Scholarlink

<p>In this paper, we investigate Diversity-aware <i>k</i>-Maximum Inner Product Search (D<i>k</i>MIPS), an essential problem in recommendation and information retrieval tasks where balancing relevance and diversity is crucial for user satisfaction and engagement. Vanilla <i>k</i>MIPS prioritizes relevance over diversity, often yielding highly homogeneous search results. In addition, existing D<i>k</i>MIPS methods remain limited in effectiveness and efficiency. To address these issues, we introduce a novel D<i>k</i>MIPS formulation that integrates relevance and diversity into a unified objective, with a controllable parameter <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\lambda \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>λ</mi> </math></EquationSource> </InlineEquation> that allows users to adjust the level of diversity to their specific needs. We propose two scan-based algorithms, Greedy and DualGreedy, that leverage submodularity to provide D<i>k</i>MIPS results with theoretical guarantees. Furthermore, we incorporate a lightweight Ball-Cone Tree (BC-Tree) index to improve the query efficiency of Greedy and DualGreedy. Extensive experiments on real-world datasets for recommendation and document retrieval tasks show that our proposed algorithms consistently achieve a better balance between diversity and relevance than several state-of-the-art <i>k</i>MIPS and D<i>k</i>MIPS methods, while outperforming existing D<i>k</i>MIPS methods in terms of efficiency and scalability. Our code is publicly available at <a href="https://github.com/HuangQiang/DiverseMIPS">https://github.com/HuangQiang/DiverseMIPS</a>.</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

Balancing Relevance and Diversity in k-Maximum Inner Product Search

  • 期刊论文

  • 发表日期：

    2026年6月19日

查看全文
  • Qiang Huang,
  • Yanhao Wang,
  • Yiqun Sun,
  • Anthony K. H. Tung,
  • Jun Yu

摘要

In this paper, we investigate Diversity-aware k-Maximum Inner Product Search (DkMIPS), an essential problem in recommendation and information retrieval tasks where balancing relevance and diversity is crucial for user satisfaction and engagement. Vanilla kMIPS prioritizes relevance over diversity, often yielding highly homogeneous search results. In addition, existing DkMIPS methods remain limited in effectiveness and efficiency. To address these issues, we introduce a novel DkMIPS formulation that integrates relevance and diversity into a unified objective, with a controllable parameter \(\lambda \) λ that allows users to adjust the level of diversity to their specific needs. We propose two scan-based algorithms, Greedy and DualGreedy, that leverage submodularity to provide DkMIPS results with theoretical guarantees. Furthermore, we incorporate a lightweight Ball-Cone Tree (BC-Tree) index to improve the query efficiency of Greedy and DualGreedy. Extensive experiments on real-world datasets for recommendation and document retrieval tasks show that our proposed algorithms consistently achieve a better balance between diversity and relevance than several state-of-the-art kMIPS and DkMIPS methods, while outperforming existing DkMIPS methods in terms of efficiency and scalability. Our code is publicly available at https://github.com/HuangQiang/DiverseMIPS.