Applying data mining to retrieve the optimal environment for plant growth

• Main Authors: Chen-Wei Lin, 林辰蔚
• Other Authors: Yu-Chang Tzeng
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/05927895261247575901

碩士 === 國立聯合大學 === 電子工程學系碩士班 === 104 === Traditionally, agriculture is the most important industry in Taiwan. Since Taiwan participated in the World Trade Organization (WTO), extreme challenges inevitably impact on Taiwan’s agriculture. Precision agriculture which includes functions of environment monitoring, information integration, real-time analysis, and decision making, is the best solution to improve the competitiveness of traditional agricultural. Microclimate has a great influence on the quality of crops, the quality of the same crop differs greatly in different countries or regions due to different climatic conditions. To traditional agriculture, the expert knowledge of the optimal microclimate conditions for crop planting is accumulated by long-term experiences. Limited by its experimental field, the expert knowledge is applicable only to small area. This thesis is to use the data mining techniques to retrieve the optimal environment for plant growth. The value of data mining is to extract valuable information from a great amount of the data. To verify the retrieval results of using data mining, this thesis presents the distance to the mean classification method according to the microclimate of the period of crop growth simulation data and actual data for quality classification, and with the LB_Keogh method of comparison. From the experimental results obtained by using data mining technology to estimate the most suitable for crop growth environment. In addition, the distance to the mean classification method is better than LB_Keogh method.