{"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/09110460","title":"Feature-Weighted Linear Stacking","arxiv_id":"0911.0460","date":"2009-11-03","proceeding":null,"authors":["Joseph Sill","Gabor Takacs","Lester Mackey","David Lin"],"abstract":"Ensemble methods, such as stacking, are designed to boost predictive accuracy\nby blending the predictions of multiple machine learning models. Recent work\nhas shown that the use of meta-features, additional inputs describing each\nexample in a dataset, can boost the performance of ensemble methods, but the\ngreatest reported gains have come from nonlinear procedures requiring\nsignificant tuning and training time. Here, we present a linear technique,\nFeature-Weighted Linear Stacking (FWLS), that incorporates meta-features for\nimproved accuracy while retaining the well-known virtues of linear regression\nregarding speed, stability, and interpretability. FWLS combines model\npredictions linearly using coefficients that are themselves linear functions of\nmeta-features. This technique was a key facet of the solution of the second\nplace team in the recently concluded Netflix Prize competition. Significant\nincreases in accuracy over standard linear stacking are demonstrated on the\nNetflix Prize collaborative filtering dataset.","url_abs":"http://arxiv.org/abs/0911.0460v2","url_pdf":"http://arxiv.org/pdf/0911.0460v2.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":"09110460","repo_url":"https://github.com/fbenites/sklearn-hierarchical-classification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"09110460","repo_url":"https://github.com/fukatani/stacked_generalization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"09110460","repo_url":"https://github.com/globality-corp/sklearn-hierarchical-classification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}