{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Sniatynski MJ"],"funding":["NCI NIH HHS","National Institutes of Health"],"pagination":["100415"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC8848007"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["3(2)"],"pubmed_abstract":["Combining classifier systems potentially improves predictive accuracy, but outcomes have proven impossible to predict. Classification most commonly improves when the classifiers are \"sufficiently good\" (generalized as \" <b><i>accuracy</i></b> \") and \"sufficiently different\" (generalized as \" <b><i>diversity</i></b> \"), but the individual and joint quantitative influence of these factors on the final outcome remains unknown. We resolve these issues. Beginning with simulated data, we develop the DIRAC framework (<i>DIversity</i> of Ranks and <i>ACcuracy</i>), which accurately predicts outcome of both score-based fusions originating from exponentially modified Gaussian distributions and rank-based fusions, which are inherently distribution independent. DIRAC was validated using biological dua"],"journal":["Patterns (New York, N.Y.)"],"pubmed_title":["Ranks underlie outcome of combining classifiers: Quantitative roles for <i>diversity</i> and <i>accuracy</i>."],"pmcid":["PMC8848007"],"funding_grant_id":["R01-HL132556","U01-ES16048","P01-CA168530","R01-HL140335","RC1-ES018411","R01-AG045713","R01-AG28996","P30 CA071789"],"pubmed_authors":["Wilkens LR","Sniatynski MJ","Kristal BS","Ernst T","Shepherd JA","Hsu DF"],"additional_accession":[]},"is_claimable":false,"name":"Ranks underlie outcome of combining classifiers: Quantitative roles for <i>diversity</i> and <i>accuracy</i>.","description":"Combining classifier systems potentially improves predictive accuracy, but outcomes have proven impossible to predict. Classification most commonly improves when the classifiers are \"sufficiently good\" (generalized as \" <b><i>accuracy</i></b> \") and \"sufficiently different\" (generalized as \" <b><i>diversity</i></b> \"), but the individual and joint quantitative influence of these factors on the final outcome remains unknown. We resolve these issues. Beginning with simulated data, we develop the DIRAC framework (<i>DIversity</i> of Ranks and <i>ACcuracy</i>), which accurately predicts outcome of both score-based fusions originating from exponentially modified Gaussian distributions and rank-based fusions, which are inherently distribution independent. DIRAC was validated using biological dua","dates":{"release":"2022-01-01T00:00:00Z","publication":"2022 Feb","modification":"2025-06-01T12:02:11.277Z","creation":"2025-06-01T12:02:11.277Z"},"accession":"S-EPMC8848007","cross_references":{"pubmed":["35199065"],"doi":["10.1016/j.patter.2021.100415"]}}