Using Weighted Hybrid Discretization Method to Analyze Climate Changes

摘要: Data mining is the process of posing queries to large quantities data and extracting information, often previously unknown, using mathematical, statistical machine learning techniques. However some techniques like classification clustering cannot deal with numeric attributes though most real dataset contains attributes. Continuous should be divided into a small distinct range nominal in order apply Correct discretization makes succinct contributes high performance algorithms. Meanwhile, several methods are presented applied, but it dependent on area. In this paper, we propose weighted hybrid technique based entropy contingency coefficient. Also analyze evaluation well-known such as Equal-width binning, 1R, MDLP ChiMerge.