碩士 / 雲林科技大學 / 工業工程與管理研究所碩士班 / 97 / The rate of breast cancer is the highest in the female of Taiwan. The medicine advances in recent years, doctors discover that the survival rate of breast cancer close to 100% in early period. Because the data mining tools are spring up many scholars gradually use data mining tools in the medical field. There is much abundant information in the hospital in Taiwan. Taiwan people''s genes and lifestyle factors differ from other countries people. Therefore, the study uses data mining technology to establish the classification of breast cancer survival patterns and offer our country the treatment decision-making reference of survival ability of breast cancer.
Studying the patients of breast cancer in midland of Taiwan specific hospital has a total of 1340 data . Taking advantage of SVM, Logistic regression, and C5.0 decision tree construct the classified model of breast cancer patient''s survival ability and using 10-fold cross-validation approach to identify the model .
The results show that the establishment of classification tools for the classification of models by an average accuracy rate of more than 90% of both, SVM is the best way in the three categories of classification method for survival mode. The results of experiment show that three methods of the classification system establish very high accuracy rate, predict more accurate survival ability of breast cancer and offer the reference to medical decision-making frame.
| Identifer | oai:union.ndltd.org:TW/097YUNT5030023 |
| Date | January 2009 |
| Creators | Sheng-yu Huang, 黃聖育 |
| Contributors | Bor-wen Cheng, 鄭博文 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 57 |
Page generated in 0.0264 seconds