerce platforms, offering custom suggestions to enhance user experience and streamline decision making. However, existing systems face significant challenges like the cold start problem, difficulty in adapting to dynamic user preferences and a lack of integration of multimodal data sources such as text, images and contextual information. This review examines recent advancements in fashion recommendation systems, focusing on the application of artificial intelligence to address these limitations. Additionally, In this review, it is proposed to consider some evaluation metrics—Personalization Index, Fashion Diversity Index and User Engagement Index—that should be considered while assessing recommendation effectiveness in the fashion domain. Our findings highlight the need for dynamic, AI-driven solutions capable of real-time personalization to better align with dynamic user preferences. These insights can serve as a foundation for future research and practical implementations, potentially improving user engagement and satisfaction across e-commerce platforms.