碩士 / 中興大學 / 應用數學系所 / 94 / Motivation
Microarray has been increasingly used in cancer research. Using expression levels of thousands of genes monitored simultaneously by microarray, tumors’ molecular variations are distinguished, and cancers are more accurately classified. While statistical methods have been extensively evaluated for dichotomous classifications, there are only limited reports on the important issue of multi-class cancer classification. It needs to explore the statistical methods of the multi-class cancer classification.
Objective
In this research, we address multi-class cancer classifications by applying logistic discrimination (LD) based methods on microarray data of nominal and ordinal scaled sample class outcomes, e.g., tissue samples of different cancer subtypes and cancer stages. LD based classifiers are assessed by misclassification rates on microarray data and comparing with normal model discrimination based classifiers.
Identifer | oai:union.ndltd.org:TW/094NCHU5507029 |
Date | January 2006 |
Creators | Yi-Xuan Wu 吳怡萱, 吳怡萱 |
Contributors | Qi-Kang Chen 陳齊康, 陳齊康 |
Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
Language | en_US |
Detected Language | English |
Type | 學位論文 ; thesis |
Format | 30 |
Page generated in 0.0149 seconds