The study of differences using CHAID, CART and QUEST models for data mining

• Main Authors: Tang,Hsueh-Jen, 湯學仁
• Other Authors: Branedt Tso
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/56642806036906917644

碩士 === 國防大學管理學院 === 資訊管理學系 === 98 === The Data Mining Techniques are widely used in the fields to analyze a great quantity of data in recent years. The main purpose is to dig out valuable knowledge or message from a large number of historical data. The decision tree is one of data mining ways that set up model faster than other methods when deal with considerable quantities of complicated information. With easy comprehended rules, it is a powerful and popular classification prediction tool. It can separate the valuable materials, set up decision tree model and find the hidden trend. With the arborescent figure of the intuition, clearly color classification, and data sheets, it can help the policymaker to confirm and assess easily. With the different mathematical calculations way of decision tree model, it has different probing way and calculating rules to the materials. In order to understand every decision tree model classification rule and difference, this research utilize three decision tree models , CART, CHAID, and QUEST, to mine and analyze the reason of result. The result of study can offer the enterprises or individual for reference when they choose prospecting method, and make correct decision.