Performance analysis of a rule based Mining framework in predicting Retail sales

Abstract

newlineData Mining is the process of extracting interesting and previously newlineunknown patterns and correlations from data stored in huge data repositories newlineAssociation Rule Mining a descriptive mining technique of Data Mining is newlinethe process of discovering items which tend to occur together in transactions newlineAssociation rules are interesting correlations among attributes in a database newlineA key issue that needs more concern when using Association Rule newlineMining is the soundness of the rules external to the dataset from which they newlineare generated Rules are usually the derivative of the patterns in a specific newlinedata set When a different phenomenon occurs the changes in the set of rules newlineobtained from the new setting could be significant This thesis provides a newlineGroup based Mining of Association Rules G MAR model by paying special newlineattention as how the variation between two different settings affects the newlinechanges of the rules based on the notion of fine partitioned groups termed as newlinefactions Using this G MAR model a simple technique called Coalescent newlineDataset is proposed to get a fine approximation of the set of rules for a new newlinesituation newline