Ifp Tree-based Frequent Pattern Mining Algorithm With Privacy Protection

Data mining is a nontrivial process which refers to discover unknown,potentially useful information and knowledge from massive data set. Associationrule mining is one of the important research contents in the field of Datamining,which reflects the interdependence and relevance in amount of dataitemsets. For improving mining efficiency and privacy information securityprotection, association rule mining method and privacy information securityprotection are in-depth researched in the thesis. the main work is as follows:Firstly, The paper presents a frequent pattern mining algorithm based on IFPtree and array technology.this algorithm reduces the memory occupation byusing the constrained subtree mining method. Then, it reduces the tree traversaltime by array technology to improve the efficiency of the algorithm. Lastly, theexperimental results show that the algorithm is feasible and effective.Secondly, The paper presents a method based on IFP tree privacy protectionway.First it mines frequent item sets fastly,and determines the sensitive frequentitem sets.Then the sensitive frequent item sets are been deleted,and the otherfeatures of the original data sets can be manintained at utmost extremely. Basedon the IFP tree, three anonymous privacies are been introduced, and the privacyof the IFP tree can efficiently identify nonymous level and recognitionpersonalized privary.The analysis shows that the algorithm is feasible andeffective.