Association Rule Classification and Regression Algorithm Based on Frequent Itemset Tree

The categorization association rules based on the Apriori algorithm can't deal with the numerical data directly. When mass rules are generated, classifying the new data enjoys matching so many rules one by one as to decrease the efficiency and accuracy. Moreover, the association rules can't be used to realize the regression prediction. In order to solve above problems, we proposed a new association rule classification and regression algorithm based on frequent itemset tree (ARCRFI-tree) according to the advantages of matrix operation and tree structure. Firstly, all frequent itemsets are obtained by constructing a new frequent tree structure, based on which the association rules are mined. Then, the consequents of the association rules are reconstructed with the least square method to realize the classification and regression prediction for new sample. Finally, the theoretical analysis and experiments compared with algorithms demonstrate our algorithm has high prediction accuracy and mining efficiency.