{"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/algorithm-selection-for-collaborative","title":"Algorithm Selection for Collaborative Filtering: the influence of graph metafeatures and multicriteria metatargets","arxiv_id":"1807.09097","date":"2018-07-23","proceeding":null,"authors":["Tiago Cunha","Carlos Soares","André C. P. L. F. de Carvalho"],"abstract":"To select the best algorithm for a new problem is an expensive and difficult\ntask. However, there are automatic solutions to address this problem: using\nMetalearning, which takes advantage of problem characteristics (i.e.\nmetafeatures), one is able to predict the relative performance of algorithms.\nIn the Collaborative Filtering scope, recent works have proposed diverse\nmetafeatures describing several dimensions of this problem. Despite interesting\nand effective findings, it is still unknown whether these are the most\neffective metafeatures. Hence, this work proposes a new set of graph\nmetafeatures, which approach the Collaborative Filtering problem from a Graph\nTheory perspective. Furthermore, in order to understand whether metafeatures\nfrom multiple dimensions are a better fit, we investigate the effects of\ncomprehensive metafeatures. These metafeatures are a selection of the best\nmetafeatures from all existing Collaborative Filtering metafeatures. The impact\nof the most representative metafeatures is investigated in a controlled\nexperimental setup. Another contribution we present is the use of a\nPareto-Efficient ranking procedure to create multicriteria metatargets. These\nnew rankings of algorithms, which take into account multiple evaluation\nmeasures, allow to explore the algorithm selection problem in a fairer and more\ndetailed way. According to the experimental results, the graph metafeatures are\na good alternative to related work metafeatures. However, the results have\nshown that the feature selection procedure used to create the comprehensive\nmetafeatures is is not effective, since there is no gain in predictive\nperformance. Finally, an extensive metaknowledge analysis was conducted to\nidentify the most influential metafeatures.","url_abs":"http://arxiv.org/abs/1807.09097v1","url_pdf":"http://arxiv.org/pdf/1807.09097v1.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":"algorithm-selection-for-collaborative","repo_url":"https://github.com/tiagodscunha/cf2vec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"algorithm-selection-for-collaborative","repo_url":"https://github.com/tiagodscunha/cf_metafeatures","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"algorithm-selection-for-collaborative","repo_url":"https://github.com/tiagodscunha/lr_alg_sel","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"},{"task_slug":"feature-selection","task_name":"feature selection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}