首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0902108-135154


URN etd-0902108-135154 Statistics This thesis had been viewed 3246 times. Download 1004 times.
Author Wen-Tin Hsu
Author's Email Address No Public.
Department Computer Science and Enginerring
Year 2007 Semester 2
Degree Master Type of Document Master's Thesis
Language English Page Count 76
Title The Study of Using Visualization Strategy to Display Ocean Salinity and Temperature Variations
Keyword
  • spatial-temporal data mining
  • inter-transaction association rules
  • Ocean salinity
  • Ocean temperature
  • Ocean temperature
  • Ocean salinity
  • inter-transaction association rules
  • spatial-temporal data mining
  • Abstract 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.
    Advisor Committee
  • none - advisor
  • Yo-Ping Huang - advisor
  • none - co-chair
  • Files indicate access worldwide
    Date of Defense 2008-07-11 Date of Submission 2008-09-10


    Browse | Search All Available ETDs