Scholarlink

<p>Users can benefit from intelligent data handling strategies by analyzing and accessing the massive amounts of data generated by automated and technological gadgets. Using recommendation systems (RS) to sift through the enormous volume of data and extract pertinent information is the most popular method. Nevertheless, early RS lacked the capability to integrate contextual information, limiting their ability to provide personalized suggestions. Context-aware recommendation systems (CARS) address this gap by incorporating contextual factors to align recommendations with dynamic user preferences and conditions. However, effectively utilizing context remains challenging, with limited resources to guide researchers and developers in designing efficient CARS. This study conducts a systematic review to summarize the current state of CARS, identify limitations, and propose future research directions. Using the Rayyan Artificial Intelligence (AI) tool for study selection, 83 primary studies, including journal articles, conference papers, and book chapters from 2020 to 2025, were analyzed. The findings highlight the diverse paradigms, context types, and methodologies adopted in CARS, along with varying evaluation techniques across multiple application domains. Key insights from this research emphasize the complexity of leveraging context to enhance recommendation quality. Persistent challenges include managing diverse contextual factors, improving scalability, and addressing evaluation inconsistencies. This study also identifies opportunities for innovation, such as dynamic context modeling and multi-objective optimization. By providing a comprehensive overview of CARS research, this work contributes to understanding the field’s progress and future potential. These findings are valuable for both academic researchers and industry professionals, offering practical guidance for developing effective, context-aware recommendation systems capable of addressing evolving user needs.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A systematic review of advancements in context-aware recommendation systems

  • 期刊论文

  • 发表日期：

    2025年12月19日

查看全文
  • Mohamed El Amine Chafiki,
  • Oumaima Stitini,
  • Soulaimane Kaloun

摘要

Users can benefit from intelligent data handling strategies by analyzing and accessing the massive amounts of data generated by automated and technological gadgets. Using recommendation systems (RS) to sift through the enormous volume of data and extract pertinent information is the most popular method. Nevertheless, early RS lacked the capability to integrate contextual information, limiting their ability to provide personalized suggestions. Context-aware recommendation systems (CARS) address this gap by incorporating contextual factors to align recommendations with dynamic user preferences and conditions. However, effectively utilizing context remains challenging, with limited resources to guide researchers and developers in designing efficient CARS. This study conducts a systematic review to summarize the current state of CARS, identify limitations, and propose future research directions. Using the Rayyan Artificial Intelligence (AI) tool for study selection, 83 primary studies, including journal articles, conference papers, and book chapters from 2020 to 2025, were analyzed. The findings highlight the diverse paradigms, context types, and methodologies adopted in CARS, along with varying evaluation techniques across multiple application domains. Key insights from this research emphasize the complexity of leveraging context to enhance recommendation quality. Persistent challenges include managing diverse contextual factors, improving scalability, and addressing evaluation inconsistencies. This study also identifies opportunities for innovation, such as dynamic context modeling and multi-objective optimization. By providing a comprehensive overview of CARS research, this work contributes to understanding the field’s progress and future potential. These findings are valuable for both academic researchers and industry professionals, offering practical guidance for developing effective, context-aware recommendation systems capable of addressing evolving user needs.