Data Preprocessing for Web Data Mining

摘要: Data preprocessing includes data cleaning, integration, transformation and reduction. cleaning is aimed to remove unrelated or redundant items through two processes. integration three main problems each of them can be solved by kinds methods. generalization property construction standardization. Three algorithms used normalize the data. The last step reduction compress in order improve quality mining models. All these four steps are interrelated other shouldn’t separated. They work together final result mining.