A Convolution-Based Function on Attribute Selections of Decision Tree

• Main Authors: Chao-Yuan Chuang, 莊朝淵
• Other Authors: 謝銘鈞
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/05021972951354604491

碩士 === 國立臺灣大學 === 醫學工程學研究所 === 92 === Recently the study of Artificial Intelligence(AI) has developed rapidly and its achievements has become a center of attraction. Many researchers and theorists in these areas are working on computational models of concepts. In the major applications of AI, Data Mining is an essential issue. Putting it briefly, to quote from Dr. Fayyad, ” Data Mining --The KDD (Knowledge Discovery in Databases) process for extracting useful knowledge from volumes of data ”; Many methods are well known in this region, such as decision tree, instance-based learning, naïve Bayes classifier and support vector machine…. Among these algorithms, decision tree will help us to get a rule map which is easily to comprehend. It is worth while examining the subject more closely, as we should concentrate on scholar Quinlan’s famous tree model and would like to propound a different principle based on another function which is similar to the concept of Convolution, the present investigation was essentially an exploration of this method. In addition, we applied this method to process medical database and hoped this preliminary study will provide us an useful reference in the related fields.