碩士 / 大同大學 / 資訊工程學系(所) / 96 / The domain experts find that ocean salinity and temperature play an important role in global climate changes. Global ocean salinity and temperature abnormal variations attract researchers to find interesting patterns. Data mining strategy is used to discover association rules from Argo ocean salinity and temperature variations. In the past, the association rules are only described in rule forms. A visualization system is constructed to help users observe the association rules and their variations. In our research, the ocean salinity and temperature variation data along the Taiwan coast were analyzed.
Traditional mining techniques focus on finding associations among items within one transaction. They are unable to discover rich contextual patterns related to location and time. FITI algorithm is used to find the association rules. The quantitative inter-transaction association rules mining algorithm is proposed to find the salinity and temperature abnormal variation patterns from the transformed data set. Example from the discovered association rules looks like, “If the salinity near the northern Taiwan rose 0.1psu to 0.2psu, then the temperature near the northeast Taiwan will rise from 0℃ to 0.8℃ in the next month.”
This study focuses on ocean salinity and temperature variations obtained from the waters surrounding Taiwan. A visualization system is constructed for users to easily understand inter-transaction association rules from ocean salinity and temperature variations.
Identifer | oai:union.ndltd.org:TW/096TTU05392062 |
Date | January 2008 |
Creators | Wen-Tin Hsu, 許雯婷 |
Contributors | Yo-Ping Huang, none, 黃有評, 謝尚琳 |
Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
Language | en_US |
Detected Language | English |
Type | 學位論文 ; thesis |
Format | 76 |
Page generated in 0.0166 seconds