Concurrent frequent itemsets mining in a shared prefix tree using the Apriori algorithm
AbstractThis paper presents a sequential frequent itemsets mining algorithm Apriori that is adapted to concurrent processing. It applies Master Slave scheme to candidate generation and support counting operations performed by threads on a single machine. Two approaches to traversing shared prefix tree and counting support of itemsets are presented and compared. Several optimization methods have been proposed for the multithreaded environment. Proposed enhancements have been successfully implemented using JAVA. This paper discusses results of the performance of concurrent Apriori algorithm against different datasets. Presented approach has been illustrated with many experiments and measurements performed using multiprocessor and multithreaded computer.
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.