Document Typethesis Author NameMoraski, Ashley M. URNetd-050406-124140 TitleClassification via distance profile nearest neighbors DegreeMS DepartmentMathematical Sciences AdvisorsJayson Wilbur, Advisor Keywordsclassification distance profile nearest neighbor Date of Presentation/Defense2006-05-04 Availabilityunrestricted

AbstractMost classification rules can be expressed in terms of a distance (or dissimilarity) from the point to be classified to each of the candidate classes. For example, linear discriminant analysis classifies points into the class for which the (sample) Mahalanobis distance is smallest. However, dependence among these point-to-group distance measures is generally ignored. The primary goal of this project is to investigate the properties of a general non-parametric classification rule which takes this dependence structure into account. A review of classification procedures and applications is presented. The distance profile nearest-neighbor classification rule is defined. Properties of the rule are then explored via application to both real and simulated data and comparisons to other classification rules are discussed.

FilesMoraski.pdf

Browse by Author | Browse by Department | Search all available ETDs

Questions? Email etd-questions@wpi.eduMaintained by webmaster@wpi.edu