{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/efficient-representative-subset-selection","title":"Efficient Representative Subset Selection over Sliding Windows","arxiv_id":"1706.04764","date":"2017-06-15","proceeding":null,"authors":["Yanhao Wang","Yuchen Li","Kian-Lee Tan"],"abstract":"Representative subset selection (RSS) is an important tool for users to draw\ninsights from massive datasets. Existing literature models RSS as the\nsubmodular maximization problem to capture the \"diminishing returns\" property\nof the representativeness of selected subsets, but often only has a single\nconstraint (e.g., cardinality), which limits its applications in many\nreal-world problems. To capture the data recency issue and support different\ntypes of constraints, we formulate dynamic RSS in data streams as maximizing\nsubmodular functions subject to general $d$-knapsack constraints (SMDK) over\nsliding windows. We propose a \\textsc{KnapWindow} framework (KW) for SMDK. KW\nutilizes the \\textsc{KnapStream} algorithm (KS) for SMDK in append-only streams\nas a subroutine. It maintains a sequence of checkpoints and KS instances over\nthe sliding window. Theoretically, KW is\n$\\frac{1-\\varepsilon}{1+d}$-approximate for SMDK. Furthermore, we propose a\n\\textsc{KnapWindowPlus} framework (KW$^{+}$) to improve upon KW. KW$^{+}$\nbuilds an index \\textsc{SubKnapChk} to manage the checkpoints and KS instances.\n\\textsc{SubKnapChk} deletes a checkpoint whenever it can be approximated by its\nsuccessors. By keeping much fewer checkpoints, KW$^{+}$ achieves higher\nefficiency than KW while still guaranteeing a\n$\\frac{1-\\varepsilon'}{2+2d}$-approximate solution for SMDK. Finally, we\nevaluate the efficiency and solution quality of KW and KW$^{+}$ in real-world\ndatasets. The experimental results demonstrate that KW achieves more than two\norders of magnitude speedups over the batch baseline and preserves high-quality\nsolutions for SMDK over sliding windows. KW$^{+}$ further runs 5-10 times\nfaster than KW while providing solutions with equivalent or even better\nutilities.","url_abs":"http://arxiv.org/abs/1706.04764v2","url_pdf":"http://arxiv.org/pdf/1706.04764v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"efficient-representative-subset-selection","repo_url":"https://github.com/yhwang1990/Representative-Subset-Selection","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}