Thesis Title

Classification, Clustering, And Their Statistical Properties

Date of Graduation

Summer 2006

Degree

Master of Science in Mathematics

Department

Mathematics

Committee Chair

Yingcai Su

Keywords

association rules, classification, clustering, multivariate statistical analysis, neural networks

Subject Categories

Mathematics

Abstract

This thesis analyzed three of the major topics inherent in multivariate statistical analysis, as well as introducing statistical theories that aid in the understanding of the topics that are discussed. There is also a focus on the statistical properties of these procedures. The first topic is classification. Classification is a procedure that attempts to place data from unknown populations into known populations based on either a pre-classified set of data, known as a training set, or certain knowledge about the populations. The second topic covered is clustering. In this technique, using statistical analysis, an attempt is made to find the underlying patterns within the data without the use of a training set. The final topic discussed is association rules. This topic finds items that occur together with a certain frequency.

Copyright

© Benjamin J. Lakin

Citation-only

Dissertation/Thesis

Share

COinS