Hybrid Quality-Based Recommender Systems: A Systematic Literature Review

Area of Science:

• Computer Science
• Information Science
• E-commerce Technology

Background:

• Consumer behavior and search patterns are evolving with technology, impacting e-commerce.
• Recommender systems are crucial for increasing product visibility and sales in online retail.
• Hybrid recommender systems, combining multiple methodologies, are a significant research focus.

Purpose of the Study:

• To conduct a systematic review of recent developments in hybrid recommender systems.
• To assess progress, identify common approaches, explore technical contexts, and highlight research gaps.
• To provide insights into the design and implementation of hybrid recommender systems, considering big data challenges.

Main Methods:

• Systematic literature review adhering to Cochrane Handbook and Kitchenham & Charters principles.
• Searched ACM, Google Scholar, Scopus, Springer (last 4 years) and Web of Science (all years).
• Utilized ASReview, an open-source active learning application for efficient literature filtering.

Main Results:

• Analysis of hybrid recommender system trends, practical applications, strengths, and limitations.
• Identification of key challenges and opportunities in big data (volume, velocity, variety) for hybrid systems.
• Overview of the current state-of-the-art in hybrid recommendation algorithms.

Conclusions:

• Hybrid recommender systems are vital for modern e-commerce, offering enhanced user experiences and sales.
• Further research is needed to address big data challenges and refine hybrid approaches.
• This review provides a foundation for future research in optimizing hybrid recommender systems.