High Dimensional Global Pattern Set Mining Based On Genetic Constraint Mechanism

Abstract--- Pattern set mining troubled with discovery a set of NP interrelated patterns in constraints. The Number of Pattern(NP) set mining crisis is a very universal problem that can be instantiated to a large range of well-known mining responsibilities including concept-learning, rule-learning, redescription mining, abstract clustering and NP-tiling. To resolve this create a large number of constraints for use in NP pattern set mining, both at the narrow level, that is, on individual patterns, and on the universal level for a whole pattern set. The genetic approach used to model by high dimensional data into low dimensional data population for speedy retrieval of frequent itemsets and reducing the storage space of the dataset. This allows exploring a large number of settings within a unified framework and solving multiple mining problems in single step. The experimental results show that there is an improvement of predictive accuracy for classifying data summarized based on Fixed Length Feature Construction With Substitution (FLFCWS) to address the problem by means of optimizing the feature construction for relational data summarization. Keywords-Data mining, Genetic approach, Constraints Constraint programming and pattern set mining