在无背景知识的面向数据的频繁模式发现研究中按照关系数据库概念重新定义了面向数据的多关系频繁模式发现任务和搜索空间.同时,使用了一个优化的精化算子构建搜索空间,这一精化算子一方面有效地利用了关系数据库隐含的数据模式特征,从而能够自然地构建有趣形态的模式,另一方面能够在不过度限制搜索空间的情况下避免等价模式的产生.建立了一个候选模式评估共享计算策略,从而降低了方法评估阶段的时间复杂性.实验表明,所提出的MRFP-DA算法整体上具有良好的效率和可扩展性.
Based on the theory and techniques of relational databases, the new definitions of task and search space are refinement operator to provide an improvement of the efficiency of candidate generation is incorporated in our algorithm. Furthermore, a new strategy of in the candidate evaluation is utilized. In the experiments, it proposed algorithm, MRFPDA, is comparable with the sharing computations to avoid redundant computations is shown that on small datasets the performance of the performance of the state-of-the-art of multi-relational fre-quent pattern discovery, and on large datasets MRFPDA is more scalable than two existing approaches.