Optimization Algorithm Improvement of Association Rule Mining Based on Particle Swarm Optimization

The frequent itemsets in the presence of large data categories lead to disturbance, accuracy of data mining association rules feature is not good, in order to improve the extraction performance of associated features in data mining, presents an association rule mining algorithm based on particle swarm algorithm, attribute object template data using the constrained concept lattice structure for mining model construction set and the constrained concept lattice structure Hasse diagram design using depth first traversal mechanism of transaction database, with frequent item list for indexing, search data clustering using particle swarm algorithm for association rules of cable, to achieve large data clustering using fuzzy C means clustering algorithm can meet the classification attribute pattern correlation dimension, The minimum support and the minimum confidence are given as the thresholds to find the valuable association rules. The simulation results show that the method of mining the frequent itemsets can accurately reflect the characteristics of big data fusion clustering association rules mining, the convergence process is better, it can effectively extract the user interest all constrained association rules.