BruteSuppression: a size reduction method for Apriori rule sets

Abstract

Association rule mining can provide genuine insight into the data being analysed; however, rule sets can be extremely large, and therefore difficult and timeconsuming for the user to interpret. We propose reducing the size of Apriori rule sets by removing overlapping rules, and compare this approach with two standard methods for reducing rule set size: increasing the minimum confidence parameter, and increasing the minimum antecedent support parameter. We evaluate the rule sets in terms of confidence and coverage, as well as two rule interestingness measures that favour rules with antecedent conditions that are poor individual predictors of the target class, as we assume that these represent potentially interesting rules. We also examine the distribution of the rules graphically, to assess whether particular classes of rules are eliminated. We show that removing overlapping rules substantially reduces rule set size in most cases, and alters the character of a rule set less than if the standard parameters are used to constrain the rule set to the same size. Based on our results, we aim to extend the Apriori algorithm to incorporate the suppression of overlapping rules.

RETRACTED: Research on the innovation of college English teaching based on curriculum thinking

“…Combined with statistical information mining methods, the effect evaluation of college English teaching innovation reform is evaluated. 8 The grade X ð0Þ of the effect evaluation of the innovation reform is divided into Ngrades, which is X ð1Þ ; X ð2Þ ; Á Á Á ; X ðNÞ , and X ð0Þ ¼ [ N i¼1 X ðiÞ . The statistical analysis and optimization evaluation of the innovative reform effect of college English teaching are carried out by rough set algorithm.…”

RETRACTED: Modeling and analysis of higher schools massive sports data based on cloud computing

“…, n, which carry out indexing instruction control. Through weight vector coding, under K-means clustering, 8,9 the power level index information of attribute set of mining output of higher school sports data is as follows…”

A Novel Approach Based on Associative Rule Mining Technique for Multi-label Classification (ARM-MLC)

“…But this method is not a proper method as it has many disadvantages related to it. So, here we are using an algorithm called brute suppression for decreasing the rule set's size which is based on eliminating the rules that is similar to other rules [4]. The idea of this method is that if rules are alike, we can keep one and delete the rest [5].…”

“…But there are two drawbacks to this method; firstly, some important rules in dominant positions having various cases maybe eliminated and secondly, there are no methods to select the most optimal threshold value. Here, we have done a process in which if two rules are similar, we will delete one of them [4].…”