Research on association rules algorithms based on cloud environment
Zhou Li-jua · Jisuanji gongcheng yu sheji · 2014
To solve the performance bottleneck problem of traditional data mining algorithms in dealing with huge amounts of data,the FP-Growth algorithm is studied.A parallel composite linked list-based FP-Growth algorithm in cloud environment(PCLFP)is proposed.Frequent patterns are mined based on composite linked list instead of building FP-Tree and conditional FPTree.With different sizes of data sets to validate the improved algorithm,the results show that PCL-FP improves the efficiency and has good flexibility and extensibility.Huge amounts of data and mine frequent itemsets are processed effectively.