Scholarlink

<p>Recommender systems (RS) play an important role in filtering vast amounts of data and providing users with personalized suggestions. However, single-model recommendation techniques often face limitations such as data sparsity, bias, or limited representation of user-item interactions. To address such issues, this paper proposes a Blended Model Approach that combines the strengths of multiple recommendation models. Each model captures unique aspects of the data, and their outputs are integrated using a weighted blending strategy to enhance recommendation accuracy. The proposed model is evaluated on the MovieLens dataset and the results demonstrate improved performance compared to exiting models. This blended strategy offers a practical solution for building more accurate and robust recommender systems.</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 recommender systems with a blended model approach

  • 期刊论文

  • 发表日期：

    2025年6月23日

查看全文
  • Mohammed Wasid,
  • Ananay Tyagi,
  • Love Khandelwal,
  • Deepak Singh

摘要

Recommender systems (RS) play an important role in filtering vast amounts of data and providing users with personalized suggestions. However, single-model recommendation techniques often face limitations such as data sparsity, bias, or limited representation of user-item interactions. To address such issues, this paper proposes a Blended Model Approach that combines the strengths of multiple recommendation models. Each model captures unique aspects of the data, and their outputs are integrated using a weighted blending strategy to enhance recommendation accuracy. The proposed model is evaluated on the MovieLens dataset and the results demonstrate improved performance compared to exiting models. This blended strategy offers a practical solution for building more accurate and robust recommender systems.