碩士 / 國立臺灣海洋大學 / 海洋環境資訊學系 / 97 / The purpose of this thesis is using data mining techniques to develop a prediction model, and to discuss the mechanism of influence on typhoon intensification by using decision tree algorithms. The related environmental data including sea surface temperature (SST), atmospheric water vapor (WV), rain rate, sea surface height anomaly (SSHA) and air-sea temperature difference are used for the analysis. The results indicate that the most important factor to affect typhoon intensity is air-sea temperature difference and the second one is SST. When typhoons pass over the ocean where its SST is larger than air temperature, about 88% of typhoons’ intensities are enhanced. The prediction model is further validated by using the data of super typhoon JANGMI (200815, category-5). The results show that the accuracy of prediction is around 82.35%, and the precision is about 85.71%. This study suggests that the data mining technique is an efficient tool for estimation and prediction of influence of typhoon intensity with marine environments.
| Identifer | oai:union.ndltd.org:TW/097NTOU5282006 |
| Date | January 2009 |
| Creators | Yu-Ching Chang, 張育菁 |
| Contributors | Chung-Ru Ho, 何宗儒 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 59 |
Page generated in 0.0018 seconds