{"title":"The effect of sample size and disease prevalence on supervised machine learning of narrative data.","authors":"Lawrence K McKnight, Adam Wilcox, George Hripcsak","doi":"","DOIUrl":null,"url":null,"abstract":"<p><p>This paper examines the independent effects of outcome prevalence and training sample sizes on inductive learning performance. We trained 3 inductive learning algorithms (MC4, IB, and Naïve-Bayes) on 60 simulated datasets of parsed radiology text reports labeled with 6 disease states. Data sets were constructed to define positive outcome states at 4 prevalence rates (1, 5, 10, 25, and 50%) in training set sizes of 200 and 2,000 cases. We found that the effect of outcome prevalence is significant when outcome classes drop below 10% of cases. The effect appeared independent of sample size, induction algorithm used, or class label. Work is needed to identify methods of improving classifier performance when output classes are rare.</p>","PeriodicalId":79712,"journal":{"name":"Proceedings. AMIA Symposium","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2002-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2244149/pdf/procamiasymp00001-0560.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. AMIA Symposium","FirstCategoryId":"1085","ListUrlMain":"","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
This paper examines the independent effects of outcome prevalence and training sample sizes on inductive learning performance. We trained 3 inductive learning algorithms (MC4, IB, and Naïve-Bayes) on 60 simulated datasets of parsed radiology text reports labeled with 6 disease states. Data sets were constructed to define positive outcome states at 4 prevalence rates (1, 5, 10, 25, and 50%) in training set sizes of 200 and 2,000 cases. We found that the effect of outcome prevalence is significant when outcome classes drop below 10% of cases. The effect appeared independent of sample size, induction algorithm used, or class label. Work is needed to identify methods of improving classifier performance when output classes are rare.