Scholarlink

<p>A context-aware recommender system has become more popular because of its capability to provide personalized recommendations by considering contextual information such as place, season, companion, and more. These days, cross-domain recommender systems have been developed, which take user preferences and behavior from one domain into account in another domain. However, like traditional recommender systems, the abovementioned recommender systems have also come up with issues like sparsity and cold start. We propose a recommender system that combines a context-aware and cross-domain recommender system (CACD) to address the aforementioned problems to provide efficient personalization to the user. This paper discusses the issues and proposes a dynamic approach to combine newer context with time for improving CACD recommendation systems. We propose a unique weighted similarity metric for improved computation of recommendations. The experiment’s findings demonstrate that the CACD recommender system outperforms cutting-edge systems in improving user personalisation.</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

Enhancing Personalisation Using Dynamic Context-Aware and Content-Based Cross Domain Recommender System

  • 期刊论文

  • 发表日期：

    2025年9月20日

查看全文
  • Jitali Patel,
  • Pavan Kumar,
  • Hetav Modi,
  • Priyal Palkhiwala,
  • Jigna Patel,
  • Rupal Kapdi,
  • Amisha Patel

摘要

A context-aware recommender system has become more popular because of its capability to provide personalized recommendations by considering contextual information such as place, season, companion, and more. These days, cross-domain recommender systems have been developed, which take user preferences and behavior from one domain into account in another domain. However, like traditional recommender systems, the abovementioned recommender systems have also come up with issues like sparsity and cold start. We propose a recommender system that combines a context-aware and cross-domain recommender system (CACD) to address the aforementioned problems to provide efficient personalization to the user. This paper discusses the issues and proposes a dynamic approach to combine newer context with time for improving CACD recommendation systems. We propose a unique weighted similarity metric for improved computation of recommendations. The experiment’s findings demonstrate that the CACD recommender system outperforms cutting-edge systems in improving user personalisation.