High-performance algorithm in mining generators of closed frequent itemsets
Abstract
Association-rule mining is one of the most important and well-researched techniques of Data Mining. Particularly, the techniques of mining are exact association rules and non-redundant, Some authors have proposed mining these association rules from generators of closed frequent item sets. In this paper, we propose the parallel MCP-GCFI algorithm to fast mining generators of closed frequent item sets on Multi-Core processor, as well as to expand the algorithm on distributed computing systems such as Hadoop, Spark. The experimental results show that the proposed algorithms perform faster than other existing algorithms on both real-life datasets of UCI and synthetic datasets generated by IBM Almaden
Moraref