Show simple item record

dc.contributor.advisorGüvenir, Altayen_US
dc.contributor.authorKurtcephe, Muraten_US
dc.date.accessioned2016-01-08T18:14:07Z
dc.date.available2016-01-08T18:14:07Z
dc.date.issued2010
dc.identifier.urihttp://hdl.handle.net/11693/15142
dc.descriptionAnkara : The Department of Computer Engineering and the Institute of Engineering and Science of Bilkent University, 2010.en_US
dc.descriptionThesis (Master's) -- Bilkent University, 2010.en_US
dc.descriptionIncludes bibliographical references leaves 56-64.en_US
dc.description.abstractRisks exist in many different domains; medical diagnoses, financial markets, fraud detection and insurance policies are some examples. Various risk measures and risk estimation systems have hitherto been proposed and this thesis suggests a new risk estimation method. Risk estimation by maximizing the area under a Receiver Operating Characteristics (ROC) curve (REMARC) defines risk estimation as a ranking problem. Since the area under ROC curve (AUC) is related to measuring the quality of ranking, REMARC aims to maximize the AUC value on a single feature basis to obtain the best ranking possible on each feature. For a given categorical feature, we prove a sufficient condition that any function must satisfy to achieve the maximum AUC. Continuous features are also discretized by a method that uses AUC as a metric. Then, a heuristic is used to extend this maximization to all features of a dataset. REMARC can handle missing data, binary classes and continuous and nominal feature values. The REMARC method does not only estimate a single risk value, but also analyzes each feature and provides valuable information to domain experts for decision making. The performance of REMARC is evaluated with many datasets in the UCI repository by using different state-of-the-art algorithms such as Support Vector Machines, naïve Bayes, decision trees and boosting methods. Evaluations of the AUC metric show REMARC achieves predictive performance significantly better compared with other machine learning classification methods and is also faster than most of them. In order to develop new risk estimation framework by using the REMARC method cardiovascular surgery domain is selected. The TurkoSCORE project is used to collect data for training phase of the REMARC algorithm. The predictive performance of REMARC is compared with one of the most popular cardiovascular surgical risk evaluation method, called EuroSCORE. EuroSCORE is evaluated on Turkish patients and it is shown that EuroSCORE model is insufficient for Turkish population. Then, the predictive performances of EuroSCORE and TurkoSCORE that uses REMARC for prediction are compared. Empirical evaluations show that REMARC achieves better prediction than EuroSCORE on Turkish patient population.en_US
dc.description.statementofresponsibilityKurtcephe, Muraten_US
dc.format.extentxiv, 77 leaves, illustrationsen_US
dc.language.isoEnglishen_US
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.subjectRisk Estimationen_US
dc.subjectAUC Maximizationen_US
dc.subjectAUCen_US
dc.subjectRankingen_US
dc.subjectCardiovascular Operation Risk Evaluationen_US
dc.subject.lccWG169 .K87 2010en_US
dc.subject.lcshHeart--Surgery.en_US
dc.subject.lcshCardiac surgical procedures--Adverse effects.en_US
dc.subject.lcshHealth risk assessment--Mathematical models.en_US
dc.titleRisk estimation by maximizing area under receiver operating characteristics curve with application to cardiovascular surgeryen_US
dc.typeThesisen_US
dc.departmentDepartment of Computer Engineeringen_US
dc.publisherBilkent Universityen_US
dc.description.degreeM.S.en_US
dc.identifier.itemidB122248


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record