碩士 / 中興大學 / 應用數學系所 / 95 / Microarray has demonstrated useful applications in cancer research. By analyzing the array generated gene expression data, cancers are distinguished by their molecular variations. In this paper, the multiclass cancer classification by using microarray data is addressed. In contrast to most existing classification procedures established without considering the class structure, we propose a new method by applying the kernel technique to generalize the proportional odds logistic regression for categorizing examples into ordered classes (e.g., cancer stages or grades). The performance of resulting classifier is demonstrated on simulated and publicly available microarray datasets.
| Identifer | oai:union.ndltd.org:TW/095NCHU5507021 |
| Date | January 2007 |
| Creators | Ling-Zi Chen, 陳陵姿 |
| Contributors | 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 | 19 |
Page generated in 0.0016 seconds