DoCluster: Efficient mining maximal biclusters without candidate maintenance in the function-resource matrix

Abstract

The functional layer is the pillar of the PHM system or avionics safety system. Its effectiveness is the core of system task effectives. In this paper we proposed a new bicluster mining algorithm: DoCluster, to effectively mine all biclusters with maximal variant usage rate and low usage rate from the discrete function-resource matrix. Firstly, DoCluster algorithm constructs a sample weighted graph which includes all resource collections between two samples that meet the definition of variant usage rate or low usage rate; secondly, all biclusters with maximal variant usage rate and low usage rate satisfying the definition are mined by using sample-growth and depth-first method in the constructed weighted graph. To improve the mining efficiency of the algorithm, DoCluster algorithm uses multiple pruning strategies to ensure the mining of maximal bicluster without candidate maintenance. The experimental results show our algorithm is more efficient than other two algorithms.