作者: Wei Zhang , Bingru Yang
DOI: 10.1109/ISDA.2006.92
关键词:
摘要: We propose MRFPDA, an efficient and scalable algorithm for multi-relational frequent pattern discovery. incorporate in the optimal refinement operator to provide improvement of efficiency candidate generation. Furthermore, MRFPDA utilizes a new strategy sharing computations avoid redundant evaluation. In our experiments, it is shown that on small datasets performance comparable with state-of-the-art discovery, large more than two existing approaches