An Optimal Association Rule Mining Algorithm Based on Knowledge Grid

Distributed data mining and in particular grid-enabled data mining has become an active area of research and development in recent years. As the amount of available digital electronic data is growing at an unprecedented rate, it is necessary to provide general data mining algorithms that help to leverage grid capacity in supporting high-performance distributed computing for solving their data mining problem in a distributed way. In this paper, an optimal multi-strategy based hybrid distribution (MBHD) algorithm based on knowledge grid is proposed for performance improvement over current grid-based association rule mining algorithms. With the optimization polices based on auction model and timestamp mechanism, MBHD algorithm effectively solves the load imbalance problem in grid environment and decreases the communication overhead. The response time performance of MBHD algorithm with different numbers of hosts and minimum supports is analyzed by experiments. The numerical results show that MBHD is efficient and performs better than count distribution (CD) algorithm, intelligent data distribution (IDD) algorithm and hybrid distribution (HD) algorithm.