Data mining is a process of extracting knowledge from underlying huge multidimensional data. Data mining techniques discovers hidden patterns from a given data. Classification is one of the techniques of data mining. Data Classification consists of categorization of data under the known class labels for its most effective and efficient utilization. There are different algorithms available for classification. Association rule mining is used to generate the rules for strongly associated attributes. It helps to uncover the association between seemingly unrelated attributes. Association rules are identified by analyzing patterns which satisfies the confidence and support criteria. This paper explains pre-processing for a given bank data set along with classification and association rule mining. Association followed by classification method helps in finding the noisy data attributes. Experimental setup uses WEKA tool for data mining. WEKA is a collection of machine learning algorithms for data mining tasks. Experiment has shown that classification with guidance of strongly supported rules from association rule mining helps in removal of noise and has increased the accuracy of classifier for a given data set.

Data mining Classification Association rule mining accuracy noisy data confidence support criteria